WHAT OTHERS ASKED
Every question put to TEGBot, by newest chat or by theme. Tap a question to see the answer.
Who finishes strongest over the last 3 holes?
John PATTERSON finishes strongest. He is the only player who scores as well over holes 16–18 as on the rest of the round: his Stableford points per hole are slightly higher (+0.06), and his gross is level at +1.56 vs par per hole. Everyone else fades, and David MULLIN fades most (−0.18 points per hole).
| Player | Last 3 gross vs par per hole | Holes 1–15 gross | Last 3 pts per hole | Holes 1–15 pts | Last 3 holes played | Points change |
|---|---|---|---|---|---|---|
| John PATTERSON | 1.56 | 1.56 | 2.03 | 1.97 | 108 | 0.06 |
| Stuart NEUMANN | 1.58 | 1.45 | 1.85 | 1.91 | 141 | -0.06 |
| Gregg WILLIAMS | 1.39 | 1.32 | 1.97 | 2.04 | 213 | -0.07 |
| Jon BAKER | 1.28 | 1.1 | 1.92 | 2.04 | 213 | -0.12 |
| Henry MELLER | 1.59 | 1.4 | 1.68 | 1.84 | 81 | -0.16 |
| Alex BAKER | 1.89 | 1.68 | 1.88 | 2.05 | 156 | -0.17 |
| David MULLIN | 1.21 | 1.02 | 1.82 | 1.99 | 213 | -0.18 |
On raw numbers, David MULLIN has the lowest gross vs par over the last 3 holes (+1.21 per hole), but that's because he's the best golfer overall.
How I worked this out: I compared each player's holes 16–18 with their own holes 1–15, across all TEGs and rounds. I led with the change in Stableford points per hole, which allows for handicap. The "Last 3 holes played" column is the number of holes each player has played on 16–18. John PATTERSON and Henry MELLER have the smallest samples, so treat their figures with caution.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
h['L']=h.Hole>=16
g=h.groupby(['Player','L']).agg(gvp=('GrossVP','mean'),pts=('Stableford','mean'),n=('Sc','size')).unstack()
o=pd.DataFrame({'Last3 gross vs par/hole':g[('gvp',True)],'Holes1-15':g[('gvp',False)],'Last3 pts/hole':g[('pts',True)],'1-15 pts/hole':g[('pts',False)],'n last3':g[('n',True)]})
o['Gross diff']=o.iloc[:,0]-o.iloc[:,1]
o['Pts diff']=o['Last3 pts/hole']-o['1-15 pts/hole']
print(o.sort_values('Pts diff',ascending=False).round(2).to_markdown())
"
→ Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 135, in import_optional_dependency
module = importlib.import_module(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1140, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'tabulate'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<string>", line 9, in <module>
File "/usr/local/lib/python3.11/site-packages/pandas/util/_decorators.py", line 333, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/core/frame.py", line 2988, in to_markdown
tabulate = import_optional_dependency("tabulate")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 138, in import_optional_dependency
raise ImportError(msg)
ImportError: Missing optional dependency 'tabulate'. Use pip or conda to install tabulate.
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
h['L']=h.Hole>=16
g=h.groupby(['Player','L']).agg(gvp=('GrossVP','mean'),pts=('Stableford','mean'),n=('Sc','size')).unstack()
o=pd.DataFrame({'Last3 gross':g[('gvp',True)],'Holes1-15 gross':g[('gvp',False)],'Last3 pts':g[('pts',True)],'1-15 pts':g[('pts',False)],'n last3':g[('n',True)]})
o['Pts diff']=o['Last3 pts']-o['1-15 pts']
o=o.sort_values('Pts diff',ascending=False).round(2).reset_index()
print('|'+'|'.join(o.columns)+'|');print('|'+'---|'*len(o.columns))
for r in o.values:print('|'+'|'.join(map(str,r))+'|')
"
→ |Player|Last3 gross|Holes1-15 gross|Last3 pts|1-15 pts|n last3|Pts diff| |---|---|---|---|---|---|---| |John PATTERSON|1.56|1.56|2.03|1.97|108|0.06| |Stuart NEUMANN|1.58|1.45|1.85|1.91|141|-0.06| |Gregg WILLIAMS|1.39|1.32|1.97|2.04|213|-0.07| |Jon BAKER|1.28|1.1|1.92|2.04|213|-0.12| |Henry MELLER|1.59|1.4|1.68|1.84|81|-0.16| |Alex BAKER|1.89|1.68|1.88|2.05|156|-0.17| |David MULLIN|1.21|1.02|1.82|1.99|213|-0.18|
What are Greggs best ever total gross scores across a full TEG
Gregg WILLIAMS's best full-TEG gross score is 346 (+58) at TEG 19 in 2026. That's his best by both measures. TEG 15 is next at 347 (+63). The top five by strokes versus par are below. TEG 16 (350) has fewer strokes than TEG 18 (354), but both are +66 because the courses had different pars.
| TEG | Year | Area | Gross | Vs par | Per round |
|---|---|---|---|---|---|
| 19 | 2026 | Algarve | 346 | +58 | 86.5 |
| 15 | 2022 | Lisbon Coast | 347 | +63 | 86.8 |
| 18 | 2025 | Catalonia | 354 | +66 | 88.5 |
| 16 | 2023 | Lisbon Coast | 350 | +66 | 87.5 |
| 13 | 2020 | Kent | 355 | +68 | 88.8 |
See the Records page for the all-time list.
How I worked this out: I used Gregg's 18 completed TEGs and ranked them by gross strokes versus par. Ties share a position. All the TEGs shown had four rounds, so the totals compare fairly.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
t=pd.read_csv('tegs.csv');t=t[(t.Player=='Gregg WILLIAMS')&(t.Complete==True)]
t['PerRound']=(t.Sc/t.Rounds).round(1)
print(len(t));print(t.sort_values('GrossVP')[['TEGNum','Year','Area','Rounds','Sc','GrossVP','PerRound']].head(8).to_markdown(index=False))"
→ 18
Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 135, in import_optional_dependency
module = importlib.import_module(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1140, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'tabulate'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<string>", line 5, in <module>
File "/usr/local/lib/python3.11/site-packages/pandas/util/_decorators.py", line 333, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/core/frame.py", line 2988, in to_markdown
tabulate = import_optional_dependency("tabulate")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 138, in import_optional_dependency
raise ImportError(msg)
ImportError: Missing optional dependency 'tabulate'. Use pip or conda to install tabulate.
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
t=pd.read_csv('tegs.csv');t=t[(t.Player=='Gregg WILLIAMS')&(t.Complete==True)]
t['PerRound']=(t.Sc/t.Rounds).round(1)
r=t.sort_values('GrossVP')[['TEGNum','Year','Area','Rounds','Sc','GrossVP','PerRound']].head(6)
print('|'+'|'.join(r.columns)+'|');print('|'+'---|'*len(r.columns))
for x in r.values:print('|'+'|'.join(map(str,x))+'|')"
→ |TEGNum|Year|Area|Rounds|Sc|GrossVP|PerRound| |---|---|---|---|---|---|---| |19|2026|Algarve, Portugal|4|346.0|58.0|86.5| |15|2022|Lisbon Coast, Portugal|4|347.0|63.0|86.8| |18|2025|Catalonia, Spain|4|354.0|66.0|88.5| |16|2023|Lisbon Coast, Portugal|4|350.0|66.0|87.5| |13|2020|Kent, England|4|355.0|68.0|88.8| |14|2021|Kent, England|4|363.0|76.0|90.8|
How many times has a player gone under 90 for all four rounds
A player has broken 90 in all four rounds of a TEG 2 times:
- Gregg WILLIAMS, TEG 19: his worst round was 89, and he totalled 346.
- Jon BAKER, TEG 13: his worst round was 88, and he totalled 337.
If "under 90" is read as 90 or better, the count rises to 4.
How I worked this out: I counted player-TEG combinations where all four rounds were completed 18-hole rounds and every round's gross strokes were below 90. TEG 2 had only three rounds, so it can't count.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv');r=r[r.Holes==18]
g=r.groupby(['Player','TEGNum']).agg(n=('Sc','size'),mx=('Sc','max'),tot=('Sc','sum'))
a=g[(g.n==4)&(g.mx<90)]
print(a);print(a.groupby('Player').size())
print(g[(g.n==4)&(g.mx<=90)].shape)
"
→ n mx tot Player TEGNum Gregg WILLIAMS 19 4 89.0 346.0 Jon BAKER 13 4 88.0 337.0 Player Gregg WILLIAMS 1 Jon BAKER 1 dtype: int64 (4, 3)
Who has the longest run without a double bogey?
Jon BAKER has the longest run without a double bogey or worse: 23 holes. He did it in TEG 7, from round 1 hole 8 to round 2 hole 12.
The next longest runs:
- Gregg WILLIAMS: 22 holes
- David MULLIN: 21 holes
- Stuart NEUMANN: 20 holes
I counted a "double bogey" as double bogey or worse. Streaks run across rounds and TEGs. See the streak records.
Show the workings (1 step)
Lookup: get_streak_records
{
"direction": "good"
}
→ {
"direction": "good",
"longest_streak_per_player": [
{
"Player": "Jon BAKER",
"Birdies": "2",
"Pars": "6",
"No +2s": "23",
"No TBPs": "41"
},
{
"Player": "Gregg WILLIAMS",
"Birdies": "2",
"Pars": "5",
"No +2s": "22",
"No TBPs": "44"
},
{
"Player": "David MULLIN",
"Birdies": "2",
"Pars": "5",
"No +2s": "21",
"No TBPs": "59"
},
{
"Player": "Stuart NEUMANN",
"Birdies": "1",
"Pars": "5",
"No +2s": "20",
"No TBPs": "25"
},
{
"Player": "Henry MELLER",
"Birdies": "2",
"Pars": "5",
"No +2s": "18",
"No TBPs": "42"
},
{
"Player": "John PATTERSON",
"Birdies": "1",
"Pars": "4",
"No +2s": "13",
"No TBPs": "27"
},
{
"Player": "Alex BAKER",
"Birdies": "1",
"Pars": "3",
"No +2s": "8",
"No TBPs": "28"
}
],
"all_time_records": [
{
"Streak Type": "Birdies",
"Record": "2",
"Player": "David MULLIN",
"When": "TEG 17 R2, H9 to H10"
},
{
"Streak Type": "Birdies",
"Record": "2",
"Player": "Gregg WILLIAMS",
"When": "TEG 15 R1, H3 to H4"
},
{
"Streak Type": "Birdies",
"Record": "2",
"Player": "Henry MELLER",
"When": "TEG 3 R3, H11 to H12"
},
{
"Streak Type": "Birdies",
"Record": "2",
"Player": "Jon BAKER",
"When": "TEG 17 R4, H4 to H5"
},
{
"Streak Type": "Pars or Better",
"Record": "6",
"Player": "Jon BAKER",
"When": "TEG 10 R2, H5 to H10"
},
{
"Streak Type": "No +2s",
"Record": "23",
"Player": "Jon BAKER",
"When": "TEG 7, R1 H8 to R2 H12"
},
{
"Streak Type": "No TBPs",
"Record": "59",
"Player": "David MULLIN",
"When": "TEG 10, R4 H1 to TEG 11, R3 H5"
}
],
"definition": "Streaks run across consecutive holes played, including across rounds and TEGs. Good: consecutive birdies, pars or better, holes without a double bogey or worse (+2s), holes without a triple bogey or worse (TBPs). Bad: holes without eagles/birdies, consecutive holes over par, +2s, TBPs.",
"page": "/scoring/streaks"
}
What is the record for most birdies on a TeG
Your question cut off after "on a TEG", so I've assumed you meant the most birdies in a TEG. The record is 6 birdies, and David MULLIN has set it twice:
- TEG 5 (Algarve, 2012)
- TEG 10 (Algarve, 2017)
If you meant the most birdies in a single round, that record is 4, held by Jon BAKER at TEG 14 Rd 4 (Prince's - Shore / Dunes, 2021).
The records page has the full list. The site counts eagles as birdies or better.
Show the workings (1 step)
Lookup: get_records
{
"scope": "score_counts"
}
→ {
"scope": "score_counts",
"records": [
{
"best": true,
"label": "Most Eagles in a TEG",
"value": "1",
"player": "Jon BAKER",
"when": "TEG 4 (East Sussex, England, 2011)"
},
{
"best": true,
"label": "Most Eagles in a TEG",
"value": "1",
"player": "Stuart NEUMANN",
"when": "TEG 12 (Catalonia, Spain, 2019)"
},
{
"best": true,
"label": "Most Eagles in a TEG",
"value": "1",
"player": "John PATTERSON",
"when": "TEG 13 (Kent, England, 2020)"
},
{
"best": true,
"label": "Most Eagles in a TEG",
"value": "1",
"player": "David MULLIN",
"when": "TEG 15 (Lisbon Coast, Portugal, 2022)"
},
{
"best": true,
"label": "Most Eagles in a Round",
"value": "1",
"player": "Jon BAKER",
"when": "TEG 4 Rd 4 (Bletchingley, Sep 2011)"
},
{
"best": true,
"label": "Most Eagles in a Round",
"value": "1",
"player": "Stuart NEUMANN",
"when": "TEG 12 Rd 4 (PGA Catalunya - Stadium, Oct 2019)"
},
{
"best": true,
"label": "Most Eagles in a Round",
"value": "1",
"player": "John PATTERSON",
"when": "TEG 13 Rd 4 (Prince's - Shore / Dunes, Oct 2020)"
},
{
"best": true,
"label": "Most Eagles in a Round",
"value": "1",
"player": "David MULLIN",
"when": "TEG 15 Rd 2 (West Cliffs, Oct 2022)"
},
{
"best": true,
"label": "Most Birdies in a TEG",
"value": "6",
"player": "David MULLIN",
"when": "TEG 5 (Algarve, Portugal, 2012)"
},
{
"best": true,
"label": "Most Birdies in a TEG",
"value": "6",
"player": "David MULLIN",
"when": "TEG 10 (Algarve, Portugal, 2017)"
},
{
"best": true,
"label": "Most Birdies in a Round",
"value": "4",
"player": "Jon BAKER",
"when": "TEG 14 Rd 4 (Prince's - Shore / Dunes, Nov 2021)"
},
{
"best": true,
"label": "Most Pars in a TEG",
"value": "34",
"player": "David MULLIN",
"when": "TEG 10 (Algarve, Portugal, 2017)"
},
{
"best": true,
"label": "Most Pars in a TEG",
"value": "34",
"player": "Jon BAKER",
"when": "TEG 13 (Kent, England, 2020)"
},
{
"best": true,
"label": "Most Pars in a Round",
"value": "11",
"player": "Henry MELLER",
"when": "TEG 7 Rd 1 (Praia D'El Rey, Oct 2014)"
},
{
"best": true,
"label": "Most Pars in a Round",
"value": "11",
"player": "David MULLIN",
"when": "TEG 7 Rd 2 (Bom Sucesso, Oct 2014)"
},
{
"best": true,
"label": "Most Pars in a Round",
"value": "11",
"player": "David MULLIN",
"when": "TEG 10 Rd 1 (Boavista, Sep 2017)"
},
{
"best": true,
"label": "Most Pars in a Round",
"value": "11",
"player": "Jon BAKER",
"when": "TEG 10 Rd 2 (Palmares - Praia / Alvor, Oct 2017)"
},
{
"best": true,
"label": "Most Pars in a Round",
"value": "11",
"player": "Gregg WILLIAMS",
"when": "TEG 13 Rd 3 (Littlestone, Oct 2020)"
},
{
"best": true,
"label": "Most Pars in a Round",
"value": "11",
"player": "Jon BAKER",
"when": "TEG 15 Rd 3 (Penha Longa, Oct 2022)"
},
{
"best": true,
"label": "Most Pars in a Round",
"value": "11",
"player": "Gregg WILLIAMS",
"when": "TEG 16 Rd 4 (Estoril, Oct 2023)"
},
{
"best": true,
"label": "Most Pars in a Round",
"value": "11",
"player": "Jon BAKER",
"when": "TEG 19 Rd 2 (Monte Rei, Oct 2026)"
},
{
"best": false,
"label": "Most TBPs in a TEG",
"value": "28",
"player": "Gregg WILLIAMS",
"when": "TEG 4 (East Sussex, England, 2011)"
},
{
"best": false,
"label": "Most TBPs in a Round",
"value": "9",
"player": "Henry MELLER",
"when": "TEG 2 Rd 1 (Boavista, May 2009)"
},
{
"best": false,
"label": "Most TBPs in a Round",
"value": "9",
"player": "Gregg WILLIAMS",
"when": "TEG 4 Rd 4 (Bletchingley, Sep 2011)"
},
{
"best": false,
"label": "Most TBPs in a Round",
"value": "9",
"player": "Alex BAKER",
"when": "TEG 8 Rd 3 (Quinta da Marinha, Nov 2015)"
},
{
"best": false,
"label": "Most TBPs in a Round",
"value": "9",
"player": "Alex BAKER",
"when": "TEG 9 Rd 1 (Royal \u00d3bidos, Sep 2016)"
},
{
"best": false,
"label": "Most TBPs in a Round",
"value": "9",
"player": "John PATTERSON",
"when": "TEG 17 Rd 2 (West Cliffs, Oct 2024)"
}
],
"notes": [
"Eagles, birdies and pars counts include better scores."
],
"page": "/records?tab=score_counts"
}
Roughly what gross score differential do Alex and Gregg need against Jon to level the stableford competition in the final round
To level the Stableford in round 4, Alex needs to finish about 3 strokes better than Jon gross, and Gregg about 14 strokes better.
Jon is on 123 points, Gregg on 113 and Alex on 110, so Alex is 13 points behind and Gregg 10. Alex gets 10 more handicap strokes than Jon, and Gregg gets 4 fewer. Those strokes cover part of each gap.
| Player | Points behind Jon | Handicap vs Jon | Gross they must beat Jon by (about) |
|---|---|---|---|
| Alex BAKER | 13 | +10 | 3 |
| Gregg WILLIAMS | 10 | −4 | 14 |
- Alex's task looks modest. Even so, Alex is 98 over par gross so far, so it would take a big round.
- Gregg's task is much harder. He would need roughly 14 strokes better than whatever Jon shoots.
- Jon is 97.4% to win the Trophy on the live chances page.
How I worked this out: I used the approximation that Stableford points for a round ≈ handicap + 36 − gross vs par. This is the site's own link between points and gross, and it ignores any holes that score 0 points. I took standings after 3 rounds (Thru 0, nobody mid-round) and compared each player's gap in points with their handicap difference from Jon. Ties would need roughly one stroke less in each case, and the exact figure depends on which holes the shots land on.
Show the workings (3 steps)
Lookup: get_live_win_chances
{}
→ {
"teg_num": 19,
"rounds_done": 3,
"rounds_total": 4,
"live_round": null,
"simulations": 10000,
"measure_label": "Stableford",
"players": [
{
"Player": "Jon BAKER",
"Pl": "JB",
"Handicap": 21,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 123,
"GrossVsParSoFar": 49,
"NetPosition": "1",
"GrossPosition": "2",
"TrophyChancePct": 97.4,
"JacketChancePct": 29.3,
"SpoonChancePct": 0.0,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 18.0
},
{
"Player": "Gregg WILLIAMS",
"Pl": "GW",
"Handicap": 17,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 113,
"GrossVsParSoFar": 47,
"NetPosition": "2",
"GrossPosition": "1",
"TrophyChancePct": 2.1,
"JacketChancePct": 70.7,
"SpoonChancePct": 0.0,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 16.5
},
{
"Player": "Alex BAKER",
"Pl": "AB",
"Handicap": 31,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 110,
"GrossVsParSoFar": 98,
"NetPosition": "3",
"GrossPosition": "5",
"TrophyChancePct": 0.5,
"JacketChancePct": 0.0,
"SpoonChancePct": 0.0,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 33.4
},
{
"Player": "David MULLIN",
"Pl": "DM",
"Handicap": 19,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 95,
"GrossVsParSoFar": 73,
"NetPosition": "4",
"GrossPosition": "3",
"TrophyChancePct": 0.0,
"JacketChancePct": 0.0,
"SpoonChancePct": 31.9,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 22.9
},
{
"Player": "John PATTERSON",
"Pl": "JP",
"Handicap": 25,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 92,
"GrossVsParSoFar": 97,
"NetPosition": "5",
"GrossPosition": "4",
"TrophyChancePct": 0.0,
"JacketChancePct": 0.0,
"SpoonChancePct": 68.1,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 30.0
}
],
"definition": "Win chances now for the TEG in progress. Completed rounds and the holes entered so far in the round being played are banked at their actual scores; the rest is simulated hole by hole from each player's form, blended with how they are playing this TEG. Trophy = most Stableford points (net vs par before TEG 8), Green Jacket = lowest gross, Wooden Spoon = fewest Stableford points. Change is in percentage points since the last finalised round. Thru = holes entered in the live round.",
"notes": [
"A prediction, not a certainty.",
"Mid-round figures move as scores come in. When live_round is set they include scores entered but not yet finalised; otherwise they are as at the last finalised round.",
"Positions share ties (2=)."
],
"page": "/simulation?tab=live"
}
Lookup: get_what_it_takes
{
"competition": "trophy"
}
→ {
"players": [
{
"teg": 19,
"player": "AB",
"name": "Alex BAKER",
"competition": "trophy",
"measure": "Stableford",
"rivals": "same_pace",
"rounds_done": 3,
"live_round": null,
"holes_left": 18,
"now": {
"total": 110,
"gross_vp_now": 98
},
"projections": [
{
"player": "JB",
"name": "Jon BAKER",
"now": 123,
"projected": 164.0
},
{
"player": "GW",
"name": "Gregg WILLIAMS",
"now": 113,
"projected": 150.67
},
{
"player": "AB",
"name": "Alex BAKER",
"now": 110,
"projected": 146.67
},
{
"player": "DM",
"name": "David MULLIN",
"now": 95,
"projected": 126.67
},
{
"player": "JP",
"name": "John PATTERSON",
"now": 92,
"projected": 122.67
}
],
"best_rival": {
"player": "JB",
"name": "Jon BAKER",
"projected": 164.0
},
"need": {
"tie": {
"target_total": 164,
"points": 54,
"net_vs_par": null,
"gross_vp": 13,
"gross": 85,
"per_round_gross": 85.0,
"per_round_points": 54.0
},
"outright": {
"target_total": 165,
"points": 55,
"net_vs_par": null,
"gross_vp": 12,
"gross": 84,
"per_round_gross": 84.0,
"per_round_points": 55.0
}
},
"reality": {
"best_round_vp": 17,
"best_round_gross": 89,
"rounds_recent": 12,
"rounds_recent_at_or_better": 0,
"recent_tegs": [
16,
17,
18
]
},
"out_of_reach": true,
"no_holes_left": false,
"notes": [
"Rivals keep their pace so far: total so far plus their average per hole times holes left.",
"Points and gross are linked by gross vs par = strokes received + 2 per hole - points; about right (exact unless a hole scores 0 points)."
],
"already_enough": false
},
{
"teg": 19,
"player": "DM",
"name": "David MULLIN",
"competition": "trophy",
"measure": "Stableford",
"rivals": "same_pace",
"rounds_done": 3,
"live_round": null,
"holes_left": 18,
"now": {
"total": 95,
"gross_vp_now": 73
},
"projections": [
{
"player": "JB",
"name": "Jon BAKER",
"now": 123,
"projected": 164.0
},
{
"player": "GW",
"name": "Gregg WILLIAMS",
"now": 113,
"projected": 150.67
},
{
"player": "AB",
"name": "Alex BAKER",
"now": 110,
"projected": 146.67
},
{
"player": "DM",
"name": "David MULLIN",
"now": 95,
"projected": 126.67
},
{
"player": "JP",
"name": "John PATTERSON",
"now": 92,
"projected": 122.67
}
],
"best_rival": {
"player": "JB",
"name": "Jon BAKER",
"projected": 164.0
},
"need": {
"tie": {
"target_total": 164,
"points": 69,
"net_vs_par": null,
"gross_vp": -14,
"gross": 58,
"per_round_gross": 58.0,
"per_round_points": 69.0
},
"outright": {
"target_total": 165,
"points": 70,
"net_vs_par": null,
"gross_vp": -15,
"gross": 57,
"per_round_gross": 57.0,
"per_round_points": 70.0
}
},
"reality": {
"best_round_vp": 4,
"best_round_gross": 76,
"rounds_recent": 12,
"rounds_recent_at_or_better": 0,
"recent_tegs": [
16,
17,
18
]
},
"out_of_reach": true,
"no_holes_left": false,
"notes": [
"Rivals keep their pace so far: total so far plus their average per hole times holes left.",
"Points and gross are linked by gross vs par = strokes received + 2 per hole - points; about right (exact unless a hole scores 0 points)."
],
"already_enough": false
},
{
"teg": 19,
"player": "GW",
"name": "Gregg WILLIAMS",
"competition": "trophy",
"measure": "Stableford",
"rivals": "same_pace",
"rounds_done": 3,
"live_round": null,
"holes_left": 18,
"now": {
"total": 113,
"gross_vp_now": 47
},
"projections": [
{
"player": "JB",
"name": "Jon BAKER",
"now": 123,
"projected": 164.0
},
{
"player": "GW",
"name": "Gregg WILLIAMS",
"now": 113,
"projected": 150.67
},
{
"player": "AB",
"name": "Alex BAKER",
"now": 110,
"projected": 146.67
},
{
"player": "DM",
"name": "David MULLIN",
"now": 95,
"projected": 126.67
},
{
"player": "JP",
"name": "John PATTERSON",
"now": 92,
"projected": 122.67
}
],
"best_rival": {
"player": "JB",
"name": "Jon BAKER",
"projected": 164.0
},
"need": {
"tie": {
"target_total": 164,
"points": 51,
"net_vs_par": null,
"gross_vp": 2,
"gross": 74,
"per_round_gross": 74.0,
"per_round_points": 51.0
},
"outright": {
"target_total": 165,
"points": 52,
"net_vs_par": null,
"gross_vp": 1,
"gross": 73,
"per_round_gross": 73.0,
"per_round_points": 52.0
}
},
"reality": {
"best_round_vp": 10,
"best_round_gross": 82,
"rounds_recent": 12,
"rounds_recent_at_or_better": 0,
"recent_tegs": [
16,
17,
18
]
},
"out_of_reach": true,
"no_holes_left": false,
"notes": [
"Rivals keep their pace so far: total so far plus their average per hole times holes left.",
"Points and gross are linked by gross vs par = strokes received + 2 per hole - points; about right (exact unless a hole scores 0 points)."
],
"already_enough": false
},
{
"teg": 19,
"player": "JB",
"name": "Jon BAKER",
"competition": "trophy",
"measure": "Stableford",
"rivals": "same_pace",
"rounds_done": 3,
"live_round": null,
"holes_left": 18,
"now": {
"total": 123,
"gross_vp_now": 49
},
"projections": [
{
"player": "JB",
"name": "Jon BAKER",
"now": 123,
"projected": 164.0
},
{
"player": "GW",
"name": "Gregg WILLIAMS",
"no
Code run in the sandbox
python3 -c "
hc={'J':21,'A':31,'G':17};pts={'J':123,'A':110,'G':113}
for p in 'AG':print(p,(pts['J']-pts[p])-(hc[p]-hc['J'])) "
→ A 3 G 14
What are the wooden spoon odds now after 3 rounds of teg 19
John PATTERSON is the favourite for the Wooden Spoon at 68.1%. David MULLIN is on 31.9%. The other three players are at 0.0%.
| Player | Stableford after 3 rounds | Spoon chance |
|---|---|---|
| John PATTERSON | 92 | 68.1% |
| David MULLIN | 95 | 31.9% |
| Alex BAKER | 110 | 0.0% |
| Gregg WILLIAMS | 113 | 0.0% |
| Jon BAKER | 123 | 0.0% |
Only 3 Stableford points separate the bottom two, so round 4 decides it. These figures come from the live chances simulator. They are as at the end of round 3, with no round in progress.
How I worked this out: the simulator banks the completed rounds at their actual scores. It plays round 4 hole by hole from each player's form, blended with how they are playing this TEG. It ran 10,000 simulations. The Wooden Spoon goes to the player with the fewest Stableford points.
Show the workings (1 step)
Lookup: get_live_win_chances
{}
→ {
"teg_num": 19,
"rounds_done": 3,
"rounds_total": 4,
"live_round": null,
"simulations": 10000,
"measure_label": "Stableford",
"players": [
{
"Player": "Jon BAKER",
"Pl": "JB",
"Handicap": 21,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 123,
"GrossVsParSoFar": 49,
"NetPosition": "1",
"GrossPosition": "2",
"TrophyChancePct": 97.4,
"JacketChancePct": 29.3,
"SpoonChancePct": 0.0,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 18.0
},
{
"Player": "Gregg WILLIAMS",
"Pl": "GW",
"Handicap": 17,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 113,
"GrossVsParSoFar": 47,
"NetPosition": "2",
"GrossPosition": "1",
"TrophyChancePct": 2.1,
"JacketChancePct": 70.7,
"SpoonChancePct": 0.0,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 16.5
},
{
"Player": "Alex BAKER",
"Pl": "AB",
"Handicap": 31,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 110,
"GrossVsParSoFar": 98,
"NetPosition": "3",
"GrossPosition": "5",
"TrophyChancePct": 0.5,
"JacketChancePct": 0.0,
"SpoonChancePct": 0.0,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 33.4
},
{
"Player": "David MULLIN",
"Pl": "DM",
"Handicap": 19,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 95,
"GrossVsParSoFar": 73,
"NetPosition": "4",
"GrossPosition": "3",
"TrophyChancePct": 0.0,
"JacketChancePct": 0.0,
"SpoonChancePct": 31.9,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 22.9
},
{
"Player": "John PATTERSON",
"Pl": "JP",
"Handicap": 25,
"Thru": 0,
"HolesPlayed": 54,
"NetSoFar": 92,
"GrossVsParSoFar": 97,
"NetPosition": "5",
"GrossPosition": "4",
"TrophyChancePct": 0.0,
"JacketChancePct": 0.0,
"SpoonChancePct": 68.1,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 30.0
}
],
"definition": "Win chances now for the TEG in progress. Completed rounds and the holes entered so far in the round being played are banked at their actual scores; the rest is simulated hole by hole from each player's form, blended with how they are playing this TEG. Trophy = most Stableford points (net vs par before TEG 8), Green Jacket = lowest gross, Wooden Spoon = fewest Stableford points. Change is in percentage points since the last finalised round. Thru = holes entered in the live round.",
"notes": [
"A prediction, not a certainty.",
"Mid-round figures move as scores come in. When live_round is set they include scores entered but not yet finalised; otherwise they are as at the last finalised round.",
"Positions share ties (2=)."
],
"page": "/simulation?tab=live"
}
If we run the predictatron using only this tournament’s rounds, what would the percentages be for wooden spoon and trophy
I can't run that. The simulation tool only reads finished TEGs as form history, and TEG 19 is still in progress. When I tried "TEG 19 only", it returned no history to work from. It also can't stage the round 3 scores into the form data.
The closest figures I can give are the live chances from my last answer. They bank the actual round 1–3 scores and simulate round 4 from each player's normal form plus how they are playing this TEG. In that run John PATTERSON was on 68.1% for the Wooden Spoon and David MULLIN on 31.9%. I didn't pull the Trophy chances then. I can fetch them if you like.
How I worked this out: I tried the simulator's "only these TEGs" option with TEG 19 alone. It stopped with an error because there was no finished history to use.
Show the workings (2 steps)
Code run in the sandbox
mkdir -p /tmp/teg && python3 -m zipfile -e "$INPUT_DIR/teg_toolkit.zip" /tmp/teg && python3 -m zipfile -e "$INPUT_DIR/teg_data.zip" /tmp/teg && cat /tmp/teg/skills/teg-simulation/SKILL.md
→ ---
name: teg-simulation
description: Run the site's Predictatron Monte Carlo engine for the next TEG. Use for forecasts, odds, "why is X favourite", "what if X's handicap were Y / last TEG was ignored", handicap-change impact and equalising handicaps.
---
# TEG simulation (Predictatron)
The next TEG is simulated 10,000 times from each player's recent hole-by-hole form, on that TEG's courses and handicaps. This is the same engine as the site's /simulation page. Never reimplement it or do the arithmetic yourself. Run the script and report its numbers.
For a plain "who will win" or "what are the odds" question, use the `get_predictions` lookup instead. It gives the same numbers as `baseline`. Use this skill for why, what-if and attribution questions.
While a TEG is in progress, this skill only knows finalised rounds and has no live staging, so its figures are the pre-tournament or last-round view. Chances now, including holes entered mid-round, come from the `get_live_win_chances` lookup. "What does X need to win" comes from `get_what_it_takes`. Use those, not `baseline`, for live and mid-TEG questions, and quote their numbers.
Setup (once per container) is in the system prompt. Then:
```
python /tmp/teg/skills/teg-simulation/scripts/sim.py <command> --json [options]
```
Defaults: seed 1, 10,000 sims (the same as `get_predictions`, so `baseline` matches it exactly), the next TEG's roster, saved (or draft) handicaps. The same command always returns the same numbers. The site page uses a fresh seed, so its figures can differ by a point or two.
## Commands
- `baseline`: the site's default run. Per player: Trophy %, Jacket %, Spoon %, mean Stableford, mean gross vs par, expected finish, and the full finishing-position table.
- `whatif`: a changed run next to the baseline, same roster/seed/sims. Reports the change in Trophy %, Jacket % and mean Stableford.
- `handicap-impact`: how the handicap changes since the previous TEG moved each player's Trophy chance, split fairly between the movers (Shapley).
- `equalising`: the handicap that would give each player about 36 Stableford points a round.
`whatif`, `handicap-impact` and `equalising` take these options:
- `--handicap GW=18` (repeatable): override a handicap. Player = code or name.
- `--exclude-tegs 18,17`: drop TEGs from form history. Recency weights then fall to the next most recent TEGs.
- `--only-tegs 16,17,18`: use only these TEGs.
- `--weights 50,35,15`: recency weights, newest TEG first. Default 50,35,15; older TEGs get 0.
- `--teg-weights 18:0,17:50`: explicit weight per TEG; the rest keep their default.
- `--method window|bands`, `--min-holes`, `--shrinkage` (bands only), `--field-alpha`: model settings. Leave at defaults unless asked.
- Common to all: `--target TEG`, `--players A,B,C`, `--n-sims`, `--seed`.
Python use: `sys.path.insert(0, "/tmp/teg/skills/teg-simulation/scripts"); import sim; sim.whatif(handicaps={"GW": 18})`. It returns the same dict as `--json`.
Errors print a one-line message (for example, an unknown player lists the valid ones). Fix the input and rerun.
## Reading the numbers
- Trophy % = share of sims where the player had the most Stableford points. Jacket % = lowest gross. Spoon % = fewest Stableford points. Each column sums to 100.
- `TrophySEpp` = Monte Carlo standard error in percentage points (about 0.5 near 50% at 10k sims). Differences smaller than about 1 pp between players, or between runs, are noise. Raise `--n-sims` (max 100,000) to tighten.
- Delta columns are scenario minus baseline, in percentage points, on paired random draws. They are steadier than the levels, but still only trust deltas above about 0.5 pp.
- `MeanGrossVP` is the expected gross total relative to par for the whole TEG.
## Method and caveats
- Each hole score is drawn from the player's own history for that par and stroke index, weighted by recency (default: last TEG 50, one before 35, one before that 15). Rare outcomes (eagles, blobs) are blended in from the whole field. This models form, not luck of the day or injuries.
- Handicaps only change the strokes received. A handicap change moves Stableford totals, not gross. So Jacket % does not move with handicap overrides.
- Effects are not additive. Changing two handicaps is not the sum of changing each. Use `handicap-impact` for attribution.
- If the TEG has no scorecard yet, each sim draws random par-72 courses. The notes in the output say so. Handicaps may be a draft calculation.
- Trophy = Stableford applies from TEG 8. All upcoming TEGs are in that era.
- This predicts a future TEG. It cannot re-run a past TEG's actual result.
## Recipes
- Why is X the favourite? Run `baseline`. Then `whatif --exclude-tegs <latest>` and `whatif --handicap X=<hc+3>` to see which input carries the lead. Quote both deltas.
- What if X's handicap were Y? `whatif --handicap X=Y`. Report X's and the field's Trophy % before and after.
- How much does last TEG matter? `whatif --teg-weights <last>:0` (or `--exclude-tegs <last>`) and compare. Also try `--weights 100,0,0` for "only last TEG".
- Who gained most from the handicap changes? `handicap-impact`. Use the `from_<code>_PP` columns for who caused what.
- Are the handicaps fair? `equalising`: compare CurrentHC and EqualisingHC, and Unrounded for size.
- Is a gap real? Look at `TrophySEpp`, or rerun with `--n-sims 100000` or another `--seed`.
Code run in the sandbox
cd /tmp/teg && python skills/teg-simulation/scripts/sim.py whatif --json --only-tegs 19 2>&1 | head -c 3000
→ {"error": "No history left after only_tegs=[19]."}
What is each persons chance of winning the wooden spoon now, having completed 2 rounds of teg 19
John PATTERSON is the heavy favourite for the Wooden Spoon at 83.7%, according to the live chances after 2 rounds of TEG 19.
| Player | Spoon chance | Stableford so far |
|---|---|---|
| John PATTERSON | 83.7% | 56 |
| David MULLIN | 9.8% | 65 |
| Alex BAKER | 6.4% | 67 |
| Jon BAKER | 0.0% | 81 |
| Gregg WILLIAMS | 0.0% | 76 |
John is last on Stableford, 9 points behind David and 11 behind Alex. David and Alex have a small chance of catching him. Jon and Gregg are well clear.
How I worked this out: these are the simulator's figures from 10,000 runs. It simulates the remaining holes from each player's form, blended with how they are playing this TEG. The Wooden Spoon goes to the fewest Stableford points. Both rounds are finalised, so no round is live. It is a prediction, not a certainty.
Show the workings (1 step)
Lookup: get_live_win_chances
{}
→ {
"teg_num": 19,
"rounds_done": 2,
"rounds_total": 4,
"live_round": null,
"simulations": 10000,
"measure_label": "Stableford",
"players": [
{
"Player": "Jon BAKER",
"Pl": "JB",
"Handicap": 21,
"Thru": 0,
"HolesPlayed": 36,
"NetSoFar": 81,
"GrossVsParSoFar": 34,
"NetPosition": "1",
"GrossPosition": "2",
"TrophyChancePct": 80.2,
"JacketChancePct": 17.9,
"SpoonChancePct": 0.0,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 18.8
},
{
"Player": "Gregg WILLIAMS",
"Pl": "GW",
"Handicap": 17,
"Thru": 0,
"HolesPlayed": 36,
"NetSoFar": 76,
"GrossVsParSoFar": 30,
"NetPosition": "2",
"GrossPosition": "1",
"TrophyChancePct": 19.2,
"JacketChancePct": 82.1,
"SpoonChancePct": 0.0,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 16.3
},
{
"Player": "Alex BAKER",
"Pl": "AB",
"Handicap": 31,
"Thru": 0,
"HolesPlayed": 36,
"NetSoFar": 67,
"GrossVsParSoFar": 72,
"NetPosition": "3",
"GrossPosition": "4=",
"TrophyChancePct": 0.5,
"JacketChancePct": 0.0,
"SpoonChancePct": 6.4,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 35.3
},
{
"Player": "David MULLIN",
"Pl": "DM",
"Handicap": 19,
"Thru": 0,
"HolesPlayed": 36,
"NetSoFar": 65,
"GrossVsParSoFar": 48,
"NetPosition": "4",
"GrossPosition": "3",
"TrophyChancePct": 0.1,
"JacketChancePct": 0.0,
"SpoonChancePct": 9.8,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 22.4
},
{
"Player": "John PATTERSON",
"Pl": "JP",
"Handicap": 25,
"Thru": 0,
"HolesPlayed": 36,
"NetSoFar": 56,
"GrossVsParSoFar": 72,
"NetPosition": "5",
"GrossPosition": "4=",
"TrophyChancePct": 0.0,
"JacketChancePct": 0.0,
"SpoonChancePct": 83.7,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 31.2
}
],
"definition": "Win chances now for the TEG in progress. Completed rounds and the holes entered so far in the round being played are banked at their actual scores; the rest is simulated hole by hole from each player's form, blended with how they are playing this TEG. Trophy = most Stableford points (net vs par before TEG 8), Green Jacket = lowest gross, Wooden Spoon = fewest Stableford points. Change is in percentage points since the last finalised round. Thru = holes entered in the live round.",
"notes": [
"A prediction, not a certainty.",
"Mid-round figures move as scores come in. When live_round is set they include scores entered but not yet finalised; otherwise they are as at the last finalised round.",
"Positions share ties (2=)."
],
"page": "/simulation?tab=live"
}
What is the worst opening 3 holes (gross vs par) that any player has had in any round. Give me the worst 5.
The worst opening three holes in any round are +10 vs par, and six rounds share that mark. That means the top 5 has a six-way tie at the cut-off, so here are all six:
| Player | TEG | Round | Course | Strokes (holes 1-3) | Vs par |
|---|---|---|---|---|---|
| Stuart NEUMANN | 15 | 1 | Estoril | 8-4-9 | +10 |
| Alex BAKER | 16 | 4 | Estoril | 9-4-8 | +10 |
| Gregg WILLIAMS | 4 | 2 | Lingfield | 6-7-9 | +10 |
| Jon BAKER | 7 | 3 | Royal Óbidos | 5-10-7 | +10 |
| John PATTERSON | 15 | 4 | Oitavos Dunes | 10-6-5 | +10 |
| Alex BAKER | 12 | 1 | PGA Catalunya - Tour | 7-7-7 | +10 |
Alex BAKER appears twice. Estoril also appears twice, which suggests it's a hard opening.
How I worked this out: I summed gross vs par over holes 1-3 for every player-round with all three holes played (360 rounds). The tie at +10 has no separation, so I've listed every tied round rather than cutting it to five.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
f=h[h.Hole<=3].groupby(['Player','TEGNum','Round','Course']).agg(g=('GrossVP','sum'),n=('Hole','count'),s=('Sc',lambda x:list(x))).reset_index()
f=f[f.n==3].sort_values('g',ascending=False)
c=f.g.iloc[4];print(len(f),c)
print(f[f.g>=c].to_markdown(index=False))
"
→ 360 10.0
Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 135, in import_optional_dependency
module = importlib.import_module(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1140, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'tabulate'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<string>", line 7, in <module>
File "/usr/local/lib/python3.11/site-packages/pandas/util/_decorators.py", line 333, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/core/frame.py", line 2988, in to_markdown
tabulate = import_optional_dependency("tabulate")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 138, in import_optional_dependency
raise ImportError(msg)
ImportError: Missing optional dependency 'tabulate'. Use pip or conda to install tabulate.
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
f=h[h.Hole<=3].groupby(['Player','TEGNum','Round','Course']).agg(g=('GrossVP','sum'),n=('Hole','count'),s=('Sc',lambda x:'-'.join(map(str,x)))).reset_index()
f=f[f.n==3].sort_values('g',ascending=False)
f=f[f.g>=10]
print('|Player|TEG|Round|Course|Strokes|Vs par|\n|-|-|-|-|-|-|')
for r in f.itertuples():print(f'|{r.Player}|{r.TEGNum}|{r.Round}|{r.Course}|{r.s}|+{r.g}|')
print(len(f))
"
→ |Player|TEG|Round|Course|Strokes|Vs par| |-|-|-|-|-|-| |Stuart NEUMANN|15|1|Estoril|8.0-4.0-9.0|+10.0| |Alex BAKER|16|4|Estoril|9.0-4.0-8.0|+10.0| |Gregg WILLIAMS|4|2|Lingfield|6.0-7.0-9.0|+10.0| |Jon BAKER|7|3|Royal Óbidos|5.0-10.0-7.0|+10.0| |John PATTERSON|15|4|Oitavos Dunes|10.0-6.0-5.0|+10.0| |Alex BAKER|12|1|PGA Catalunya - Tour|7.0-7.0-7.0|+10.0| 6
What are Greggs best ever first round scores
Gregg's best opening round by gross was 84 (+12) at PGA Catalunya - Tour in TEG 12. That round also gave him 46 Stableford points.
| TEG | Course | Strokes | Vs par | Stableford | Net vs par |
|---|---|---|---|---|---|
| 12 | PGA Catalunya - Tour | 84 | +12 | 46 | -10 |
| 15 | Estoril | 86 | +17 | 39 | -1 |
| 19 | Quinta da Ria | 87 | +15 | 38 | -2 |
| 16 | Oitavos Dunes | 89 | +18 | 34 | +2 |
| 6 | PGA Catalunya - Tour | 92 | +20 | 48 | -12 |
By Stableford, his best first round was the 48 points in TEG 6, even though that was a 92 gross. The table shows only his top five by gross, and the 87 at Quinta da Ria is this year's opening round. You can see every round on the all rounds page.
How I worked this out: I used all 18 of Gregg's completed first rounds, which includes TEG 19, and ranked them by gross strokes. Stableford is shown alongside.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv');g=r[(r.Player=='Gregg WILLIAMS')&(r.Round==1)&(r.Holes==18)]
print(len(g))
print(g.sort_values('Sc')[['TEGNum','Course','Sc','GrossVP','Stableford','NetVP']].to_markdown(index=False))
print(g.Sc.min(),g.Stableford.max())
"
→ 18
Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 135, in import_optional_dependency
module = importlib.import_module(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1140, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'tabulate'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<string>", line 5, in <module>
File "/usr/local/lib/python3.11/site-packages/pandas/util/_decorators.py", line 333, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/core/frame.py", line 2988, in to_markdown
tabulate = import_optional_dependency("tabulate")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 138, in import_optional_dependency
raise ImportError(msg)
ImportError: Missing optional dependency 'tabulate'. Use pip or conda to install tabulate.
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv');g=r[(r.Player=='Gregg WILLIAMS')&(r.Round==1)&(r.Holes==18)]
g=g.sort_values(['Sc','Stableford'],ascending=[True,False])[['TEGNum','Course','Sc','GrossVP','Stableford','NetVP']].head(6)
print('|TEG|Course|Strokes|Vs par|Stableford|Net vs par|\n|-|-|-|-|-|-|')
for x in g.values: print('|'+'|'.join(map(str,x))+'|')
"
→ |TEG|Course|Strokes|Vs par|Stableford|Net vs par| |-|-|-|-|-|-| |12|PGA Catalunya - Tour|84.0|12.0|46.0|-10.0| |15|Estoril|86.0|17.0|39.0|-1.0| |19|Quinta da Ria|87.0|15.0|38.0|-2.0| |16|Oitavos Dunes|89.0|18.0|34.0|2.0| |6|PGA Catalunya - Tour|92.0|20.0|48.0|-12.0| |14|Royal Cinque Ports|93.0|21.0|33.0|4.0|
Who is now favourite for wooden spoon in teg 19 after the 1st round
David MULLIN is the slight favourite for the Wooden Spoon at 46.4%. John PATTERSON is close behind on 42.9%. The two are level on 31 Stableford points after round 1, tied 4th= (the bottom of the field), so it's close to a two-horse race. These are the live chances.
| Player | Spoon chance | Stableford after R1 |
|---|---|---|
| David MULLIN | 46.4% | 31 |
| John PATTERSON | 42.9% | 31 |
| Alex BAKER | 6.6% | 38 |
| Gregg WILLIAMS | 2.9% | 38 |
| Jon BAKER | 1.2% | 39 |
Mullin's edge is probably down to his handicap of 19, against 25 for Patterson. Patterson gets more shots, but he is also expected to play worse than Mullin.
How I worked this out: these are the simulator's figures after the finalised round 1, with 10,000 simulations. The Wooden Spoon goes to the player with the fewest Stableford points after four rounds. Round 1 is a small sample, so the odds will move a lot.
Show the workings (1 step)
Lookup: get_live_win_chances
{}
→ {
"teg_num": 19,
"rounds_done": 1,
"rounds_total": 4,
"live_round": null,
"simulations": 10000,
"measure_label": "Stableford",
"players": [
{
"Player": "Jon BAKER",
"Pl": "JB",
"Handicap": 21,
"Thru": 0,
"HolesPlayed": 18,
"NetSoFar": 39,
"GrossVsParSoFar": 18,
"NetPosition": "1",
"GrossPosition": "2",
"TrophyChancePct": 49.5,
"JacketChancePct": 17.4,
"SpoonChancePct": 1.2,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 19.7
},
{
"Player": "Gregg WILLIAMS",
"Pl": "GW",
"Handicap": 17,
"Thru": 0,
"HolesPlayed": 18,
"NetSoFar": 38,
"GrossVsParSoFar": 15,
"NetPosition": "2=",
"GrossPosition": "1",
"TrophyChancePct": 27.0,
"JacketChancePct": 81.2,
"SpoonChancePct": 2.9,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 16.8
},
{
"Player": "Alex BAKER",
"Pl": "AB",
"Handicap": 31,
"Thru": 0,
"HolesPlayed": 18,
"NetSoFar": 38,
"GrossVsParSoFar": 30,
"NetPosition": "2=",
"GrossPosition": "4",
"TrophyChancePct": 20.0,
"JacketChancePct": 0.0,
"SpoonChancePct": 6.6,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 33.1
},
{
"Player": "John PATTERSON",
"Pl": "JP",
"Handicap": 25,
"Thru": 0,
"HolesPlayed": 18,
"NetSoFar": 31,
"GrossVsParSoFar": 33,
"NetPosition": "4=",
"GrossPosition": "5",
"TrophyChancePct": 2.1,
"JacketChancePct": 0.0,
"SpoonChancePct": 42.9,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 28.6
},
{
"Player": "David MULLIN",
"Pl": "DM",
"Handicap": 19,
"Thru": 0,
"HolesPlayed": 18,
"NetSoFar": 31,
"GrossVsParSoFar": 25,
"NetPosition": "4=",
"GrossPosition": "3",
"TrophyChancePct": 1.3,
"JacketChancePct": 1.3,
"SpoonChancePct": 46.4,
"TrophyChangePp": 0.0,
"JacketChangePp": 0.0,
"ExpectedGrossVsParPer18": 22.2
}
],
"definition": "Win chances now for the TEG in progress. Completed rounds and the holes entered so far in the round being played are banked at their actual scores; the rest is simulated hole by hole from each player's form, blended with how they are playing this TEG. Trophy = most Stableford points (net vs par before TEG 8), Green Jacket = lowest gross, Wooden Spoon = fewest Stableford points. Change is in percentage points since the last finalised round. Thru = holes entered in the live round.",
"notes": [
"A prediction, not a certainty.",
"Mid-round figures move as scores come in. When live_round is set they include scores entered but not yet finalised; otherwise they are as at the last finalised round.",
"Positions share ties (2=)."
],
"page": "/simulation?tab=live"
}
Who plays worse than their usual standard in round 1
Stuart NEUMANN is the one who most often starts badly. He averages 5.7 strokes worse than his usual in round 1 and gives up 4.8 Stableford points. Alex BAKER (+3.7) and John PATTERSON (+3.3) follow. David MULLIN is level, and Jon BAKER is slightly better in round 1.
| Player | Round 1 gross vs par | Other rounds | Round 1 pts | Other pts | Round 1s (n) | Gross diff | Pts diff |
|---|---|---|---|---|---|---|---|
| Stuart NEUMANN | 30.7 | 25.0 | 30.7 | 35.5 | 12 | 5.7 | -4.8 |
| Alex BAKER | 33.5 | 29.8 | 34.2 | 37.3 | 12 | 3.7 | -3.1 |
| John PATTERSON | 29.9 | 26.6 | 34.1 | 36.9 | 8 | 3.3 | -2.8 |
| Henry MELLER | 27.9 | 25.0 | 31.7 | 33.0 | 7 | 2.8 | -1.2 |
| Gregg WILLIAMS | 26.1 | 24.1 | 35.4 | 36.7 | 17 | 1.9 | -1.4 |
| David MULLIN | 18.7 | 18.7 | 35.3 | 35.6 | 17 | 0.0 | -0.3 |
| Jon BAKER | 19.8 | 20.7 | 36.8 | 35.8 | 17 | -0.9 | 0.9 |
How I worked this out: for each player I compared their average round 1 with their average over rounds 2–4, across all completed TEGs, using gross strokes vs par and Stableford points. Sample sizes are small (7 to 17 round 1s each), so the smaller gaps are within normal noise. Courses differ between round 1s and later rounds, and I haven't adjusted for that.
Show the workings (2 steps)
Code run in the sandbox
mkdir -p /tmp/teg && python3 -m zipfile -e "$INPUT_DIR/teg_toolkit.zip" /tmp/teg && python3 -m zipfile -e "$INPUT_DIR/teg_data.zip" /tmp/teg; cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv')
r['R1']=r.Round==1
# per player: R1 vs rounds 2-4 avg, within same TEG to net out handicap? use GrossVP and Stableford
g=r.groupby(['Player','R1']).agg(gvp=('GrossVP','mean'),sf=('Stableford','mean'),n=('Sc','size')).unstack()
o=pd.DataFrame({'R1 gross vs par':g['gvp'][True],'Other rounds':g['gvp'][False],'R1 pts':g['sf'][True],'Other pts':g['sf'][False],'n R1':g['n'][True]})
o['Gross diff']=o['R1 gross vs par']-o['Other rounds'];o['Pts diff']=o['R1 pts']-o['Other pts']
print(o.sort_values('Gross diff',ascending=False).round(1).to_markdown())
"
→ Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 135, in import_optional_dependency
module = importlib.import_module(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1140, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'tabulate'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<string>", line 9, in <module>
File "/usr/local/lib/python3.11/site-packages/pandas/util/_decorators.py", line 333, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/core/frame.py", line 2988, in to_markdown
tabulate = import_optional_dependency("tabulate")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 138, in import_optional_dependency
raise ImportError(msg)
ImportError: Missing optional dependency 'tabulate'. Use pip or conda to install tabulate.
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv')
r['R1']=r.Round==1
g=r.groupby(['Player','R1']).agg(gvp=('GrossVP','mean'),sf=('Stableford','mean'),n=('Sc','size')).unstack()
o=pd.DataFrame({'R1 gross vs par':g['gvp'][True],'Other rounds':g['gvp'][False],'R1 pts':g['sf'][True],'Other pts':g['sf'][False],'n R1':g['n'][True]})
o['Gross diff']=o['R1 gross vs par']-o['Other rounds'];o['Pts diff']=o['R1 pts']-o['Other pts']
o=o.sort_values('Gross diff',ascending=False).round(1).reset_index()
print('|'+'|'.join(o.columns)+'|');print('|'+'---|'*len(o.columns))
for _,x in o.iterrows():print('|'+'|'.join(map(str,x))+'|')
"
→ |Player|R1 gross vs par|Other rounds|R1 pts|Other pts|n R1|Gross diff|Pts diff| |---|---|---|---|---|---|---|---| |Stuart NEUMANN|30.7|25.0|30.7|35.5|12|5.7|-4.8| |Alex BAKER|33.5|29.8|34.2|37.3|12|3.7|-3.1| |John PATTERSON|29.9|26.6|34.1|36.9|8|3.3|-2.8| |Henry MELLER|27.9|25.0|31.7|33.0|7|2.8|-1.2| |Gregg WILLIAMS|26.1|24.1|35.4|36.7|17|1.9|-1.4| |David MULLIN|18.7|18.7|35.3|35.6|17|0.0|-0.3| |Jon BAKER|19.8|20.7|36.8|35.8|17|-0.9|0.9|
Based on the last 5 TEGs which player’s scoring is most erratic / hardest to predict? At both hole and round level.
At round level the most erratic player is Stuart NEUMANN, though he has only 8 rounds in this window. At hole level it is Alex BAKER, with a hole-by-hole spread of 1.45 strokes around his own average. Alex is the most erratic of the five regulars, with 20 rounds and 360 holes. Stuart's hole-by-hole spread (1.42) is almost as high.
| Player | Holes | Hole spread (SD) | Rounds | Round spread (SD) | Avg round vs par |
|---|---|---|---|---|---|
| Stuart NEUMANN | 144 | 1.42 | 8 | 7.53 | +29.12 |
| John PATTERSON | 216 | 1.38 | 12 | 6.98 | +27.67 |
| David MULLIN | 360 | 1.16 | 20 | 6.12 | +20.15 |
| Alex BAKER | 360 | 1.45 | 20 | 5.58 | +32.90 |
| Gregg WILLIAMS | 360 | 1.07 | 20 | 5.32 | +18.00 |
| Jon BAKER | 360 | 1.21 | 20 | 5.21 | +20.95 |
- Most predictable: Gregg WILLIAMS has the smallest hole spread. Jon BAKER has the smallest round spread, and Gregg is close behind him.
- Alex's pattern: he is wild hole to hole but fairly steady round to round. His bad holes and good holes seem to cancel out over 18.
- Mullin's round spread (6.12) is the biggest among the players with a full 20 rounds.
How I worked this out: I used TEGs 14–18 and gross strokes vs par. "Erratic" is the standard deviation of a player's own scores, so a higher figure means harder to predict. Hole spread uses every hole played. Round spread uses completed 18-hole rounds. I did not adjust for course difficulty, so rounds on tougher courses add some spread. Stuart (8 rounds) and John (12) have small samples, so their round figures are less reliable.
Show the workings (3 steps)
Code run in the sandbox
mkdir -p /tmp/teg && python3 -m zipfile -e "$INPUT_DIR/teg_toolkit.zip" /tmp/teg && python3 -m zipfile -e "$INPUT_DIR/teg_data.zip" /tmp/teg; head -c 1500 /tmp/teg/skills/teg-analysis/SKILL.md; ls $INPUT_DIR
→ --- name: teg-analysis description: Rigorous ad-hoc analysis of TEG scores with tested helpers (site definitions, tie-aware ranks, own-baseline comparisons, bootstrap CIs, handicap re-scoring). Use for any question that needs your own pandas work on hole, round or TEG data. --- # TEG analysis Use this for questions the site's lookups do not answer directly. The helpers encode the site's definitions so you do not reinvent them. Read `reference/rules.md` first if you are unsure about competitions, eras, handicaps or score names. Setup (once per container) is in the system prompt. Then: ```python import sys; sys.path.insert(0, "/tmp/teg/skills/teg-analysis/scripts") import tegstats as ts f = ts.load_frames() # f.holes, f.rounds, f.tegs, f.winners, f.complete ts.describe_data(f) # frames, columns, TEGs, players ``` Do not do arithmetic in your head. Compute everything in code and quote the output. ## Helpers - `load_precomputed(name)`: one of the site's calculated tables (streaks, bestball, commentary_*). Call with no name to list them. - `net_measure(teg)` / `net_score(df)`: net competition value by era (NetVP up to TEG 7, Stableford from TEG 8). - `rank_min(df, value, by, ascending)` and `top_n(df, value, n, ascending, by)`: tie-aware ranks; `top_n` keeps every row tied inside the top n. - `vs_own_baseline(df, metric, situation_mask)`: each player's metric in a situation vs their own other rows, with n and bootstrap CI, plus an `ALL` row. Use for "doescourse_holes.csv course_notes.csv courses.csv handicaps.csv holes.csv rounds.csv schedule.csv teg_data.zip teg_toolkit.zip tegs.csv winners.csv
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv');r=pd.read_csv('rounds.csv')
h=h[h.TEGNum.between(14,18)];r=r[r.TEGNum.between(14,18)&(r.Holes==18)]
a=h.groupby('Player').agg(Holes=('Sc','size'),Hole_SD=('GrossVP','std'),Avg_hole=('GrossVP','mean'))
# within-round hole SD
a['Stab_hole_SD']=h.groupby('Player').Stableford.std()
b=r.groupby('Player').agg(Rounds=('Sc','size'),Round_SD=('GrossVP','std'),Round_range=('GrossVP',lambda x:x.max()-x.min()),Avg_round=('GrossVP','mean'),Stab_SD=('Stableford','std'))
print(a.join(b).round(2).sort_values('Round_SD',ascending=False).to_markdown())
"
→ Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 135, in import_optional_dependency
module = importlib.import_module(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1140, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'tabulate'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<string>", line 9, in <module>
File "/usr/local/lib/python3.11/site-packages/pandas/util/_decorators.py", line 333, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/core/frame.py", line 2988, in to_markdown
tabulate = import_optional_dependency("tabulate")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 138, in import_optional_dependency
raise ImportError(msg)
ImportError: Missing optional dependency 'tabulate'. Use pip or conda to install tabulate.
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv');r=pd.read_csv('rounds.csv')
h=h[h.TEGNum.between(14,18)];r=r[r.TEGNum.between(14,18)&(r.Holes==18)]
a=h.groupby('Player').agg(Holes=('Sc','size'),Hole_SD=('GrossVP','std'))
b=r.groupby('Player').agg(Rounds=('Sc','size'),Round_SD=('GrossVP','std'),Avg_round=('GrossVP','mean'))
t=a.join(b).round(2).sort_values('Round_SD',ascending=False)
print('| Player | Holes | Hole SD | Rounds | Round SD | Avg round vs par |\n|--|--|--|--|--|--|')
for p,x in t.iterrows(): print('|',p,'|',x.Holes,'|',x.Hole_SD,'|',x.Rounds,'|',x.Round_SD,'|',x.Avg_round,'|')
"
→ | Player | Holes | Hole SD | Rounds | Round SD | Avg round vs par | |--|--|--|--|--|--| | Stuart NEUMANN | 144.0 | 1.42 | 8.0 | 7.53 | 29.12 | | John PATTERSON | 216.0 | 1.38 | 12.0 | 6.98 | 27.67 | | David MULLIN | 360.0 | 1.16 | 20.0 | 6.12 | 20.15 | | Alex BAKER | 360.0 | 1.45 | 20.0 | 5.58 | 32.9 | | Gregg WILLIAMS | 360.0 | 1.07 | 20.0 | 5.32 | 18.0 | | Jon BAKER | 360.0 | 1.21 | 20.0 | 5.21 | 20.95 |
how do the teg19 courses compare to other courses we've played in the last 5 years
Monte Rei, which hosts three of the four rounds, would be the hardest course we've played in the last five years bar one. Quinta da Ria is mid-pack. Neither course has been played by TEG before, so this comes from published ratings, not our scores.
- Monte Rei is a parkland course. It is worth about 5.3 extra shots to an 18-handicapper, which ranks it 3rd hardest of 28 TEG courses. Among recent courses only PGA Catalunya - Stadium (about 6.8 extra shots) is tougher. Monte Rei is well above the next group, PGA Catalunya - Tour, Royal Óbidos and Littlestone, at roughly 2 to 3 extra shots.
- Quinta da Ria is also parkland. It is worth about 1.6 extra shots, ranking 12th of 28. That is about the same as Troia and tougher than Estoril, Penha Longa and Oitavos Dunes.
How we've actually scored on recent courses (since 2021, average gross vs par per round, all players) is a useful yardstick. The hardest were Oitavos Dunes at +27.3, Praia D'El Rey at +26.9 and West Cliffs at +26.5. The easiest was Prince's - Shore / Dunes at +18.6. PGA Catalunya - Stadium averaged +21.5, so our scoring there was no worse than at the supposedly easier courses. Our results don't follow the published difficulty closely, and the ratings are only good to about a point.
Monte Rei is the big unknown. Handicaps already allow for some of that, but three rounds there are likely to hurt the higher-handicap players most.
How I worked this out: "Last 5 years" means rounds from 2021 onwards (TEG 14 to TEG 18). Extra shots and rank come from the course ratings on file. Some courses have only 4 to 5 rounds, so their scoring averages are small samples.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
c=pd.read_csv('courses.csv');s=pd.read_csv('schedule.csv')
print(s[s.TEGNum==19])
print(s.Year.max())
r=pd.read_csv('rounds.csv')
r=r[r.Year>=2021]
print(r.groupby('Course').agg(n=('Sc','size'),gvp=('GrossVP','mean')).round(1))
print(c[['Course','type','par','course_rating','slope_rating','extra_strokes','difficulty_band','difficulty_rank','TEGsPlayed']].to_string())
"
→ TEGNum Year Area Round Course Date Status
67 19 2026 Algarve, Portugal 1 Quinta da Ria 2026-10-10 upcoming
68 19 2026 Algarve, Portugal 2 Monte Rei 2026-10-11 upcoming
69 19 2026 Algarve, Portugal 3 Monte Rei 2026-10-12 upcoming
70 19 2026 Algarve, Portugal 4 Monte Rei 2026-10-13 upcoming
2026
n gvp
Course
Estoril 11 24.5
Littlestone 4 22.5
Oitavos Dunes 11 27.3
PGA Catalunya - Stadium 10 21.5
PGA Catalunya - Tour 10 24.2
Penha Longa 11 22.1
Praia D'El Rey 10 26.9
Prince's - Shore / Dunes 8 18.6
Royal Cinque Ports 4 20.5
Royal Óbidos 5 24.6
Troia 5 26.2
West Cliffs 11 26.5
Course type par course_rating slope_rating extra_strokes difficulty_band difficulty_rank TEGsPlayed
0 Ashdown Woodland 68.0 67.2 117.0 -0.2 standard 21 TEG 4
1 Bletchingley Parkland 72.0 71.3 129.0 1.8 tough 10 TEG 4
2 Boavista Parkland 71.0 68.4 118.0 -1.8 kind 27 TEG 2, TEG 3, TEG 5, TEG 10
3 Bom Sucesso Parkland 72.0 71.4 126.0 1.5 standard 13 TEG 7
4 Costa Brava Parkland/Woodland 73.0 70.9 132.0 0.9 standard 17 TEG 6
5 Crowborough Heathland 71.0 68.5 115.0 -2.2 kind 28 TEG 4
6 El Prat - Azul Parkland 72.0 71.1 127.0 1.3 standard 16 TEG 11
7 El Prat - Rosa Parkland 72.0 72.6 143.0 5.4 brutal 2 TEG 6
8 Estoril Parkland 69.0 66.8 120.0 -1.1 kind 24 TEG 15, TEG 16
9 Lingfield Parkland 72.0 70.4 134.0 1.7 tough 11 TEG 4
10 Littlestone Links 71.0 70.3 133.0 2.5 tough 6 TEG 13, TEG 14
11 Monte Rei Parkland 72.0 72.7 142.0 5.3 brutal 3 NaN
12 Oitavos Dunes Links 71.0 70.3 120.0 0.4 standard 19 TEG 8, TEG 15, TEG 16
13 PGA Catalunya - Stadium Championship Parkland 72.0 73.2 148.0 6.8 brutal 1 TEG 6, TEG 11, TEG 12, TEG 18
14 PGA Catalunya - Tour Parkland 72.0 70.4 136.0 2.1 tough 8 TEG 6, TEG 11, TEG 12, TEG 18
15 Palmares - Alvor / Lagos Parkland 72.0 71.9 130.0 2.6 tough 5 TEG 10
16 Palmares - Lagos / Praia Parkland and Links 72.0 69.7 125.0 -0.4 standard 23 TEG 5
17 Palmares - Praia / Alvor Links and Parkland 72.0 70.4 128.0 0.8 standard 18 TEG 10
18 Penha Longa Mountain Parkland 72.0 69.8 125.0 -0.3 standard 22 TEG 8, TEG 15, TEG 16
19 Praia D'El Rey Links and Parkland 73.0 70.3 123.0 -1.1 kind 24 TEG 7, TEG 9, TEG 17
20 Prince's - Dunes / Himalayas Links 72.0 70.3 123.0 -0.1 standard 20 TEG 13
21 Prince's - Shore / Dunes Links 72.0 70.2 116.0 -1.3 kind 26 TEG 13, TEG 14
22 Quinta da Marinha Parkland 71.0 69.6 131.0 1.5 standard 13 TEG 8
23 Quinta da Ria Parkland 72.0 70.9 130.0 1.6 tough 12 NaN
24 Royal Cinque Ports Links 72.0 72.0 125.0 1.9 tough 9 TEG 13, TEG 14
25 Royal Óbidos Parkland 72.0 71.8 131.0 2.7 tough 4 TEG 7, TEG 9, TEG 17
26 Troia Championship Course 72.0 72.4 119.0 1.4 standard 15 TEG 16
27 West Cliffs Links 72.0 70.3 138.0 2.3 tough 7 TEG 15, TEG 17
Where is the wooden spoon
The Wooden Spoon is currently with Jon BAKER, who finished last in the net competition at TEG 18 (2025). I've read "where is it" as "who holds it now". If you meant something else, tell me.
David MULLIN has won it most often, with 4 Spoons. The full count:
| Player | Wooden Spoons |
|---|---|
| David MULLIN | 4 |
| Henry MELLER | 3 |
| Alex BAKER | 3 |
| Gregg WILLIAMS | 2 |
| Stuart NEUMANN | 2 |
| Jon BAKER | 2 |
| John PATTERSON | 1 |
TEG 19 starts on 10 October 2026, so the Spoon changes hands soon. The Honours board has the full list.
How I worked this out: I used the official honours record for completed TEGs, TEG 2 to TEG 18. The Wooden Spoon goes to last place in the net competition.
Show the workings (1 step)
Lookup: get_honours
{
"competition": "spoon"
}
→ {
"winners_by_teg": [
{
"TEG": "TEG 2",
"Year": 2009,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 3",
"Year": 2010,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 4",
"Year": 2011,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 5",
"Year": 2012,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Stuart NEUMANN*",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 6",
"Year": 2013,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 7",
"Year": 2014,
"TEG Trophy": "Henry MELLER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 8",
"Year": 2015,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 9",
"Year": 2016,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 10",
"Year": 2017,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "John PATTERSON"
},
{
"TEG": "TEG 11",
"Year": 2018,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 12",
"Year": 2019,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 13",
"Year": 2020,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 14",
"Year": 2021,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Jon BAKER"
},
{
"TEG": "TEG 15",
"Year": 2022,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 16",
"Year": 2023,
"TEG Trophy": "Stuart NEUMANN",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 17",
"Year": 2024,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 18",
"Year": 2025,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Jon BAKER"
}
],
"counts": {
"HMM Wooden Spoon": {
"David MULLIN": 4,
"Henry MELLER": 3,
"Alex BAKER": 3,
"Gregg WILLIAMS": 2,
"Stuart NEUMANN": 2,
"Jon BAKER": 2,
"John PATTERSON": 1
}
},
"page": "/honours?tab=spoon",
"notes": [
"Only completed TEGs are counted.",
"TEG Trophy = net competition (net vs par up to TEG 7, Stableford points from TEG 8).",
"Green Jacket = best gross. Wooden Spoon = last in the net competition.",
"TEG 5 Green Jacket was awarded for best Stableford round; David MULLIN had the best gross score."
]
}
Who has the best eclectic score?
David MULLIN and Jon BAKER tie for the best eclectic score, both at 19 under par. Stuart NEUMANN is next at 13 under, then Gregg WILLIAMS at 12 under.
| Player | Eclectic vs par |
|---|---|
| David MULLIN | -19 |
| Jon BAKER | -19 |
| Stuart NEUMANN | -13 |
| Gregg WILLIAMS | -12 |
| John PATTERSON | -10 |
| Alex BAKER | -10 |
| Henry MELLER | -7 |
See the Eclectic page for the full version.
How I worked this out: I took each player's best gross score vs par on each hole number (1–18) across every TEG and course they've played, then summed the 18 holes. I did not check the result against the Eclectic page. If that page sets the eclectic up differently, for example per course, the figures could differ.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
e=h.groupby(['Player','Course','Hole']).Sc.min().reset_index()
# eclectic across all courses: best per hole number
e=h.groupby(['Player','Hole']).agg(Sc=('Sc','min'),P=('PAR','first')).reset_index()
g=e.groupby('Player').agg(Sc=('Sc','sum'),P=('P','sum'),n=('Hole','count'))
g['vp']=g.Sc-g.P
print(g.sort_values('vp').to_markdown())
"
→ Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 135, in import_optional_dependency
module = importlib.import_module(name)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/importlib/__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1204, in _gcd_import
File "<frozen importlib._bootstrap>", line 1176, in _find_and_load
File "<frozen importlib._bootstrap>", line 1140, in _find_and_load_unlocked
ModuleNotFoundError: No module named 'tabulate'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<string>", line 9, in <module>
File "/usr/local/lib/python3.11/site-packages/pandas/util/_decorators.py", line 333, in wrapper
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/core/frame.py", line 2988, in to_markdown
tabulate = import_optional_dependency("tabulate")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/pandas/compat/_optional.py", line 138, in import_optional_dependency
raise ImportError(msg)
ImportError: Missing optional dependency 'tabulate'. Use pip or conda to install tabulate.
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
e=h.groupby(['Player','Hole']).GrossVP.min().reset_index()
g=e.groupby('Player').agg(vp=('GrossVP','sum'),n=('Hole','count')).sort_values('vp').reset_index()
g.columns=['Player','Eclectic vs par','Holes']
print(g.to_string(index=False))
"
→ Player Eclectic vs par Holes
David MULLIN -19.0 18
Jon BAKER -19.0 18
Stuart NEUMANN -13.0 18
Gregg WILLIAMS -12.0 18
John PATTERSON -10.0 18
Alex BAKER -10.0 18
Henry MELLER -7.0 18
Who's won the most TEG Trophies?
Jon BAKER and Gregg WILLIAMS are tied for the most TEG Trophies, with 4 each.
| Player | Trophies |
|---|---|
| Jon BAKER | 4 |
| Gregg WILLIAMS | 4 |
| David MULLIN | 3 |
| John PATTERSON | 2 |
| Alex BAKER | 2 |
| Henry MELLER | 1 |
| Stuart NEUMANN | 1 |
See the Honours board for the full list by TEG.
How I worked this out: I used the official honours record, which counts completed TEGs only (TEG 2 to TEG 18). The Trophy is the net competition.
Show the workings (1 step)
Lookup: get_honours
{
"competition": "trophy"
}
→ {
"winners_by_teg": [
{
"TEG": "TEG 2",
"Year": 2009,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 3",
"Year": 2010,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 4",
"Year": 2011,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 5",
"Year": 2012,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Stuart NEUMANN*",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 6",
"Year": 2013,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 7",
"Year": 2014,
"TEG Trophy": "Henry MELLER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 8",
"Year": 2015,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 9",
"Year": 2016,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 10",
"Year": 2017,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "John PATTERSON"
},
{
"TEG": "TEG 11",
"Year": 2018,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 12",
"Year": 2019,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 13",
"Year": 2020,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 14",
"Year": 2021,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Jon BAKER"
},
{
"TEG": "TEG 15",
"Year": 2022,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 16",
"Year": 2023,
"TEG Trophy": "Stuart NEUMANN",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 17",
"Year": 2024,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 18",
"Year": 2025,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Jon BAKER"
}
],
"counts": {
"TEG Trophy": {
"Jon BAKER": 4,
"Gregg WILLIAMS": 4,
"David MULLIN": 3,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1,
"Stuart NEUMANN": 1
}
},
"page": "/honours?tab=trophy",
"notes": [
"Only completed TEGs are counted.",
"TEG Trophy = net competition (net vs par up to TEG 7, Stableford points from TEG 8).",
"Green Jacket = best gross. Wooden Spoon = last in the net competition.",
"TEG 5 Green Jacket was awarded for best Stableford round; David MULLIN had the best gross score."
]
}
which players have held the lead during the final round of the net competition and failed to go on to win? list the players and the tegs, and the max lead they held in round 4.
Five player-TEG cases fit: Jon BAKER (TEG 4 and TEG 6), Alex BAKER (TEG 9 and TEG 11) and Gregg WILLIAMS (TEG 10). The biggest collapse was Jon BAKER in TEG 4, who led by 7 strokes at one point in round 4 and lost to David MULLIN.
| TEG | Player | Holes in round 4 at the top | Max sole lead in round 4 | Winner |
|---|---|---|---|---|
| 4 | Jon BAKER | 10 (6 outright) | 7 strokes | David MULLIN |
| 6 | Jon BAKER | 2 (both outright) | 2 strokes | Gregg WILLIAMS |
| 9 | Alex BAKER | 1 (tied) | 0 (joint lead only) | John PATTERSON |
| 10 | Gregg WILLIAMS | 1 (tied) | 0 (joint lead only) | Alex BAKER |
| 11 | Alex BAKER | 6 (5 outright) | 4 points | Jon BAKER |
Lead units follow the net competition of the day: strokes (net vs par) up to TEG 7, Stableford points from TEG 8.
How I worked this out: I built the running net total hole by hole through round 4 of each completed TEG, including the earlier rounds' scores. A player "held the lead" if they were top, or joint top, of the net standings after any hole in round 4. Max lead is their biggest margin over second place at any such hole, and it shows 0 where they were only tied. I checked them against the official Trophy winner. TEG 2 was skipped because it only had three rounds. Leads held only before round 4 started aren't counted. Sample: 16 completed TEGs with a round 4.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; head -3 winners.csv; cat > /tmp/a.py <<'E'
import pandas as pd
h=pd.read_csv('holes.csv');w=pd.read_csv('winners.csv');t=pd.read_csv('tegs.csv')
done=t[t.Complete]['TEGNum'].unique()
out=[]
for n in sorted(done):
d=h[h.TEGNum==n].sort_values(['Round','Hole'])
if d.Round.max()<4: continue
col='Stableford' if n>=8 else 'NetVP'
sign=1 if n>=8 else -1
d=d.assign(v=d[col]*sign)
d['cum']=d.groupby('Player').v.cumsum()
r4=d[d.Round==4]
win=w[w.TEG==f'TEG {n}']['TEG Trophy'].iloc[0]
for hole,g in r4.groupby('Hole'):
s=g.set_index('Player').cum.sort_values(ascending=False)
if len(s)<2: continue
top=s.iloc[0]
leaders=list(s[s==top].index)
lead=top-s[s<top].max() if (s<top).any() else 0
for p in leaders:
out.append((n,p,hole,len(leaders),lead if len(leaders)==1 else 0,win))
o=pd.DataFrame(out,columns=['teg','p','hole','nl','lead','win'])
o=o[~o.apply(lambda r:r.p in r.win,axis=1)]
print(o.groupby(['teg','p']).agg(holes=('hole','count'),sole=('nl',lambda x:(x==1).sum()),maxlead=('lead','max')).join(o.groupby(['teg','p']).win.first()))
E
python3 /tmp/a.py
→ TEG,Year,TEG Trophy,Green Jacket,HMM Wooden Spoon
TEG 2,2009,David MULLIN,David MULLIN,Henry MELLER
TEG 3,2010,Jon BAKER,David MULLIN,Henry MELLER
holes sole maxlead win
teg p
4 Jon BAKER 10 6 7.0 David MULLIN
6 Jon BAKER 2 2 2.0 Gregg WILLIAMS
9 Alex BAKER 1 0 0.0 John PATTERSON
10 Gregg WILLIAMS 1 0 0.0 Alex BAKER
11 Alex BAKER 6 5 4.0 Jon BAKER
Taking the most recent 5 tournaments only, which player has the biggest problem on a specific type of hole relative to their normal play - e.g. who struggles most on a specific par, or on a SI range (e.g.the hardest 3 holes vs the rest), or front 9 vs back 9, or later rounds vs earlier rounds etc.
No single weak spot stands out. Three players are almost level on the biggest gap, all at about +0.54 strokes per hole worse than their own normal. Jon BAKER and David MULLIN both struggle on the three hardest holes, and Stuart NEUMANN does worse in rounds 1-2. This covers TEGs 14-18.
| Player | Weak spot | Avg vs par there | Avg vs par elsewhere | Gap | Holes |
|---|---|---|---|---|---|
| Jon BAKER | Hardest 3 holes (SI 1-3) | +1.62 | +1.07 | +0.54 | 60 |
| Stuart NEUMANN | Rounds 1-2 versus rounds 3-4 | +1.89 | +1.35 | +0.54 | 72 |
| David MULLIN | Hardest 3 holes (SI 1-3) | +1.57 | +1.03 | +0.54 | 60 |
| Alex BAKER | Holes 16-18 | +2.22 | +1.75 | +0.47 | 60 |
| Gregg WILLIAMS | Hardest 3 holes (SI 1-3) | +1.37 | +0.93 | +0.44 | 60 |
| John PATTERSON | Par 4s | +1.74 | +1.32 | +0.42 | 112 |
- Hardest holes: the SI 1-3 holes cost four of the six players more than their normal, so this is a group-wide weakness.
- Alex BAKER: he is the clearest case of a late fade, with +2.22 per hole on holes 16-18. He is also worse than his normal on par 4s (+0.40) and on SI 4-9 holes.
- Stuart NEUMANN: his rounds 1-2 gap is the odd one out. He plays worse early and improves later, and it rests on only 72 holes, so treat it with caution.
How I worked this out: I used the five most recent TEGs (14-18) and gross strokes vs par per hole. For each player I compared a category against all their other holes in the same period. The categories were par 3/4/5, hardest 3 holes (SI 1-3), SI 4-9 versus SI 10-18, front versus back 9, rounds 1-2 versus 3-4, and holes 16-18 versus 1-15. The three leaders are separated by tiny amounts, so the order among them isn't meaningful. Samples are small: Stuart NEUMANN has 144 holes in total and John PATTERSON 216.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
t=sorted(h.TEGNum.unique())[-5:];print(t)
h=h[h.TEGNum.isin(t)].copy()
h['SIg']=pd.cut(h.SI,[0,3,9,18],labels=['SI1-3','SI4-9','SI10-18'])
h['Par']='Par '+h.PAR.astype(str)
h['Half']=h.FrontBack
h['Late']=h.Round.map(lambda r:'R1-2' if r<=2 else 'R3-4')
h['Last3']=h.Hole.map(lambda x:'H16-18' if x>=16 else 'H1-15')
res=[]
for col in ['Par','SIg','Half','Late','Last3']:
g=h.groupby(['Player',col],observed=True).GrossVP.agg(['mean','count']).reset_index()
for p,d in g.groupby('Player'):
for _,r in d.iterrows():
rest=h[(h.Player==p)&(h[col]!=r[col])].GrossVP.mean()
res.append((p,col,r[col],r['count'],round(r['mean'],2),round(rest,2),round(r['mean']-rest,2)))
df=pd.DataFrame(res,columns=['Player','Split','Type','n','Avg','RestAvg','Diff']).sort_values('Diff',ascending=False)
print(df.head(15).to_string())
print(h.groupby('Player').size())
"
→ [np.int64(14), np.int64(15), np.int64(16), np.int64(17), np.int64(18)]
Player Split Type n Avg RestAvg Diff
30 Jon BAKER SIg SI1-3 60 1.62 1.07 0.54
58 Stuart NEUMANN Late R1-2 72 1.89 1.35 0.54
21 David MULLIN SIg SI1-3 60 1.57 1.03 0.54
61 Alex BAKER Last3 H16-18 60 2.22 1.75 0.47
24 Gregg WILLIAMS SIg SI1-3 60 1.37 0.93 0.44
69 Jon BAKER Last3 H16-18 60 1.52 1.09 0.42
10 John PATTERSON Par Par 4 112 1.74 1.32 0.42
33 Stuart NEUMANN SIg SI1-3 24 1.96 1.55 0.41
1 Alex BAKER Par Par 4 195 2.01 1.61 0.40
28 John PATTERSON SIg SI4-9 72 1.78 1.42 0.36
18 Alex BAKER SIg SI1-3 60 2.12 1.77 0.35
19 Alex BAKER SIg SI4-9 120 2.04 1.72 0.32
2 Alex BAKER Par Par 5 79 2.06 1.76 0.30
7 Gregg WILLIAMS Par Par 4 195 1.14 0.84 0.30
27 John PATTERSON SIg SI1-3 36 1.78 1.49 0.29
Player
Alex BAKER 360
David MULLIN 360
Gregg WILLIAMS 360
John PATTERSON 216
Jon BAKER 360
Stuart NEUMANN 144
dtype: int64
Outside of SI differentials, what does this table look like. Top 10 rows. Only TEG 19 competitors.
Alex BAKER's late fade is still the biggest problem. He is +0.47 strokes per hole worse on holes 16-18 than on holes 1-15. Par 4s then dominate the table, because five of the next nine rows are par 4s. This uses TEGs 14-18 and the five players in the TEG 19 field.
| # | Player | Weak spot | Avg vs par there | Avg vs par elsewhere | Gap | Holes |
|---|---|---|---|---|---|---|
| 1 | Alex BAKER | Holes 16-18 | +2.22 | +1.75 | +0.47 | 60 |
| 2 | John PATTERSON | Par 4s | +1.74 | +1.32 | +0.42 | 112 |
| 3 | Jon BAKER | Holes 16-18 | +1.52 | +1.09 | +0.42 | 60 |
| 4 | Alex BAKER | Par 4s | +2.01 | +1.61 | +0.40 | 195 |
| 5 | Alex BAKER | Par 5s | +2.06 | +1.76 | +0.30 | 79 |
| 6 | Gregg WILLIAMS | Par 4s | +1.14 | +0.84 | +0.30 | 195 |
| 7 | David MULLIN | Par 4s | +1.24 | +0.98 | +0.27 | 195 |
| 8 | Jon BAKER | Par 4s | +1.27 | +1.04 | +0.22 | 195 |
| 9 | Gregg WILLIAMS | Rounds 1-2 versus 3-4 | +1.10 | +0.90 | +0.20 | 180 |
| 10 | Gregg WILLIAMS | Front 9 versus back 9 | +1.08 | +0.92 | +0.17 | 180 |
- Par 4s: every player in the TEG 19 field does worse than their own normal on par 4s.
- Holes 16-18: Alex BAKER and Jon BAKER both fade at the end. John PATTERSON is the opposite and does better on holes 16-18 than on the rest of his round.
- Stuart NEUMANN: he is left out because he isn't in the TEG 19 field, according to the TEG Predictatron 3100.
How I worked this out: I used the same method as before, with gross strokes vs par per hole over TEGs 14-18. I dropped the stroke index splits and kept par 3/4/5, front vs back 9, rounds 1-2 vs 3-4, and holes 16-18 vs 1-15. Each gap compares a category with that player's other holes. I treated the TEG 19 field as the five players the Predictatron lists (Jon BAKER, John PATTERSON, Gregg WILLIAMS, Alex BAKER and David MULLIN). I left out one mirror-image row, John PATTERSON's holes 1-15, because it only reflects him doing better late. Samples are small for the 60-hole and 79-hole rows, and gaps this close in size shouldn't be read as a firm ranking.
Show the workings (3 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv');print(sorted(h.TEGNum.unique())[-3:])
t=pd.read_csv('tegs.csv');print(t[t.TEGNum>=18][['Player','TEGNum','Complete']])"
→ [np.int64(16), np.int64(17), np.int64(18)]
Player TEGNum Complete
85 Alex BAKER 18 True
86 John PATTERSON 18 True
87 Gregg WILLIAMS 18 True
88 David MULLIN 18 True
89 Jon BAKER 18 True
Lookup: get_predictions
{}
→ {
"teg_num": 19,
"simulations": 10000,
"players": [
{
"Player": "Jon BAKER",
"Handicap": "21",
"TrophyChancePct": 41.8,
"TrophyOdds": "11/8",
"JacketChancePct": 17.8,
"JacketOdds": "9/2",
"SpoonChancePct": 5.4,
"SpoonOdds": "16/1",
"ExpectedStableford": 147.5,
"ExpectedGrossVsPar": 83.0
},
{
"Player": "John PATTERSON",
"Handicap": "25",
"TrophyChancePct": 19.8,
"TrophyOdds": "4/1",
"JacketChancePct": 0.2,
"JacketOdds": "500/1",
"SpoonChancePct": 19.3,
"SpoonOdds": "4/1",
"ExpectedStableford": 141.9,
"ExpectedGrossVsPar": 107.4
},
{
"Player": "Gregg WILLIAMS",
"Handicap": "17",
"TrophyChancePct": 14.9,
"TrophyOdds": "11/2",
"JacketChancePct": 68.6,
"JacketOdds": "4/9",
"SpoonChancePct": 16.9,
"SpoonOdds": "5/1",
"ExpectedStableford": 141.6,
"ExpectedGrossVsPar": 73.0
},
{
"Player": "Alex BAKER",
"Handicap": "31",
"TrophyChancePct": 13.2,
"TrophyOdds": "13/2",
"JacketChancePct": 0.0,
"JacketOdds": "1000/1+",
"SpoonChancePct": 30.4,
"SpoonOdds": "9/4",
"ExpectedStableford": 139.1,
"ExpectedGrossVsPar": 137.5
},
{
"Player": "David MULLIN",
"Handicap": "19",
"TrophyChancePct": 10.3,
"TrophyOdds": "9/1",
"JacketChancePct": 13.4,
"JacketOdds": "13/2",
"SpoonChancePct": 28.0,
"SpoonOdds": "5/2",
"ExpectedStableford": 139.0,
"ExpectedGrossVsPar": 85.1
}
],
"notes": [
"A prediction from past form, not a certainty.",
"The Predictatron page reruns the simulation, so its figures can differ by a point or two, and lets you change the settings."
],
"definition": "Monte Carlo simulation of TEG 19 (10,000 runs). Each player's hole scores are drawn from their recent history by par and stroke index, using this TEG's courses and handicaps. Trophy = most Stableford points, Green Jacket = lowest gross, Wooden Spoon = fewest Stableford points. Chances are the share of simulations each player won; odds are the fair fractional equivalent.",
"page": "/simulation"
}
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
h=h[h.TEGNum.isin([14,15,16,17,18])&h.Player.isin(['Jon BAKER','John PATTERSON','Gregg WILLIAMS','Alex BAKER','David MULLIN'])].copy()
h['Par']='Par '+h.PAR.astype(str)
h['Half']=h.FrontBack
h['Late']=h.Round.map(lambda r:'R1-2' if r<=2 else 'R3-4')
h['Last3']=h.Hole.map(lambda x:'H16-18' if x>=16 else 'H1-15')
res=[]
for col in ['Par','Half','Late','Last3']:
g=h.groupby(['Player',col]).GrossVP.agg(['mean','count']).reset_index()
for _,r in g.iterrows():
rest=h[(h.Player==r.Player)&(h[col]!=r[col])].GrossVP.mean()
res.append((r.Player,col,r[col],r['count'],round(r['mean'],2),round(rest,2),round(r['mean']-rest,3)))
df=pd.DataFrame(res,columns=['Player','Split','Type','n','Avg','Rest','Diff']).sort_values('Diff',ascending=False)
print(df.head(12).to_string())"
→ Player Split Type n Avg Rest Diff 36 Alex BAKER Last3 H16-18 60 2.22 1.75 0.467 10 John PATTERSON Par Par 4 112 1.74 1.32 0.424 44 Jon BAKER Last3 H16-18 60 1.52 1.09 0.423 1 Alex BAKER Par Par 4 195 2.01 1.61 0.398 2 Alex BAKER Par Par 5 79 2.06 1.76 0.302 7 Gregg WILLIAMS Par Par 4 195 1.14 0.84 0.302 4 David MULLIN Par Par 4 195 1.24 0.98 0.265 41 John PATTERSON Last3 H1-15 180 1.58 1.33 0.244 13 Jon BAKER Par Par 4 195 1.27 1.04 0.224 29 Gregg WILLIAMS Late R1-2 180 1.10 0.90 0.200 20 Gregg WILLIAMS Half Front 180 1.08 0.92 0.167 22 John PATTERSON Half Front 108 1.62 1.45 0.167
what's the easiest course that has hosted at least 4 rounds. And the hardest.
The easiest course is Bom Sucesso, and the hardest is Quinta da Marinha. I measured difficulty as average gross score vs par per round, using courses with at least 4 rounds played.
| Course | Rounds | Avg vs par | Avg Stableford points | |
|---|---|---|---|---|
| Easiest | Bom Sucesso | 6 | +17.3 | 44.8 |
| Hardest | Quinta da Marinha | 12 | +30.5 | 29.5 |
Both were played in a single TEG, so each reflects one year's conditions and field. Bom Sucesso also had just 6 rounds. Of the courses played across several TEGs, the easiest is Prince's - Shore / Dunes (+18.5) and the hardest is Oitavos Dunes (+27.1).
See all course averages.
How I worked this out: I took every completed 18-hole round, grouped by course, and kept courses with at least 4 rounds. I ranked them by average strokes vs par, gross, per round. Handicaps vary between players and years, so I used gross rather than net. Net and Stableford averages point the same way at the extremes.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv')
r=r[r.Holes==18]
g=r.groupby('Course').agg(n=('Sc','size'),rounds_played=('Round','size'),gross=('GrossVP','mean'),sc=('Sc','mean'),stb=('Stableford','mean'),net=('NetVP','mean'),tegs=('TEGNum','nunique'))
g=g[g.n>=4].sort_values('gross')
print(g.round(2))
"
→ n rounds_played gross ... stb net tegs Course ... Bom Sucesso 6 6 17.33 ... 44.83 -8.67 1 Prince's - Shore / Dunes 13 13 18.54 ... 37.62 -0.77 2 Palmares - Praia / Alvor 6 6 19.33 ... 42.00 -4.67 1 Littlestone 9 9 19.44 ... 36.56 -0.11 2 Prince's - Dunes / Himalayas 5 5 21.40 ... 35.20 1.20 1 PGA Catalunya - Tour 32 32 21.53 ... 38.44 -1.38 4 Costa Brava 5 5 21.60 ... 37.40 -1.40 1 Palmares - Lagos / Praia 5 5 22.60 ... 36.20 0.60 1 Royal Cinque Ports 9 9 22.78 ... 33.44 3.22 2 Praia D'El Rey 34 34 22.79 ... 39.09 -2.21 3 El Prat - Azul 5 5 22.80 ... 34.20 2.40 1 Penha Longa 17 17 23.00 ... 35.82 1.24 3 Palmares - Alvor / Lagos 6 6 24.17 ... 37.17 0.17 1 PGA Catalunya - Stadium 37 37 24.49 ... 35.03 1.92 4 Estoril 11 11 24.55 ... 35.45 2.73 2 Boavista 62 62 24.84 ... 35.02 2.10 4 Crowborough 5 5 26.00 ... 31.60 4.80 1 Troia 5 5 26.20 ... 32.60 3.40 1 West Cliffs 11 11 26.45 ... 34.18 4.18 2 Royal Óbidos 23 23 26.78 ... 35.57 1.78 3 Oitavos Dunes 17 17 27.12 ... 32.00 5.35 3 Lingfield 5 5 27.40 ... 31.20 6.20 1 Bletchingley 5 5 28.20 ... 31.40 7.00 1 Ashdown 5 5 29.00 ... 30.20 7.80 1 El Prat - Rosa 5 5 29.20 ... 31.40 6.20 1 Quinta da Marinha 12 12 30.50 ... 29.50 8.83 1 [26 rows x 7 columns]
Who plays best in first rounds?
Jon BAKER plays best in first rounds, both in raw terms and against his own standard. He averages 36.8 Stableford points in round 1, which is +0.9 better than in his other rounds. He is the only player who scores better in round 1 than later. David MULLIN is level, and everyone else is worse in round 1.
| Player | Round 1 avg points | Other rounds avg points | Difference | Round 1 avg vs par | Round 1 rounds (n) |
|---|---|---|---|---|---|
| Jon BAKER | 36.8 | 35.8 | +0.9 | +19.8 | 17 |
| David MULLIN | 35.3 | 35.6 | -0.3 | +18.7 | 17 |
| Gregg WILLIAMS | 35.4 | 36.7 | -1.4 | +26.1 | 17 |
| Henry MELLER | 31.7 | 33.0 | -1.2 | +27.9 | 7 |
| John PATTERSON | 34.1 | 36.9 | -2.8 | +29.9 | 8 |
| Alex BAKER | 34.2 | 37.3 | -3.1 | +33.5 | 12 |
| Stuart NEUMANN | 30.7 | 35.5 | -4.8 | +30.7 | 12 |
Stuart NEUMANN is the slowest starter, ahead of Alex BAKER. Note that "Avg vs par" is gross.
See all rounds for the full data.
How I worked this out: I used every completed 18-hole round and compared round 1 with rounds 2–4 for each player. I led with Stableford points, where higher is better. Gross vs par is shown alongside. Sample sizes are small, especially for Henry MELLER and John PATTERSON, so treat the gaps loosely. Handicaps change over time, and this doesn't adjust for that.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv')
r=r[r.Holes==18]
r['R1']=r.Round==1
g=r.groupby(['Player','R1']).agg(vp=('GrossVP','mean'),st=('Stableford','mean'),n=('Sc','size')).unstack()
g['dvp']=g[('vp',True)]-g[('vp',False)]
g['dst']=g[('st',True)]-g[('st',False)]
print(g.round(2).sort_values('dst',ascending=False).to_string())
"
→ vp st n dvp dst R1 False True False True False True Player Jon BAKER 20.72 19.82 35.84 36.76 50 17 -0.90 0.92 David MULLIN 18.66 18.71 35.62 35.29 50 17 0.05 -0.33 Henry MELLER 25.05 27.86 32.95 31.71 20 7 2.81 -1.24 Gregg WILLIAMS 24.12 26.06 36.72 35.35 50 17 1.94 -1.37 John PATTERSON 26.58 29.88 36.92 34.12 24 8 3.29 -2.79 Alex BAKER 29.81 33.50 37.31 34.17 36 12 3.69 -3.14 Stuart NEUMANN 24.97 30.67 35.46 30.67 35 12 5.70 -4.79
Who is best on the back 9?
John PATTERSON is the most improved on the back 9. He scores +0.7 points per nine better than on his front 9, just ahead of Gregg WILLIAMS at +0.6. In raw terms, Gregg has the highest back 9 score at 18.5 points per nine. David MULLIN has the best gross back 9, at +9.8 vs par per nine.
Only John and Gregg play better on the back than the front. Everyone else gets worse.
| Player | Back 9 avg points | Front 9 avg points | Back minus front (points) | Back 9 avg vs par | Front 9 avg vs par | Nines (n) |
|---|---|---|---|---|---|---|
| John PATTERSON | 18.5 | 17.8 | +0.7 | +13.4 | +14.0 | 32 |
| Gregg WILLIAMS | 18.5 | 17.9 | +0.6 | +12.1 | +12.5 | 67 |
| Jon BAKER | 17.9 | 18.2 | -0.4 | +10.6 | +9.9 | 67 |
| Alex BAKER | 18.0 | 18.5 | -0.5 | +15.9 | +14.8 | 48 |
| David MULLIN | 17.5 | 18.1 | -0.6 | +9.8 | +8.9 | 67 |
| Henry MELLER | 16.0 | 16.7 | -0.7 | +13.4 | +12.4 | 27 |
| Stuart NEUMANN | 16.7 | 17.6 | -0.9 | +13.9 | +12.5 | 47 |
I led with the relative measure (back 9 against the player's own front 9). John's sample is small, so treat his gap loosely.
How I worked this out: I used every hole ever played, including in-progress TEGs. I summed Stableford points and gross strokes vs par on holes 1-9 and 10-18 for each player. Then I scaled them to an average per nine. Stableford points are net of handicap, so Gregg's raw lead partly reflects his handicap. Gross vs par is the pure scoring measure, and David is best on that.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
g=h.groupby(['Player','FrontBack']).agg(st=('Stableford','sum'),vp=('GrossVP','sum'),n=('Sc','size'))
g['stp']=g.st/g.n*9; g['vpp']=g.vp/g.n*9
u=g[['stp','vpp','n']].unstack()
u['dst']=u[('stp','Back')]-u[('stp','Front')]
u['dvp']=u[('vpp','Back')]-u[('vpp','Front')]
print(u.round(2).sort_values('dst',ascending=False).to_string())
"
→ stp vpp n dst dvp FrontBack Back Front Back Front Back Front Player John PATTERSON 18.47 17.75 13.44 13.97 288 288 0.72 -0.53 Gregg WILLIAMS 18.51 17.87 12.10 12.51 603 603 0.64 -0.40 Jon BAKER 17.85 18.22 10.64 9.85 603 603 -0.37 0.79 Alex BAKER 18.02 18.50 15.90 14.83 432 432 -0.48 1.06 David MULLIN 17.46 18.07 9.79 8.88 603 603 -0.61 0.91 Henry MELLER 15.96 16.67 13.41 12.37 243 243 -0.70 1.04 Stuart NEUMANN 16.66 17.57 13.94 12.49 423 423 -0.91 1.45
What happened the last time we played Boavista?
The last time we played Boavista was TEG 10 (Algarve, 30 September to 3 October 2017). Boavista hosted rounds 1 and 4. Alex BAKER won the TEG Trophy, David MULLIN took the Green Jacket and John PATTERSON took the Wooden Spoon.
- Round 1: David MULLIN scored 49 points (gross 75, +4), the best score of the day.
- Round 4: Alex BAKER scored 51 points, which was enough to win the Trophy. John PATTERSON had 21 points (gross 111).
| Player | Round 1 points | Round 4 points | Final Trophy position | Final Jacket position |
|---|---|---|---|---|
| Alex BAKER | 41 | 51 | 1 | 4 |
| Gregg WILLIAMS | 38 | 40 | 2 | 3 |
| David MULLIN | 49 | 40 | 3 | 1 |
| Jon BAKER | 37 | 40 | 4 | 2 |
| Stuart NEUMANN | 32 | 39 | 5 | 5 |
| John PATTERSON | 29 | 21 | 6 | 6 |
How I worked this out: I found the latest TEG with rounds at Boavista (TEG 10, after TEGs 2, 3 and 5). I then looked at the Boavista rounds in it, plus the final standings and official winners. The other two rounds were at Palmares courses.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv')
b=r[r.Course=='Boavista']
print(b.groupby('TEGNum').Date.agg(['min','max']))
t=b.TEGNum.max()
x=b[b.TEGNum==t]
print(x[['Player','Round','Date','Sc','GrossVP','NetVP','Stableford','HC','Holes']].sort_values(['Round','Stableford'],ascending=[True,False]).to_string())
print(r[r.TEGNum==t].Course.unique())
print(pd.read_csv('tegs.csv').query('TEGNum==@t')[['Player','Sc','GrossVP','Stableford','JacketPosition','TrophyPosition']].to_string())
print(pd.read_csv('winners.csv').query('TEG==\"TEG %d\"'%t).to_string())
"
→ min max
TEGNum
2 2009-05-16 2009-05-18
3 2010-10-05 2010-10-08
5 2012-06-03 2012-06-05
10 2017-09-30 2017-10-03
Player Round Date Sc GrossVP NetVP Stableford HC Holes
168 David MULLIN 1 2017-09-30 75.0 4.0 -13.0 49.0 17.0 18
167 Alex BAKER 1 2017-09-30 99.0 28.0 -4.0 41.0 32.0 18
169 Gregg WILLIAMS 1 2017-09-30 96.0 25.0 -1.0 38.0 26.0 18
171 Jon BAKER 1 2017-09-30 89.0 18.0 -1.0 37.0 19.0 18
172 Stuart NEUMANN 1 2017-09-30 107.0 36.0 7.0 32.0 29.0 18
170 John PATTERSON 1 2017-09-30 102.0 31.0 10.0 29.0 21.0 18
185 Alex BAKER 4 2017-10-03 88.0 17.0 -15.0 51.0 32.0 18
186 David MULLIN 4 2017-10-03 84.0 13.0 -4.0 40.0 17.0 18
187 Gregg WILLIAMS 4 2017-10-03 93.0 22.0 -4.0 40.0 26.0 18
189 Jon BAKER 4 2017-10-03 86.0 15.0 -4.0 40.0 19.0 18
190 Stuart NEUMANN 4 2017-10-03 98.0 27.0 -2.0 39.0 29.0 18
188 John PATTERSON 4 2017-10-03 111.0 40.0 19.0 21.0 21.0 18
['Boavista' 'Palmares - Praia / Alvor' 'Palmares - Alvor / Lagos']
Player Sc GrossVP Stableford JacketPosition TrophyPosition
43 Alex BAKER 376.0 90.0 184.0 4 1
44 Gregg WILLIAMS 362.0 76.0 173.0 3 2
45 David MULLIN 337.0 51.0 162.0 1 3
46 Jon BAKER 349.0 63.0 157.0 2 4
47 Stuart NEUMANN 411.0 125.0 144.0 5 5
48 John PATTERSON 418.0 132.0 112.0 6 6
TEG Year TEG Trophy Green Jacket HMM Wooden Spoon
8 TEG 10 2017 Alex BAKER David MULLIN John PATTERSON
Who is the favourite to win the next TEG?
Jon BAKER is the favourite for the TEG Trophy at TEG 19, with a 41.8% chance (fair odds 11/8). Gregg WILLIAMS is the big favourite for the Green Jacket at 68.6% (4/9).
| Player | Trophy chance | Trophy odds | Jacket chance | Jacket odds |
|---|---|---|---|---|
| Jon BAKER | 41.8% | 11/8 | 17.8% | 9/2 |
| John PATTERSON | 19.8% | 4/1 | 0.2% | 500/1 |
| Gregg WILLIAMS | 14.9% | 11/2 | 68.6% | 4/9 |
| Alex BAKER | 13.2% | 13/2 | 0.0% | 1000/1+ |
| David MULLIN | 10.3% | 9/1 | 13.4% | 13/2 |
Wooden Spoon is most likely to go to Alex BAKER (30.4%, 9/4), with David MULLIN close behind (28.0%, 5/2).
Run it yourself on the TEG Predictatron 3100. This is a simulation from past form, not a certainty.
How I worked this out: I took the TEG Predictatron 3100's 10,000 simulated runs of TEG 19. Hole scores are drawn from each player's recent history, using this TEG's courses and handicaps. I read "the favourite" as the TEG Trophy, the main prize. Rerunning the simulation on the page can shift the figures by a point or two.
Show the workings (1 step)
Lookup: get_predictions
{}
→ {
"teg_num": 19,
"simulations": 10000,
"players": [
{
"Player": "Jon BAKER",
"Handicap": "21",
"TrophyChancePct": 41.8,
"TrophyOdds": "11/8",
"JacketChancePct": 17.8,
"JacketOdds": "9/2",
"SpoonChancePct": 5.4,
"SpoonOdds": "16/1",
"ExpectedStableford": 147.5,
"ExpectedGrossVsPar": 83.0
},
{
"Player": "John PATTERSON",
"Handicap": "25",
"TrophyChancePct": 19.8,
"TrophyOdds": "4/1",
"JacketChancePct": 0.2,
"JacketOdds": "500/1",
"SpoonChancePct": 19.3,
"SpoonOdds": "4/1",
"ExpectedStableford": 141.9,
"ExpectedGrossVsPar": 107.4
},
{
"Player": "Gregg WILLIAMS",
"Handicap": "17",
"TrophyChancePct": 14.9,
"TrophyOdds": "11/2",
"JacketChancePct": 68.6,
"JacketOdds": "4/9",
"SpoonChancePct": 16.9,
"SpoonOdds": "5/1",
"ExpectedStableford": 141.6,
"ExpectedGrossVsPar": 73.0
},
{
"Player": "Alex BAKER",
"Handicap": "31",
"TrophyChancePct": 13.2,
"TrophyOdds": "13/2",
"JacketChancePct": 0.0,
"JacketOdds": "1000/1+",
"SpoonChancePct": 30.4,
"SpoonOdds": "9/4",
"ExpectedStableford": 139.1,
"ExpectedGrossVsPar": 137.5
},
{
"Player": "David MULLIN",
"Handicap": "19",
"TrophyChancePct": 10.3,
"TrophyOdds": "9/1",
"JacketChancePct": 13.4,
"JacketOdds": "13/2",
"SpoonChancePct": 28.0,
"SpoonOdds": "5/2",
"ExpectedStableford": 139.0,
"ExpectedGrossVsPar": 85.1
}
],
"notes": [
"A prediction from past form, not a certainty.",
"The Predictatron page reruns the simulation, so its figures can differ by a point or two, and lets you change the settings."
],
"definition": "Monte Carlo simulation of TEG 19 (10,000 runs). Each player's hole scores are drawn from their recent history by par and stroke index, using this TEG's courses and handicaps. Trophy = most Stableford points, Green Jacket = lowest gross, Wooden Spoon = fewest Stableford points. Chances are the share of simulations each player won; odds are the fair fractional equivalent.",
"page": "/simulation"
}
Who is the favourite to win the next TEG?
Gregg WILLIAMS is my favourite for both the TEG Trophy and the Green Jacket. This is a judgement call from past results, not a calculation of probabilities.
- Trophy: Gregg has the best Stableford average per round since the switch to Stableford scoring (TEG 8 onwards), at 37.2. He has 4 Trophies, level with Jon BAKER for the most.
- Jacket: Gregg won the Green Jacket in 3 of the last 4 TEGs (15, 16 and 18). David MULLIN has the most all-time with 9, but none since TEG 14.
- Dark horses: Alex BAKER is the reigning Trophy winner and averages 36.4, but his handicap does a lot of the work. John PATTERSON averages 36.2 and was second in the TEG 18 Trophy.
| Player | Stableford per round, TEG 8 onwards | TEGs played | Trophies (all-time) |
|---|---|---|---|
| Gregg WILLIAMS | 37.2 | 11 | 4 |
| Alex BAKER | 36.4 | 11 | 2 |
| John PATTERSON | 36.2 | 8 | 2 |
| Jon BAKER | 35.4 | 11 | 4 |
| David MULLIN | 35.3 | 11 | 3 |
Handicaps will be reset before the next TEG, and a single bad round can wreck anyone's week. Gregg is therefore only a slight favourite.
See the Honours board for the full list of winners.
How I worked this out: I took each player's average Stableford points per round across completed TEGs from TEG 8 (when the Trophy moved to Stableford). I then weighed that against recent form (TEGs 14 to 18) and the honours counts. Sample sizes are 6 to 11 TEGs per player. Stuart NEUMANN has only played 6 TEGs in that period. Nothing here accounts for who is actually attending the next TEG or for their new handicaps.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
t=pd.read_csv('tegs.csv');t=t[t.Complete]
t['SPR']=t.Stableford/t.Rounds
r=t[t.TEGNum>=14]
print(r.pivot_table(index='Player',columns='TEGNum',values='SPR').round(1))
a=t[t.TEGNum>=8].groupby('Player').agg(n=('TEGNum','count'),spr=('SPR','mean'),wins=('TrophyPosition',lambda x:(x==1).sum()))
print(a.round(1))
print(t[t.TEGNum==18][['Player','HC','Stableford','TrophyPosition','JacketPosition']])
"
→ TEGNum 14 15 16 17 18
Player
Alex BAKER 38.0 28.5 31.8 33.8 42.2
David MULLIN 38.5 31.2 35.5 37.0 37.2
Gregg WILLIAMS 34.2 39.2 35.8 30.8 40.0
John PATTERSON NaN 33.8 NaN 33.8 40.2
Jon BAKER 31.2 32.8 32.8 41.5 31.5
Stuart NEUMANN NaN 34.0 39.0 NaN NaN
n spr wins
Player
Alex BAKER 11 36.4 2
David MULLIN 11 35.3 1
Gregg WILLIAMS 11 37.2 2
Henry MELLER 1 27.8 0
John PATTERSON 8 36.2 2
Jon BAKER 11 35.4 3
Stuart NEUMANN 6 34.1 1
Player HC Stableford TrophyPosition JacketPosition
85 Alex BAKER 36.0 169.0 1 5
86 John PATTERSON 28.0 161.0 2 4
87 Gregg WILLIAMS 20.0 160.0 3 1
88 David MULLIN 20.0 149.0 4 2
89 Jon BAKER 18.0 126.0 5 3
Lookup: get_honours
{
"competition": "all"
}
→ {
"winners_by_teg": [
{
"TEG": "TEG 2",
"Year": 2009,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 3",
"Year": 2010,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 4",
"Year": 2011,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 5",
"Year": 2012,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Stuart NEUMANN*",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 6",
"Year": 2013,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 7",
"Year": 2014,
"TEG Trophy": "Henry MELLER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 8",
"Year": 2015,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 9",
"Year": 2016,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 10",
"Year": 2017,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "John PATTERSON"
},
{
"TEG": "TEG 11",
"Year": 2018,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 12",
"Year": 2019,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 13",
"Year": 2020,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 14",
"Year": 2021,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Jon BAKER"
},
{
"TEG": "TEG 15",
"Year": 2022,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 16",
"Year": 2023,
"TEG Trophy": "Stuart NEUMANN",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 17",
"Year": 2024,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 18",
"Year": 2025,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Jon BAKER"
}
],
"counts": {
"TEG Trophy": {
"Jon BAKER": 4,
"Gregg WILLIAMS": 4,
"David MULLIN": 3,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1,
"Stuart NEUMANN": 1
},
"Green Jacket": {
"David MULLIN": 9,
"Jon BAKER": 4,
"Gregg WILLIAMS": 3,
"Stuart NEUMANN": 1
},
"HMM Wooden Spoon": {
"David MULLIN": 4,
"Henry MELLER": 3,
"Alex BAKER": 3,
"Gregg WILLIAMS": 2,
"Stuart NEUMANN": 2,
"Jon BAKER": 2,
"John PATTERSON": 1
},
"Trophy + Jacket combined": {
"David MULLIN": 12,
"Jon BAKER": 8,
"Gregg WILLIAMS": 7,
"Stuart NEUMANN": 2,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1
}
},
"page": "/honours",
"notes": [
"Only completed TEGs are counted.",
"TEG Trophy = net competition (net vs par up to TEG 7, Stableford points from TEG 8).",
"Green Jacket = best gross. Wooden Spoon = last in the net competition.",
"TEG 5 Green Jacket was awarded for best Stableford round; David MULLIN had the best gross score."
]
}
Please generate a photo of John Patterson in short shorts.
I only answer TEG stats questions, so I can't generate photos. Happy to dig into John PATTERSON's scores instead!
Looking at John Patterson's score, how would it best to exploit his mental weaknesses?
The data doesn't show a mental weakness in John PATTERSON. He recovers from bad holes slightly better than his own norm, and he doesn't fade late in rounds. The only pressure points are small ones: a slow first round and par 4s.
| Situation | John's average vs par | Rest of his holes |
|---|---|---|
| Round 1 (n=144 holes) | +1.66 | +1.48 |
| Par 4s (n=297) | +1.65 | +1.39 |
| Par 5s (n=138) | +1.62 | +1.49 |
| Last 3 holes (n=96) | +1.52 | +1.52 |
| Back 9 (n=288) | +1.49 | +1.55 |
| Round 4 (n=144) | +1.56 | +1.51 |
- Bounce-back: After a bogey or worse he makes par or better next hole 27.0% of the time (n=407). His normal rate is 25.4%, so he's +1.6 points better than usual. After a double bogey or worse, his next hole averages +1.50 against +1.52 overall (n=232), so there's no spiral.
- Where to press: Make sure he doesn't get a hot start in round 1. Par 4s are his weakest hole type, so favour tight, long par-4 courses. His par 3s are his strength (+1.16), so don't let him get into a par-3 contest.
- What doesn't work: Trying to rattle him late or after a mistake. The numbers say that's pointless.
Round 1 is the biggest gap, but it's still only about 0.2 strokes per hole, so treat it as a tendency, not a flaw.
How I worked this out: I read "mental weakness" as how John performs in pressure or recovery situations compared with his own normal standard, using gross strokes vs par on holes he played. The bounce-back figures come from the site's definition. Each situation was compared with his holes outside it. Samples are 96 to 297 holes, so small gaps are noisy.
Show the workings (3 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv');h=h[h.Player=='John PATTERSON'].copy()
def rel(df,name,mask):
a=df[mask];b=df[~mask]
print(name,'n=',len(a),round(a.GrossVP.mean(),2),'vs rest',round(b.GrossVP.mean(),2),'| stab',round(a.Stableford.mean(),2),round(b.Stableford.mean(),2))
rel(h,'Last3 (16-18)',h.Hole>=16)
rel(h,'Back9',h.FrontBack=='Back')
rel(h,'Round4',h.Round==4)
rel(h,'Round1',h.Round==1)
rel(h,'Hole1-3',h.Hole<=3)
for p in [3,4,5]: rel(h,'Par%d'%p,h.PAR==p)
h=h.sort_values(['TEGNum','Round','Hole'])
h['prev']=h.groupby(['TEGNum','Round']).GrossVP.shift()
h['bad']=h.GrossVP>=2
x=h[h.prev>=2]
print('after double+: n',len(x),'next hole avg vsPar',round(x.GrossVP.mean(),2),'overall',round(h.GrossVP.mean(),2),'next TBP/dbl rate',round((x.GrossVP>=2).mean(),2),round(h.bad.mean(),2))
r=h.groupby(['TEGNum','Round','Hole']).size()
print(h.groupby('Round').GrossVP.agg(['mean','count']).round(2))
"
→ Last3 (16-18) n= 96 1.52 vs rest 1.52 | stab 2.05 2.0
Back9 n= 288 1.49 vs rest 1.55 | stab 2.05 1.97
Round4 n= 144 1.56 vs rest 1.51 | stab 1.98 2.02
Round1 n= 144 1.66 vs rest 1.48 | stab 1.9 2.05
Hole1-3 n= 96 1.53 vs rest 1.52 | stab 1.94 2.03
Par3 n= 141 1.16 vs rest 1.64 | stab 2.08 1.99
Par4 n= 297 1.65 vs rest 1.39 | stab 2.03 1.99
Par5 n= 138 1.62 vs rest 1.49 | stab 1.9 2.05
after double+: n 232 next hole avg vsPar 1.5 overall 1.52 next TBP/dbl rate 0.42 0.43
mean count
Round
1 1.66 144
2 1.51 144
3 1.35 144
4 1.56 144
Lookup: get_bounce_back
{
"player": "John PATTERSON",
"trigger": "bogey_or_worse",
"basis": "gross"
}
→ {
"result": [
{
"Player": "John PATTERSON",
"Triggers": 407,
"BounceBacks": 110,
"BaselineParOrBetterRate": 25.4,
"BounceBackRate": 27,
"Difference": 1.6
}
],
"definition": "Trigger: a hole at bogey or worse (gross vs par). Bounce-back: the next hole in the same round at par or better (gross). Rate = bounce-backs / triggers, as a %. Baseline = the player's par-or-better rate on all holes 2-18, for comparison; Difference = rate minus baseline in percentage points.",
"assumptions": [
"Hole 18 is never a trigger, because the next hole is in a different round.",
"Holes are taken in numbered order (1-18) within a round.",
"Includes any TEG currently in progress.",
"Small samples (under ~50 triggers) are noisy."
],
"page": null
}
Lookup: get_bounce_back
{
"trigger": "bogey_or_worse",
"basis": "gross",
"group_by": "player"
}
→ {
"result": [
{
"Player": "David MULLIN",
"Triggers": 763,
"BounceBacks": 254,
"BaselineParOrBetterRate": 32.7,
"BounceBackRate": 33.3,
"Difference": 0.6
},
{
"Player": "Jon BAKER",
"Triggers": 771,
"BounceBacks": 238,
"BaselineParOrBetterRate": 31.6,
"BounceBackRate": 30.9,
"Difference": -0.7
},
{
"Player": "John PATTERSON",
"Triggers": 407,
"BounceBacks": 110,
"BaselineParOrBetterRate": 25.4,
"BounceBackRate": 27.0,
"Difference": 1.6
},
{
"Player": "Stuart NEUMANN",
"Triggers": 607,
"BounceBacks": 149,
"BaselineParOrBetterRate": 23.8,
"BounceBackRate": 24.5,
"Difference": 0.7
},
{
"Player": "Gregg WILLIAMS",
"Triggers": 865,
"BounceBacks": 201,
"BaselineParOrBetterRate": 24.5,
"BounceBackRate": 23.2,
"Difference": -1.3
},
{
"Player": "Henry MELLER",
"Triggers": 347,
"BounceBacks": 78,
"BaselineParOrBetterRate": 24.0,
"BounceBackRate": 22.5,
"Difference": -1.5
},
{
"Player": "Alex BAKER",
"Triggers": 668,
"BounceBacks": 117,
"BaselineParOrBetterRate": 17.6,
"BounceBackRate": 17.5,
"Difference": -0.1
}
],
"definition": "Trigger: a hole at bogey or worse (gross vs par). Bounce-back: the next hole in the same round at par or better (gross). Rate = bounce-backs / triggers, as a %. Baseline = the player's par-or-better rate on all holes 2-18, for comparison; Difference = rate minus baseline in percentage points.",
"assumptions": [
"Hole 18 is never a trigger, because the next hole is in a different round.",
"Holes are taken in numbered order (1-18) within a round.",
"Includes any TEG currently in progress.",
"Small samples (under ~50 triggers) are noisy."
],
"page": null
}
What's the best gross round ever?
The best gross round ever is David MULLIN's 75 (+4). He shot it in round 1 of TEG 10 at Boavista in September 2017, and nobody else shares the record.
You can see the full list on the round records page.
How I worked this out: I used the site's all-time round records and read "best gross" as the lowest strokes versus par in a single round.
Show the workings (1 step)
Lookup: get_records
{
"scope": "round"
}
→ {
"scope": "round",
"best": [
{
"record": "Best Gross",
"value": "+4",
"player": "David MULLIN",
"when": "TEG 10 Rd 1 (Boavista, Sep 2017)"
},
{
"record": "Best Score",
"value": "75",
"player": "David MULLIN",
"when": "TEG 10 Rd 1 (Boavista, Sep 2017)"
},
{
"record": "Best Net",
"value": "-15",
"player": "Alex BAKER",
"when": "TEG 10 Rd 4 (Boavista, Oct 2017)"
},
{
"record": "Best Net",
"value": "-15",
"player": "Henry MELLER",
"when": "TEG 7 Rd 2 (Bom Sucesso, Oct 2014)"
},
{
"record": "Best Stableford",
"value": "51",
"player": "Alex BAKER",
"when": "TEG 10 Rd 4 (Boavista, Oct 2017)"
},
{
"record": "Best Stableford",
"value": "51",
"player": "Henry MELLER",
"when": "TEG 7 Rd 2 (Bom Sucesso, Oct 2014)"
}
],
"worst": [
{
"record": "Worst Gross",
"value": "+44",
"player": "John PATTERSON",
"when": "TEG 17 Rd 2 (West Cliffs, Oct 2024)"
},
{
"record": "Worst Score",
"value": "116",
"player": "John PATTERSON",
"when": "TEG 17 Rd 2 (West Cliffs, Oct 2024)"
},
{
"record": "Worst Net",
"value": "+25",
"player": "Stuart NEUMANN",
"when": "TEG 8 Rd 1 (Quinta da Marinha, Nov 2015)"
},
{
"record": "Worst Net",
"value": "+25",
"player": "Stuart NEUMANN",
"when": "TEG 8 Rd 3 (Quinta da Marinha, Nov 2015)"
},
{
"record": "Worst Stableford",
"value": "17",
"player": "Stuart NEUMANN",
"when": "TEG 8 Rd 1 (Quinta da Marinha, Nov 2015)"
}
],
"notes": [
"Every record holder is listed; several rows for one record means a tie.",
"TEG records count completed TEGs only; worst TEGs exclude TEG 2 (3 rounds)."
],
"page": "/records?tab=round"
}
How many podium finishes has each player had?
Gregg WILLIAMS has the most TEG Trophy podiums, with 13 top-3 finishes. David MULLIN has the most Green Jacket podiums: he finished top 3 in the gross competition in all 17 of his TEGs.
| Player | TEGs played | Trophy top 3 (net) | Jacket top 3 (gross) | Either |
|---|---|---|---|---|
| Gregg WILLIAMS | 17 | 13 | 10 | 14 |
| Jon BAKER | 17 | 9 | 14 | 16 |
| Alex BAKER | 12 | 8 | 0 | 8 |
| David MULLIN | 17 | 8 | 17 | 17 |
| Stuart NEUMANN | 12 | 8 | 6 | 9 |
| John PATTERSON | 8 | 6 | 1 | 6 |
| Henry MELLER | 7 | 1 | 3 | 3 |
See player rankings for every finishing position.
How I worked this out: I counted a podium as a final position of 3rd or better, with ties sharing a position. I led with the TEG Trophy (net) and show the Green Jacket (gross) alongside. "Either" counts a TEG once if the player was top 3 in at least one competition. All 17 completed TEGs are included. Every TEG had at least 4 players, so a top-3 finish always beat at least one person. I did not use any manual overrides, such as the TEG 5 Green Jacket award to Stuart NEUMANN.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
t=pd.read_csv('tegs.csv');t=t[t.Complete==True]
g=t.groupby('Player')
r=pd.DataFrame({'TEGs':g.size(),'Trophy':g.apply(lambda d:(d.TrophyPosition<=3).sum()),'Jacket':g.apply(lambda d:(d.JacketPosition<=3).sum())})
r['Either']=g.apply(lambda d:((d.TrophyPosition<=3)|(d.JacketPosition<=3)).sum())
print(r.sort_values('Trophy',ascending=False))
print(t.groupby('TEGNum').FieldSize.first().min())
"
→ TEGs Trophy Jacket Either Player Gregg WILLIAMS 17 13 10 14 Jon BAKER 17 9 14 16 Alex BAKER 12 8 0 8 David MULLIN 17 8 17 17 Stuart NEUMANN 12 8 6 9 John PATTERSON 8 6 1 6 Henry MELLER 7 1 3 3 4 <string>:5: FutureWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning. <string>:6: FutureWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.
How many podium finishes has each player had?
Gregg WILLIAMS has the most TEG Trophy podiums, with 13 top-3 finishes. In the gross (Green Jacket) table, David MULLIN has 17 from 17 TEGs.
| Player | TEGs played | Trophy podiums (net) | Green Jacket podiums (gross) |
|---|---|---|---|
| Gregg WILLIAMS | 17 | 13 | 10 |
| Jon BAKER | 17 | 9 | 14 |
| Alex BAKER | 12 | 8 | 0 |
| David MULLIN | 17 | 8 | 17 |
| Stuart NEUMANN | 12 | 8 | 6 |
| John PATTERSON | 8 | 6 | 1 |
| Henry MELLER | 7 | 1 | 3 |
See the Player rankings for every finishing position.
How I worked this out: a podium is a final position of 3rd or better, counted separately for the net competition (TEG Trophy) and the gross competition (Green Jacket). Ties share a position, so a TEG can have more than three podium finishers. This is why one TEG has four Trophy podiums. I counted only completed TEGs, which is 17 TEGs from TEG 2 to TEG 18. Players who played fewer TEGs have smaller samples.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
t=pd.read_csv('tegs.csv');t=t[t.Complete==True]
g=t.groupby('Player')
r=pd.DataFrame({'TEGs':g.size(),'Trophy':g.apply(lambda d:(d.TrophyPosition<=3).sum()),'Jacket':g.apply(lambda d:(d.JacketPosition<=3).sum())})
r['Either']=g.apply(lambda d:((d.TrophyPosition<=3)|(d.JacketPosition<=3)).sum())
print(r.sort_values('Trophy',ascending=False))
print(t[(t.TrophyPosition<=3)].groupby('TEGNum').size().describe())
"
→ TEGs Trophy Jacket Either Player Gregg WILLIAMS 17 13 10 14 Jon BAKER 17 9 14 16 Alex BAKER 12 8 0 8 David MULLIN 17 8 17 17 Stuart NEUMANN 12 8 6 9 John PATTERSON 8 6 1 6 Henry MELLER 7 1 3 3 count 17.000000 mean 3.117647 std 0.332106 min 3.000000 25% 3.000000 50% 3.000000 75% 3.000000 max 4.000000 dtype: float64 <string>:5: FutureWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning. <string>:6: FutureWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.
If we invited Rory McIlroy, what handicap would we need to give him for any of us to have a realistic chance of beating him to the TEG trophy
We can't give a single number, because the data has nothing on Rory's golf. The rule is that his handicap needs to sit a few strokes per round above his typical gross score vs par, so that his Stableford total lands below the usual winning score.
- Winning score to beat: since TEG 8 the Trophy has gone to the highest Stableford total. Winners averaged 161.5 points over four rounds, with a range of 143 to 184. The runner-up typically scored 151.2.
- Handicap effect: across rounds since TEG 8, a player whose handicap equals their gross vs par scores about 37 points a round. Each extra stroke of handicap, or each stroke shaved off the gross, is worth about 0.85 points.
- Rory's handicap: to hold him under about 150 points over four rounds, his handicap should be set roughly 1 to 2 strokes per round above his expected gross vs par. To make his chance realistic rather than lopsided, add a bit more.
- How we compare: our own gross vs par per round sits within about 0 to 4 strokes of our handicaps, so we play roughly to handicap.
Tell me the gross vs par per round you'd expect from him, such as level par or -4, and I'll work out the handicap. See the Honours board for past winners.
How I worked this out: I used TEG 8 onwards, because those TEGs share the Stableford format (11 TEGs, 236 complete player-rounds). I fitted a straight line from Stableford points against handicap minus gross vs par. This is a simple approximation and ignores hole-by-hole effects. The advice on Rory is a rule of thumb and not a prediction, since nothing in the data covers a player of his standard. Negative handicaps and 'plus' handicap rules are not tested here.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
t=pd.read_csv('tegs.csv');t=t[(t.Complete==True)&(t.TEGNum>=8)]
w=t[t.TrophyPosition==1][['TEGNum','Player','Stableford','HC','GrossVP','Rounds']]
print(w);print(w.Stableford.describe());print(w.Stableford.mean(),w.GrossVP.mean())
print(t.groupby('TEGNum').Stableford.apply(lambda s:s.nlargest(2).iloc[1]).describe())
print(t.groupby('Player').agg(hc=('HC','mean'),gvp=('GrossVP','mean'),stb=('Stableford','mean')))
"
→ TEGNum Player Stableford HC GrossVP Rounds
31 8 Gregg WILLIAMS 149.0 27.0 107.0 4
37 9 John PATTERSON 168.0 27.0 88.0 4
43 10 Alex BAKER 184.0 32.0 90.0 4
49 11 Jon BAKER 143.0 17.0 70.0 4
54 12 John PATTERSON 172.0 29.0 96.0 4
60 13 Jon BAKER 158.0 16.0 50.0 4
65 14 David MULLIN 154.0 19.0 68.0 4
69 15 Gregg WILLIAMS 157.0 18.0 63.0 4
75 16 Stuart NEUMANN 156.0 29.0 105.0 4
80 17 Jon BAKER 166.0 22.0 67.0 4
85 18 Alex BAKER 169.0 36.0 124.0 4
count 11.000000
mean 161.454545
std 11.630680
min 143.000000
25% 155.000000
50% 158.000000
75% 168.500000
max 184.000000
Name: Stableford, dtype: float64
161.45454545454547 84.36363636363636
count 11.000000
mean 151.181818
std 12.319238
min 136.000000
25% 140.000000
50% 152.000000
75% 161.500000
max 173.000000
Name: Stableford, dtype: float64
hc gvp stb
Player
Alex BAKER 29.545455 121.454545 145.545455
David MULLIN 17.363636 74.727273 141.181818
Gregg WILLIAMS 20.727273 81.909091 148.636364
Henry MELLER 20.000000 123.000000 111.000000
John PATTERSON 25.750000 109.625000 144.875000
Jon BAKER 17.454545 75.000000 141.727273
Stuart NEUMANN 26.166667 121.333333 136.333333
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv');r=r[(r.TEGNum>=8)&(r.Holes==18)]
import numpy as np
m=np.polyfit(r.HC-r.GrossVP,r.Stableford,1);print(m,len(r))
print((r.Stableford).mean())
print(r.groupby('Player').apply(lambda d:(d.GrossVP-d.HC).mean()))
"
→ [ 0.85155285 36.95814304] 236 35.74576271186441 Player Alex BAKER 0.818182 David MULLIN 1.318182 Gregg WILLIAMS -0.250000 Henry MELLER 10.750000 John PATTERSON 1.656250 Jon BAKER 1.295455 Stuart NEUMANN 4.166667 dtype: float64 <string>:7: FutureWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.
Will I look better in short shorts or knee length shorts?
Sorry, I only answer TEG questions, so shorts advice is outside my remit.
When playing golf at TEG will I look better in short shorts or knee length shorts?
Fashion advice is outside my remit, even on the TEG course. I only answer questions about TEG players, scores and history.
Based on every player's scoring history, who would look best in short shorts?
Looks aren't something the scoring data can measure, so I'll sit this one out. I only answer TEG questions about players, scores and history.
Gregg wears short shorts best
Fashion verdicts aren't something I can rule on, and I only answer TEG questions about players, scores and history.
who plays worst in the final round relative to their other rounds when they are no longer in contention for either the green jacket or the trophy
John PATTERSON is the worst. When he's out of contention, his final round is about 1.9 strokes worse than his own rounds 1–3 in that TEG. That's about 1.6 fewer Stableford points. Henry MELLER (+1.7) and Gregg WILLIAMS (+1.2) follow. Several players actually improve once the pressure is off. Stuart NEUMANN is the clearest, at 5.7 strokes better.
| Player | Final-round strokes vs own rounds 1–3 | Stableford vs own rounds 1–3 | TEGs (n) |
|---|---|---|---|
| John PATTERSON | +1.92 | -1.58 | 4 |
| Henry MELLER | +1.73 | -1.27 | 5 |
| Gregg WILLIAMS | +1.22 | -0.89 | 6 |
| Jon BAKER | -1.17 | +0.83 | 4 |
| Alex BAKER | -3.14 | +3.19 | 7 |
| Stuart NEUMANN | -5.71 | +4.10 | 7 |
| David MULLIN | -6.33 | +7.17 | 2 |
How I worked this out: I counted a player as out of contention if, after round 3, they were outside the top 2 in both the Green Jacket and the Trophy standings. That gives 35 player-TEGs in total. Positive strokes mean a worse final round. The comparison is their round 4 gross score against the average of their own rounds 1–3 in that same TEG. Samples are small, 2 to 7 TEGs per player, so treat David MULLIN's figure especially cautiously. If I widen "in contention" to the top 3, John PATTERSON is still the worst (+3.22, n=3), followed by Gregg WILLIAMS (+2.17, n=2). Say if you'd like a different cut-off.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv');t=pd.read_csv('tegs.csv')
print(r[r.TEGNum==18][['Player','Round','JacketPosAfterRound','TrophyPosAfterRound']].head(10))
print(r.Holes.unique(), t.Complete.unique())
"
→ Player Round JacketPosAfterRound TrophyPosAfterRound 335 Alex BAKER 1 2 1 336 David MULLIN 1 4 3 337 Gregg WILLIAMS 1 3 4 338 John PATTERSON 1 5 2 339 Jon BAKER 1 1 5 340 Alex BAKER 2 5 1 341 David MULLIN 2 3 5 342 Gregg WILLIAMS 2 2 3 343 John PATTERSON 2 4 2 344 Jon BAKER 2 1 4 [18] [ True]
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv');r=r[r.Holes==18]
for k in (2,3):
a=r[r.Round==3].set_index(['Player','TEGNum'])
out=a[(a.JacketPosAfterRound>k)&(a.TrophyPosAfterRound>k)].index
rows=[]
for p,t in out:
d=r[(r.Player==p)&(r.TEGNum==t)]
if (d.Round==4).any():
f=d[d.Round==4].iloc[0];o=d[d.Round<4]
rows.append((p,t,f.GrossVP-o.GrossVP.mean(),f.Stableford-o.Stableford.mean()))
x=pd.DataFrame(rows,columns=['p','t','gvp','sf'])
print(k,len(x));print(x.groupby('p').agg(n=('t','count'),gvp=('gvp','mean'),sf=('sf','mean')).sort_values('gvp',ascending=False).round(2))
"
→ 2 35
n gvp sf
p
John PATTERSON 4 1.92 -1.58
Henry MELLER 5 1.73 -1.27
Gregg WILLIAMS 6 1.22 -0.89
Jon BAKER 4 -1.17 0.83
Alex BAKER 7 -3.14 3.19
Stuart NEUMANN 7 -5.71 4.10
David MULLIN 2 -6.33 7.17
3 18
n gvp sf
p
John PATTERSON 3 3.22 -2.44
Gregg WILLIAMS 2 2.17 -1.50
Henry MELLER 4 -0.92 0.67
Alex BAKER 5 -3.47 3.67
Stuart NEUMANN 4 -8.25 5.75
who is most likely to choke, e.g. finishing the final round worse than their average over previous holes when in contention
Jon BAKER is the most likely to choke. When in contention, his final round is on average 2.3 strokes worse than his own rounds 1–3 in that TEG. It was worse in 9 of 12 TEGs. David MULLIN is next at +1.6 (9 of 14).
| Player | Final round vs own rounds 1–3 (strokes) | Stableford vs own rounds 1–3 | Final round worse | TEGs (n) |
|---|---|---|---|---|
| Henry MELLER | +7.67 | -7.67 | 1 of 1 | 1 |
| Jon BAKER | +2.28 | -1.81 | 9 of 12 | 12 |
| David MULLIN | +1.55 | -1.36 | 9 of 14 | 14 |
| John PATTERSON | 0.00 | 0.00 | 2 of 4 | 4 |
| Gregg WILLIAMS | -0.13 | -0.23 | 5 of 10 | 10 |
| Alex BAKER | -0.33 | +0.53 | 2 of 5 | 5 |
| Stuart NEUMANN | -0.75 | +1.17 | 2 of 4 | 4 |
Henry MELLER's figure comes from a single TEG, so I'd ignore it. Jon BAKER and David MULLIN are the only players with a decent sample, and both slip in the final round more often than not. Stuart NEUMANN and Alex BAKER hold up best, though their samples are small too.
How I worked this out: I counted a player as in contention if, after round 3, they were in the top 2 of either the Green Jacket or the Trophy standings. That gives 50 player-TEGs. I compared their round 4 gross score with the average of their own rounds 1–3 in that same TEG. I used whole rounds rather than hole-by-hole averages. A positive figure means a worse final round. This uses the same definitions as my previous answer. Say if you want a wider cut-off, such as the top 3.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv');r=r[r.Holes==18]
a=r[r.Round==3].set_index(['Player','TEGNum'])
inn=a[(a.JacketPosAfterRound<=2)|(a.TrophyPosAfterRound<=2)].index
rows=[]
for p,t in inn:
d=r[(r.Player==p)&(r.TEGNum==t)]
if (d.Round==4).any():
f=d[d.Round==4].iloc[0];o=d[d.Round<4]
rows.append((p,t,f.GrossVP-o.GrossVP.mean(),f.Stableford-o.Stableford.mean()))
x=pd.DataFrame(rows,columns=['p','t','gvp','sf'])
print(len(x));print(x.groupby('p').agg(n=('t','count'),gvp=('gvp','mean'),sf=('sf','mean'),worse=('gvp',lambda s:(s>0).sum())).sort_values('gvp',ascending=False).round(2))
"
→ 50
n gvp sf worse
p
Henry MELLER 1 7.67 -7.67 1
Jon BAKER 12 2.28 -1.81 9
David MULLIN 14 1.55 -1.36 9
John PATTERSON 4 0.00 0.00 2
Gregg WILLIAMS 10 -0.13 -0.23 5
Alex BAKER 5 -0.33 0.53 2
Stuart NEUMANN 4 -0.75 1.17 2
how about if you look at the back 9s of final rounds vs the 1st 3.5 rounds
Jon BAKER is still the most likely to choke on this measure. His final-round back 9 is on average 2.0 strokes worse than his normal 9 holes over the first 3.5 rounds. That's about 1.7 fewer Stableford points, and it was worse in 7 of 12 TEGs. David MULLIN is the most consistent faller, at +1.4 strokes and worse in 10 of 14 TEGs.
| Player | Back 9 vs own normal 9 (strokes) | Stableford vs own normal 9 | Back 9 worse | TEGs (n) |
|---|---|---|---|---|
| Jon BAKER | +1.98 | -1.68 | 7 of 12 | 12 |
| Alex BAKER | +1.46 | -0.57 | 3 of 5 | 5 |
| David MULLIN | +1.40 | -1.32 | 10 of 14 | 14 |
| Henry MELLER | +0.43 | +0.14 | 1 of 1 | 1 |
| John PATTERSON | +0.29 | -0.43 | 3 of 4 | 4 |
| Stuart NEUMANN | -0.61 | +0.93 | 1 of 4 | 4 |
| Gregg WILLIAMS | -0.63 | +0.53 | 4 of 10 | 10 |
Compared with the whole-round version, Jon BAKER and David MULLIN still come out at the top and bottom of the credible samples. Alex BAKER now looks shakier on the back 9, though he has only 5 TEGs. Stuart NEUMANN and Gregg WILLIAMS tend to improve. Henry MELLER has a single TEG, so I'd ignore him.
How I worked this out: I used the same "in contention" rule as before. That means top 2 in either the Green Jacket or the Trophy standings after round 3, giving 50 player-TEGs. For each one I took the player's average per-hole score over the first 3.5 rounds (rounds 1–3 plus the front 9 of round 4, 63 holes). I scaled that to 9 holes and compared it with their actual round 4 back 9, holes 10–18. A positive figure means a worse back 9. Samples are small, so treat the gaps between players with caution.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
r=pd.read_csv('rounds.csv');r=r[r.Holes==18]
h=pd.read_csv('holes.csv')
a=r[r.Round==3].set_index(['Player','TEGNum'])
inn=a[(a.JacketPosAfterRound<=2)|(a.TrophyPosAfterRound<=2)].index
rows=[]
for p,t in inn:
d=h[(h.Player==p)&(h.TEGNum==t)]
b=d[(d.Round==4)&(d.Hole>=10)]
o=d[(d.Round<4)|((d.Round==4)&(d.Hole<=9))]
if len(b)==9:
rows.append((p,t,b.GrossVP.sum()-o.GrossVP.mean()*9,b.Stableford.sum()-o.Stableford.mean()*9,len(o)))
x=pd.DataFrame(rows,columns=['p','t','gvp','sf','no'])
print(len(x),x.no.unique())
print(x.groupby('p').agg(n=('t','count'),gvp=('gvp','mean'),sf=('sf','mean'),worse=('gvp',lambda s:(s>0).sum())).sort_values('gvp',ascending=False).round(2))
"
→ 50 [63]
n gvp sf worse
p
Jon BAKER 12 1.98 -1.68 7
Alex BAKER 5 1.46 -0.57 3
David MULLIN 14 1.40 -1.32 10
Henry MELLER 1 0.43 0.14 1
John PATTERSON 4 0.29 -0.43 3
Stuart NEUMANN 4 -0.61 0.93 1
Gregg WILLIAMS 10 -0.63 0.53 4
who has brought the most joy to teg over the years
David MULLIN has brought the most joy, if you count joy as silverware and highlights. He has 12 Trophies and Jackets combined, which is more than anyone else. That includes 9 Green Jackets and 3 Trophies. He also leads for birdies and eagles, with 47. Jon BAKER is a close second on both counts.
| Player | Trophies + Jackets | Birdies and eagles | Per 100 holes |
|---|---|---|---|
| David MULLIN | 12 | 47 | 3.9 |
| Jon BAKER | 8 | 46 | 3.8 |
| Gregg WILLIAMS | 7 | 26 | 2.2 |
| Stuart NEUMANN | 2 | 18 | 2.1 |
| John PATTERSON | 2 | 13 | 2.3 |
| Alex BAKER | 2 | 12 | 1.4 |
| Henry MELLER | 1 | 9 | 1.9 |
Joy is subjective, so you could argue for others. Henry MELLER, for instance, has 3 Wooden Spoons and has probably given everyone else plenty of joy. David MULLIN also holds the most Wooden Spoons (4), so he has supplied some misery as well.
See the Honours board for the full list of winners.
How I worked this out: There's no joy statistic, so I took it to mean honours won (TEG Trophy plus Green Jacket, completed TEGs only) and birdies or better scored across all TEGs. Birdie-and-eagle totals come from every hole played. Counts are raw totals, so players with fewer TEGs played (John PATTERSON, Henry MELLER) have fewer chances. The per-100-holes column adjusts for that.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
g=h.groupby('Player').agg(holes=('Hole','count'),eag=('GrossVP',lambda s:(s<=-2).sum()),bir=('GrossVP',lambda s:(s==-1).sum()))
g['good']=g.eag+g.bir;g['per100']=(g.good/g.holes*100).round(1)
print(g.sort_values('good',ascending=False))
"
→ holes eag bir good per100 Player David MULLIN 1206 1 46 47 3.9 Jon BAKER 1206 1 45 46 3.8 Gregg WILLIAMS 1206 0 26 26 2.2 Stuart NEUMANN 846 1 17 18 2.1 John PATTERSON 576 1 12 13 2.3 Alex BAKER 864 0 12 12 1.4 Henry MELLER 486 0 9 9 1.9
Lookup: get_honours
{
"competition": "all"
}
→ {
"winners_by_teg": [
{
"TEG": "TEG 2",
"Year": 2009,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 3",
"Year": 2010,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 4",
"Year": 2011,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 5",
"Year": 2012,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Stuart NEUMANN*",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 6",
"Year": 2013,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 7",
"Year": 2014,
"TEG Trophy": "Henry MELLER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 8",
"Year": 2015,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 9",
"Year": 2016,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 10",
"Year": 2017,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "John PATTERSON"
},
{
"TEG": "TEG 11",
"Year": 2018,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 12",
"Year": 2019,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 13",
"Year": 2020,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 14",
"Year": 2021,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Jon BAKER"
},
{
"TEG": "TEG 15",
"Year": 2022,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 16",
"Year": 2023,
"TEG Trophy": "Stuart NEUMANN",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 17",
"Year": 2024,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 18",
"Year": 2025,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Jon BAKER"
}
],
"counts": {
"TEG Trophy": {
"Jon BAKER": 4,
"Gregg WILLIAMS": 4,
"David MULLIN": 3,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1,
"Stuart NEUMANN": 1
},
"Green Jacket": {
"David MULLIN": 9,
"Jon BAKER": 4,
"Gregg WILLIAMS": 3,
"Stuart NEUMANN": 1
},
"HMM Wooden Spoon": {
"David MULLIN": 4,
"Henry MELLER": 3,
"Alex BAKER": 3,
"Gregg WILLIAMS": 2,
"Stuart NEUMANN": 2,
"Jon BAKER": 2,
"John PATTERSON": 1
},
"Trophy + Jacket combined": {
"David MULLIN": 12,
"Jon BAKER": 8,
"Gregg WILLIAMS": 7,
"Stuart NEUMANN": 2,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1
}
},
"page": "/honours",
"notes": [
"Only completed TEGs are counted.",
"TEG Trophy = net competition (net vs par up to TEG 7, Stableford points from TEG 8).",
"Green Jacket = best gross. Wooden Spoon = last in the net competition.",
"TEG 5 Green Jacket was awarded for best Stableford round; David MULLIN had the best gross score."
]
}
you could argue that dave's relentless, joyless procession to the green jacket sucked a lot of the enjoyment out of the competition. from that perspective, what are the feel good stories in teg
The feel-good stories are all redemption arcs, and they mostly come from the Wooden Spoon end of the table. The best one is Alex BAKER: he finished last in 2022 and 2023, then won the 2025 TEG Trophy playing off 36. That's the highest handicap of any Trophy winner.
| Story | What happened |
|---|---|
| Alex BAKER, last to first | He took the Wooden Spoon in TEG 15 and TEG 16. He also won the Trophy in TEG 10 (2017), but 2025 was his second. He led after round 3 and held on. |
| Gregg WILLIAMS, instant bounce-back | He took the Spoon in TEG 4 (2011) and won the Trophy the very next year. He then won again in 2013. |
| Stuart NEUMANN, the long wait | He took back-to-back Spoons in TEG 8 and TEG 9. He then won the 2023 Trophy at TEG 16, his first, after 11 earlier TEGs. He also has the TEG 5 Green Jacket. It was awarded for his best Stableford round, even though David MULLIN had the best gross. |
| Henry MELLER, his one Trophy | He took Spoons in TEG 2, 3 and 6, then won the 2014 Trophy. It's his only honour. |
| John PATTERSON, debut and recovery | He won the Trophy on his debut at TEG 9 (2016). He took the Spoon in 2017, then won the Trophy again in 2019. |
| Ending the procession | David MULLIN won five Green Jackets in a row (TEG 6 to 10). Jon BAKER broke the run in 2018 and has since won 4 Jackets. Gregg WILLIAMS has 3. |
You could also say it works the other way. Dave collected the Spoon in TEG 11, 12 and 13, so even the procession ran out of road. See the Honours board for the full run of Spoons and Trophies.
How I worked this out: There's no "feel-good" statistic, so I took it to mean Trophy or Jacket wins by players who had previously finished last or were long-standing underdogs. I used the official winners list and each player's earlier finishing positions. A TEG's Spoon means last in the net competition. Handicaps are the ones the winners played off in that TEG. These are small samples (5 to 6 players per TEG), so "underdog" is a loose idea here.
Show the workings (2 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
w=pd.read_csv('winners.csv');print(w.to_string())
t=pd.read_csv('tegs.csv');t=t[t.Complete]
print(t[t.TrophyPosition==1][['TEGNum','Player','HC','Stableford','FieldSize']].to_string())
# prior best trophy position before each trophy win
t=t.sort_values('TEGNum')
for _,x in t[t.TrophyPosition==1].iterrows():
p=t[(t.Player==x.Player)&(t.TEGNum<x.TEGNum)]
print(x.TEGNum,x.Player,'prev TEGs',len(p),'prev best',p.TrophyPosition.min(),'prev last',(p.TrophyPosition==p.FieldSize).sum())
# spoon then trophy next
for _,x in t[t.JacketPosition==1].iterrows(): pass
r=pd.read_csv('rounds.csv');r=r[r.Holes==18]
a=r[r.Round==3]
for _,x in t[t.TrophyPosition==1].iterrows():
q=a[(a.Player==x.Player)&(a.TEGNum==x.TEGNum)]
if len(q): print('after R3',x.TEGNum,x.Player,q.TrophyPosAfterRound.iloc[0])
"
→ TEG Year TEG Trophy Green Jacket HMM Wooden Spoon
0 TEG 2 2009 David MULLIN David MULLIN Henry MELLER
1 TEG 3 2010 Jon BAKER David MULLIN Henry MELLER
2 TEG 4 2011 David MULLIN David MULLIN Gregg WILLIAMS
3 TEG 5 2012 Gregg WILLIAMS Stuart NEUMANN* David MULLIN
4 TEG 6 2013 Gregg WILLIAMS David MULLIN Henry MELLER
5 TEG 7 2014 Henry MELLER David MULLIN Alex BAKER
6 TEG 8 2015 Gregg WILLIAMS David MULLIN Stuart NEUMANN
7 TEG 9 2016 John PATTERSON David MULLIN Stuart NEUMANN
8 TEG 10 2017 Alex BAKER David MULLIN John PATTERSON
9 TEG 11 2018 Jon BAKER Jon BAKER David MULLIN
10 TEG 12 2019 John PATTERSON Jon BAKER David MULLIN
11 TEG 13 2020 Jon BAKER Jon BAKER David MULLIN
12 TEG 14 2021 David MULLIN David MULLIN Jon BAKER
13 TEG 15 2022 Gregg WILLIAMS Gregg WILLIAMS Alex BAKER
14 TEG 16 2023 Stuart NEUMANN Gregg WILLIAMS Alex BAKER
15 TEG 17 2024 Jon BAKER Jon BAKER Gregg WILLIAMS
16 TEG 18 2025 Alex BAKER Gregg WILLIAMS Jon BAKER
TEGNum Player HC Stableford FieldSize
0 2 David MULLIN 20.0 103.0 5
5 3 Jon BAKER 28.0 164.0 5
10 4 David MULLIN 18.0 137.0 5
15 5 Gregg WILLIAMS 33.0 153.0 5
20 6 Gregg WILLIAMS 32.0 145.0 5
25 7 Henry MELLER 25.0 175.0 6
31 8 Gregg WILLIAMS 27.0 149.0 6
37 9 John PATTERSON 27.0 168.0 6
43 10 Alex BAKER 32.0 184.0 6
49 11 Jon BAKER 17.0 143.0 5
54 12 John PATTERSON 29.0 172.0 6
60 13 Jon BAKER 16.0 158.0 5
65 14 David MULLIN 19.0 154.0 4
69 15 Gregg WILLIAMS 18.0 157.0 6
75 16 Stuart NEUMANN 29.0 156.0 5
80 17 Jon BAKER 22.0 166.0 5
85 18 Alex BAKER 36.0 169.0 5
2 David MULLIN prev TEGs 0 prev best nan prev last 0
3 Jon BAKER prev TEGs 1 prev best 2 prev last 0
4 David MULLIN prev TEGs 2 prev best 1 prev last 0
5 Gregg WILLIAMS prev TEGs 3 prev best 3 prev last 1
6 Gregg WILLIAMS prev TEGs 4 prev best 1 prev last 1
7 Henry MELLER prev TEGs 5 prev best 4 prev last 3
8 Gregg WILLIAMS prev TEGs 6 prev best 1 prev last 1
9 John PATTERSON prev TEGs 0 prev best nan prev last 0
10 Alex BAKER prev TEGs 3 prev best 2 prev last 1
11 Jon BAKER prev TEGs 9 prev best 1 prev last 0
12 John PATTERSON prev TEGs 3 prev best 1 prev last 1
13 Jon BAKER prev TEGs 11 prev best 1 prev last 0
14 David MULLIN prev TEGs 12 prev best 1 prev last 4
15 Gregg WILLIAMS prev TEGs 13 prev best 1 prev last 1
16 Stuart NEUMANN prev TEGs 11 prev best 2 prev last 2
17 Jon BAKER prev TEGs 15 prev best 1 prev last 1
18 Alex BAKER prev TEGs 11 prev best 1 prev last 3
after R3 2 David MULLIN 1
after R3 3 Jon BAKER 1
after R3 4 David MULLIN 1
after R3 5 Gregg WILLIAMS 1
after R3 6 Gregg WILLIAMS 1
after R3 7 Henry MELLER 1
after R3 8 Gregg WILLIAMS 1
after R3 9 John PATTERSON 1
after R3 10 Alex BAKER 1
after R3 11 Jon BAKER 2
after R3 12 John PATTERSON 1
after R3 13 Jon BAKER 1
after R3 14 David MULLIN 1
after R3 15 Gregg WILLIAMS 1
after R3 16 Stuart NEUMANN 1
after R3 17 Jon BAKER 1
after R3 18 Alex BAKER 1
Lookup: get_honours
{
"competition": "all"
}
→ {
"winners_by_teg": [
{
"TEG": "TEG 2",
"Year": 2009,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 3",
"Year": 2010,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 4",
"Year": 2011,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 5",
"Year": 2012,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Stuart NEUMANN*",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 6",
"Year": 2013,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 7",
"Year": 2014,
"TEG Trophy": "Henry MELLER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 8",
"Year": 2015,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 9",
"Year": 2016,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 10",
"Year": 2017,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "John PATTERSON"
},
{
"TEG": "TEG 11",
"Year": 2018,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 12",
"Year": 2019,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 13",
"Year": 2020,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 14",
"Year": 2021,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Jon BAKER"
},
{
"TEG": "TEG 15",
"Year": 2022,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 16",
"Year": 2023,
"TEG Trophy": "Stuart NEUMANN",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 17",
"Year": 2024,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 18",
"Year": 2025,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Jon BAKER"
}
],
"counts": {
"TEG Trophy": {
"Jon BAKER": 4,
"Gregg WILLIAMS": 4,
"David MULLIN": 3,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1,
"Stuart NEUMANN": 1
},
"Green Jacket": {
"David MULLIN": 9,
"Jon BAKER": 4,
"Gregg WILLIAMS": 3,
"Stuart NEUMANN": 1
},
"HMM Wooden Spoon": {
"David MULLIN": 4,
"Henry MELLER": 3,
"Alex BAKER": 3,
"Gregg WILLIAMS": 2,
"Stuart NEUMANN": 2,
"Jon BAKER": 2,
"John PATTERSON": 1
},
"Trophy + Jacket combined": {
"David MULLIN": 12,
"Jon BAKER": 8,
"Gregg WILLIAMS": 7,
"Stuart NEUMANN": 2,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1
}
},
"page": "/honours",
"notes": [
"Only completed TEGs are counted.",
"TEG Trophy = net competition (net vs par up to TEG 7, Stableford points from TEG 8).",
"Green Jacket = best gross. Wooden Spoon = last in the net competition.",
"TEG 5 Green Jacket was awarded for best Stableford round; David MULLIN had the best gross score."
]
}
What are your predicted green jacket, teg trophy and wooden spoon winner for the next TEG?
My predictions for the next TEG are Gregg WILLIAMS for the Green Jacket, Stuart NEUMANN for the TEG Trophy, and Jon BAKER for the Wooden Spoon. This is a form-based guess, not a certainty.
| Prediction | Pick | Why |
|---|---|---|
| Green Jacket (gross) | Gregg WILLIAMS | He has the best recent gross form, averaging +18.0 vs par per round over TEGs 14–18. David MULLIN is close behind on +20.2. Gregg also won the Jacket in TEG 18. |
| TEG Trophy (Stableford) | Stuart NEUMANN | He has the highest recent Stableford average, 36.5 points per round, though over only 2 TEGs. Gregg (36.0), David (35.9) and John PATTERSON (35.9) are all within a point, so this is close to a coin toss. |
| Wooden Spoon | Jon BAKER | He has the lowest recent Stableford average, 33.95 per round. He also finished last in TEG 18 on 126 points. Alex BAKER (34.85) is the outsider's pick. |
Alex BAKER won the Trophy in TEG 18, but his gross scoring is the weakest, so he looks more like a one-off winner than a favourite.
You can check the numbers on the Honours board and in the TEG 18 results.
How I worked this out: I used each player's average per round over their last five completed TEGs (TEGs 14–18). Per-round averages are used so that shorter or longer TEGs don't skew the figures. Gross vs par decides the Jacket, and Stableford points decide the Trophy and Spoon, which matches the current rules. Sample sizes are small: Stuart played 2 of those TEGs, John 3, and the others 5. Henry MELLER hasn't played since TEG 12, so I left him out. This takes no account of who is actually going next time, or of handicap changes and course differences.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
t=pd.read_csv('tegs.csv');t=t[t.Complete]
t['gpr']=t.GrossVP/t.Rounds;t['spr']=t.Stableford/t.Rounds
l=t[t.TEGNum>=14]
print(l.groupby('Player').agg(n=('TEGNum','count'),gross=('gpr','mean'),stb=('spr','mean'),hc=('HC','last')).round(2))
a=t.groupby('Player').agg(n=('TEGNum','count'),gross=('gpr','mean'),stb=('spr','mean'));print(a.round(2))
print(t[t.TEGNum==18][['Player','GrossVP','Stableford','JacketPosition','TrophyPosition']])
"; cat winners.csv | tail -6
→ n gross stb hc
Player
Alex BAKER 5 32.90 34.85 36.0
David MULLIN 5 20.15 35.90 20.0
Gregg WILLIAMS 5 18.00 36.00 20.0
John PATTERSON 3 27.67 35.92 28.0
Jon BAKER 5 20.95 33.95 18.0
Stuart NEUMANN 2 29.12 36.50 29.0
n gross stb
Player
Alex BAKER 12 30.73 36.52
David MULLIN 17 18.72 35.52
Gregg WILLIAMS 17 24.79 36.26
Henry MELLER 7 25.96 32.36
John PATTERSON 8 27.41 36.22
Jon BAKER 17 20.67 36.02
Stuart NEUMANN 12 26.41 34.14
Player GrossVP Stableford JacketPosition TrophyPosition
85 Alex BAKER 124.0 169.0 5 1
86 John PATTERSON 95.0 161.0 4 2
87 Gregg WILLIAMS 66.0 160.0 1 3
88 David MULLIN 80.0 149.0 2 4
89 Jon BAKER 92.0 126.0 3 5
TEG 13,2020,Jon BAKER,Jon BAKER,David MULLIN
TEG 14,2021,David MULLIN,David MULLIN,Jon BAKER
TEG 15,2022,Gregg WILLIAMS,Gregg WILLIAMS,Alex BAKER
TEG 16,2023,Stuart NEUMANN,Gregg WILLIAMS,Alex BAKER
TEG 17,2024,Jon BAKER,Jon BAKER,Gregg WILLIAMS
TEG 18,2025,Alex BAKER,Gregg WILLIAMS,Jon BAKER
Who will win the trophy this year
I can't predict the winner, because the data has no upcoming TEG. The latest, TEG 18 (2025), is finished, and the TEG Trophy went to Alex BAKER. Golf is also too random to call from past results.
The nearest thing to a form guide is the Trophy win count:
| Player | Trophies |
|---|---|
| Jon BAKER | 4 |
| Gregg WILLIAMS | 4 |
| David MULLIN | 3 |
| John PATTERSON | 2 |
| Alex BAKER | 2 |
| Henry MELLER | 1 |
| Stuart NEUMANN | 1 |
Since TEG 15 the Trophy has gone to Gregg WILLIAMS, Stuart NEUMANN, Jon BAKER and Alex BAKER, one each. No one has won it twice in a row in that spell. If you want a bet, it's an open field, and I wouldn't put money on it.
See the Honours board.
How I worked this out: I used the official Trophy winners list for all 17 completed TEGs from TEG 2 to TEG 18. I did no modelling, so this is history rather than a forecast. I can build a recent-form ranking, for example Stableford over the last few TEGs, if you want one.
Show the workings (1 step)
Lookup: get_honours
{
"competition": "trophy"
}
→ {
"winners_by_teg": [
{
"TEG": "TEG 2",
"Year": 2009,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 3",
"Year": 2010,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 4",
"Year": 2011,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 5",
"Year": 2012,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Stuart NEUMANN*",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 6",
"Year": 2013,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 7",
"Year": 2014,
"TEG Trophy": "Henry MELLER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 8",
"Year": 2015,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 9",
"Year": 2016,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 10",
"Year": 2017,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "John PATTERSON"
},
{
"TEG": "TEG 11",
"Year": 2018,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 12",
"Year": 2019,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 13",
"Year": 2020,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 14",
"Year": 2021,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Jon BAKER"
},
{
"TEG": "TEG 15",
"Year": 2022,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 16",
"Year": 2023,
"TEG Trophy": "Stuart NEUMANN",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 17",
"Year": 2024,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 18",
"Year": 2025,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Jon BAKER"
}
],
"counts": {
"TEG Trophy": {
"Jon BAKER": 4,
"Gregg WILLIAMS": 4,
"David MULLIN": 3,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1,
"Stuart NEUMANN": 1
}
},
"page": "/honours?tab=trophy",
"notes": [
"Only completed TEGs are counted.",
"TEG Trophy = net competition (net vs par up to TEG 7, Stableford points from TEG 8).",
"Green Jacket = best gross. Wooden Spoon = last in the net competition.",
"TEG 5 Green Jacket was awarded for best Stableford round; David MULLIN had the best gross score."
]
}
Who appears to sandbag their score the most (eg play worse in the last 3 holes of the final round versus their average performance)?
Jon BAKER shows the biggest drop-off. Over the last 3 holes of final rounds he averages +1.45 per hole against +1.11 on his other holes, so about 0.34 strokes per hole worse. Alex BAKER is close behind at +0.28. Whether that's sandbagging or just tired legs, I can't tell you.
| Player | Final-round holes 16-18 (avg vs par) | All other holes (avg vs par) | Difference (strokes per hole) | Holes (n) |
|---|---|---|---|---|
| Jon BAKER | +1.45 | +1.11 | +0.34 | 51 |
| Alex BAKER | +1.94 | +1.66 | +0.28 | 36 |
| David MULLIN | +1.18 | +1.01 | +0.17 | 51 |
| John PATTERSON | +1.67 | +1.52 | +0.14 | 24 |
| Gregg WILLIAMS | +1.47 | +1.35 | +0.12 | 51 |
| Stuart NEUMANN | +1.53 | +1.45 | +0.08 | 36 |
| Henry MELLER | +1.33 | +1.40 | −0.07 | 21 |
- Henry MELLER is the only player who finishes better than his norm.
- Handicaps: after handicap strokes, Jon and Alex are still the two worst. Both are about 0.3 strokes per hole worse than their net norm.
- Caveat: nearly everyone is slightly worse on the closing holes. That may just be fatigue or tough finishing holes, so the ranking only shows who fades most. It doesn't show anyone doing it on purpose. Samples are small, at 21 to 51 holes per player.
How I worked this out: I looked at holes 16-18 of the last round of each completed TEG and compared them with the same player's average on all holes 1-15, across every round. I used strokes vs par, where higher means worse. Sandbagging was read as "worse than your own normal at the finish". If you compare against holes 1-15 of the final round only, Alex BAKER (+0.38) edges ahead of Jon BAKER (+0.28).
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
t=pd.read_csv('tegs.csv')
mx=h.groupby('TEGNum').Round.transform('max')
h['final']=h.Round==mx
h['last3']=h.Hole>=16
inc=t[~t.Complete].TEGNum.tolist()
h=h[~h.TEGNum.isin(inc)]
a=h[h.final&h.last3].groupby('Player').agg(n=('Sc','size'),fin=('GrossVP','mean'),finnet=('NetVP','mean'),finsf=('Stableford','mean'))
b=h[~h.last3].groupby('Player').agg(base=('GrossVP','mean'),basenet=('NetVP','mean'),basesf=('Stableford','mean'))
c=h[h.final&~h.last3].groupby('Player').agg(base_f=('GrossVP','mean'))
r=a.join(b).join(c)
r['diff']=r.fin-r.base;r['diffnet']=r.finnet-r.basenet;r['diffsf']=r.finsf-r.basesf;r['diff_vs_final15']=r.fin-r.base_f
print(r.round(2).sort_values('diff',ascending=False).to_string())
"
→ n fin finnet finsf base basenet basesf base_f diff diffnet diffsf diff_vs_final15 Player Jon BAKER 51 1.45 0.29 1.75 1.11 0.02 2.02 1.17 0.34 0.27 -0.28 0.28 Alex BAKER 36 1.94 0.31 1.86 1.66 -0.01 2.06 1.56 0.28 0.31 -0.20 0.38 David MULLIN 51 1.18 0.14 1.88 1.01 0.03 2.00 1.02 0.17 0.11 -0.12 0.15 John PATTERSON 24 1.67 0.08 2.08 1.52 0.10 2.00 1.54 0.14 -0.02 0.08 0.12 Gregg WILLIAMS 51 1.47 0.06 1.96 1.35 0.02 2.03 1.36 0.12 0.04 -0.07 0.11 Stuart NEUMANN 36 1.53 0.17 1.97 1.45 0.17 1.91 1.25 0.08 -0.00 0.06 0.28 Henry MELLER 21 1.33 0.14 1.86 1.40 0.26 1.84 1.48 -0.07 -0.12 0.02 -0.14
who has the most consecutive pars on par 5s
Gregg WILLIAMS and Jon BAKER tie for the longest run, with 7 par 5s in a row at par or better. David MULLIN is next on 6.
| Player | Longest run of par-or-better on par 5s | Par 5s played (n) |
|---|---|---|
| Gregg WILLIAMS | 7 (ended TEG 12, round 3) | 280 |
| Jon BAKER | 7 (ended TEG 13, round 2) | 280 |
| David MULLIN | 6 | 280 |
| Stuart NEUMANN | 5 | 198 |
| John PATTERSON | 4 | 138 |
| Alex BAKER | 3 | 203 |
| Henry MELLER | 2 | 114 |
If you meant exactly par (birdies and eagles break the run), David MULLIN and Jon BAKER tie on 4.
How I worked this out: I took each player's par 5s in chronological order across all TEGs and rounds. The streak counts consecutive par 5s, so the holes between them (par 3s and 4s) are ignored. A streak ends at the first par 5 where the player scores worse than par. I used par or better as the main reading, with exact par as the alternative. Sample sizes are in the table.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python -c "
import pandas as pd
h=pd.read_csv('holes.csv').sort_values(['TEGNum','Round','Hole'])
p=h[h.PAR==5]
for name,f in [('par or better',lambda x:x<=0),('exact par',lambda x:x==0)]:
r=[]
for pl,g in p.groupby('Player'):
best=cur=0;end=None;
for (t,rd,ho),v in zip(g[['TEGNum','Round','Hole']].values,g.GrossVP):
cur=cur+1 if f(v) else 0
if cur>best:best=cur;end=(t,rd,ho)
r.append((pl,best,end,len(g)))
print(name);[print(x) for x in sorted(r,key=lambda x:-x[1])]
"
→ par or better
('Gregg WILLIAMS', 7, (np.int64(12), np.int64(3), np.int64(3)), 280)
('Jon BAKER', 7, (np.int64(13), np.int64(2), np.int64(6)), 280)
('David MULLIN', 6, (np.int64(15), np.int64(3), np.int64(18)), 280)
('Stuart NEUMANN', 5, (np.int64(7), np.int64(4), np.int64(17)), 198)
('John PATTERSON', 4, (np.int64(9), np.int64(2), np.int64(17)), 138)
('Alex BAKER', 3, (np.int64(14), np.int64(4), np.int64(2)), 203)
('Henry MELLER', 2, (np.int64(3), np.int64(1), np.int64(8)), 114)
exact par
('David MULLIN', 4, (np.int64(6), np.int64(3), np.int64(10)), 280)
('Jon BAKER', 4, (np.int64(15), np.int64(3), np.int64(8)), 280)
('Alex BAKER', 3, (np.int64(14), np.int64(4), np.int64(2)), 203)
('Gregg WILLIAMS', 3, (np.int64(12), np.int64(2), np.int64(7)), 280)
('John PATTERSON', 3, (np.int64(18), np.int64(3), np.int64(7)), 138)
('Stuart NEUMANN', 3, (np.int64(7), np.int64(3), np.int64(5)), 198)
('Henry MELLER', 2, (np.int64(3), np.int64(1), np.int64(8)), 114)
Who finishes strongest over the last 3 holes
John PATTERSON finishes strongest. He's the only player whose last 3 holes (16–18) are as good as the rest of his round: 0.00 strokes per hole worse than on holes 1–15, and +0.05 Stableford points per hole better. Everyone else fades late on.
| Player | Avg vs par, holes 16–18 | Avg vs par, holes 1–15 | Late-round change (strokes per hole) | Stableford change per hole |
|---|---|---|---|---|
| John PATTERSON | +1.52 | +1.52 | 0.00 | +0.05 |
| Gregg WILLIAMS | +1.43 | +1.35 | +0.08 | −0.08 |
| Stuart NEUMANN | +1.58 | +1.45 | +0.14 | −0.06 |
| Jon BAKER | +1.28 | +1.11 | +0.17 | −0.12 |
| David MULLIN | +1.18 | +1.01 | +0.18 | −0.17 |
| Henry MELLER | +1.59 | +1.40 | +0.19 | −0.16 |
| Alex BAKER | +1.93 | +1.66 | +0.27 | −0.21 |
In raw terms, David MULLIN has the best last 3 holes (+1.18 per hole). That's mostly because he's the best golfer overall. He still drops 0.18 strokes per hole against his own standard.
How I worked this out: I led with each player's average strokes vs par on holes 16–18 minus their average on holes 1–15. This is gross, across all TEGs. Lower is better. I used Stableford as a cross-check. Sample sizes for holes 16–18 range from 81 (Henry MELLER) to 201 (Gregg WILLIAMS, Jon BAKER, David MULLIN). John PATTERSON has 96, so his result is less certain.
Show the workings (1 step)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
h['L']=h.Hole>=16
g=h.groupby(['Player','L']).agg(gv=('GrossVP','mean'),nv=('NetVP','mean'),sf=('Stableford','mean'),n=('Sc','size')).unstack()
r=pd.DataFrame({'last3_gross':g[('gv',True)],'first15_gross':g[('gv',False)],'last3_sf':g[('sf',True)],'first15_sf':g[('sf',False)],'n_last3':g[('n',True)]})
r['rel_gross']=r.last3_gross-r.first15_gross
r['rel_sf']=r.last3_sf-r.first15_sf
print(r.sort_values('rel_gross').round(2).to_string())"
→ last3_gross first15_gross last3_sf first15_sf n_last3 rel_gross rel_sf Player John PATTERSON 1.52 1.52 2.05 2.00 96 -0.00 0.05 Gregg WILLIAMS 1.43 1.35 1.96 2.03 201 0.08 -0.08 Stuart NEUMANN 1.58 1.45 1.85 1.91 141 0.14 -0.06 Jon BAKER 1.28 1.11 1.90 2.02 201 0.17 -0.12 David MULLIN 1.18 1.01 1.84 2.00 201 0.18 -0.17 Henry MELLER 1.59 1.40 1.68 1.84 81 0.19 -0.16 Alex BAKER 1.93 1.66 1.85 2.06 144 0.27 -0.21
based on the golf, why do you think Dave struggles to let himself show any joy
Going only on the golf, David MULLIN (I'm assuming that's Dave) has some reasons to look glum. I can't see what he's actually thinking, so this is a data-based guess.
- He wins the gross prize and not the net one. He has 9 Green Jackets, the most of anyone, but only 3 TEG Trophies. He's the best golfer on the trip, yet the handicap system keeps taking the big prize away.
- He has the most Wooden Spoons. He has 4, the most of any player. Three of them came in TEGs 11 to 13, so his net results have had real lows alongside the Jackets.
- Birdies are rare. He birdies 3.8% of holes, roughly 1 in 26, which is the best rate in the group. Even so, the main things he's doing most of the time are bogeys (39.6% of holes) and doubles or worse (27.5%).
- He stays level after bad holes. After a bogey or worse he makes par or better on the next hole 33.3% of the time, the best recovery in the group. His normal par-or-better rate is 32.7%, so he neither collapses nor rebounds. That fits a man who never looks either too pleased or too upset.
| Player | Birdie % of holes | Bogey % | Double or worse % |
|---|---|---|---|
| David MULLIN | 3.8 | 39.6 | 27.5 |
| Jon BAKER | 3.7 | 37.6 | 30.9 |
| Gregg WILLIAMS | 2.2 | 37.2 | 39.0 |
| Stuart NEUMANN | 2.0 | 34.3 | 42.1 |
See the Honours board for the full list of winners.
How I worked this out: I took "Dave" to mean David MULLIN, the only Dave in the player list. Honours counts come from the official winners list, which only includes completed TEGs. Hole rates use every hole he has played (1,206 holes), measured gross against par. The bounce-back figure uses 763 bad holes, counting a bogey or worse as bad and par or better on the next hole as a bounce-back. The "why" is a light-hearted reading of the numbers, because golf data can't show what he feels.
Show the workings (3 steps)
Code run in the sandbox
cd $INPUT_DIR; python3 -c "
import pandas as pd
h=pd.read_csv('holes.csv')
g=h.groupby('Player').agg(n=('Sc','size'),eag=('GrossVP',lambda x:(x<=-2).mean()*100),bird=('GrossVP',lambda x:(x==-1).mean()*100),par=('GrossVP',lambda x:(x==0).mean()*100),bog=('GrossVP',lambda x:(x==1).mean()*100),dbl=('GrossVP',lambda x:(x>=2).mean()*100),tbp=('GrossVP',lambda x:(x>=3).mean()*100),avg=('GrossVP','mean'))
print(g.round(1))
"
→ n eag bird par bog dbl tbp avg Player Alex BAKER 864 0.0 1.4 16.1 32.1 50.5 24.8 1.7 David MULLIN 1206 0.1 3.8 28.9 39.6 27.5 9.6 1.0 Gregg WILLIAMS 1206 0.0 2.2 21.6 37.2 39.0 15.5 1.4 Henry MELLER 486 0.0 1.9 21.6 38.1 38.5 17.7 1.4 John PATTERSON 576 0.2 2.1 22.7 31.6 43.4 21.2 1.5 Jon BAKER 1206 0.1 3.7 27.6 37.6 30.9 12.2 1.1 Stuart NEUMANN 846 0.1 2.0 21.5 34.3 42.1 19.1 1.5
Lookup: get_honours
{}
→ {
"winners_by_teg": [
{
"TEG": "TEG 2",
"Year": 2009,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 3",
"Year": 2010,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 4",
"Year": 2011,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 5",
"Year": 2012,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Stuart NEUMANN*",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 6",
"Year": 2013,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 7",
"Year": 2014,
"TEG Trophy": "Henry MELLER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 8",
"Year": 2015,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 9",
"Year": 2016,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 10",
"Year": 2017,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "John PATTERSON"
},
{
"TEG": "TEG 11",
"Year": 2018,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 12",
"Year": 2019,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 13",
"Year": 2020,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 14",
"Year": 2021,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Jon BAKER"
},
{
"TEG": "TEG 15",
"Year": 2022,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 16",
"Year": 2023,
"TEG Trophy": "Stuart NEUMANN",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 17",
"Year": 2024,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 18",
"Year": 2025,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Jon BAKER"
}
],
"counts": {
"TEG Trophy": {
"Jon BAKER": 4,
"Gregg WILLIAMS": 4,
"David MULLIN": 3,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1,
"Stuart NEUMANN": 1
},
"Green Jacket": {
"David MULLIN": 9,
"Jon BAKER": 4,
"Gregg WILLIAMS": 3,
"Stuart NEUMANN": 1
},
"HMM Wooden Spoon": {
"David MULLIN": 4,
"Henry MELLER": 3,
"Alex BAKER": 3,
"Gregg WILLIAMS": 2,
"Stuart NEUMANN": 2,
"Jon BAKER": 2,
"John PATTERSON": 1
},
"Trophy + Jacket combined": {
"David MULLIN": 12,
"Jon BAKER": 8,
"Gregg WILLIAMS": 7,
"Stuart NEUMANN": 2,
"John PATTERSON": 2,
"Alex BAKER": 2,
"Henry MELLER": 1
}
},
"page": "/honours",
"notes": [
"Only completed TEGs are counted.",
"TEG Trophy = net competition (net vs par up to TEG 7, Stableford points from TEG 8).",
"Green Jacket = best gross. Wooden Spoon = last in the net competition.",
"TEG 5 Green Jacket was awarded for best Stableford round; David MULLIN had the best gross score."
]
}
Lookup: get_bounce_back
{
"trigger": "bogey_or_worse",
"basis": "gross"
}
→ {
"result": [
{
"Player": "David MULLIN",
"Triggers": 763,
"BounceBacks": 254,
"BaselineParOrBetterRate": 32.7,
"BounceBackRate": 33.3,
"Difference": 0.6
},
{
"Player": "Jon BAKER",
"Triggers": 771,
"BounceBacks": 238,
"BaselineParOrBetterRate": 31.6,
"BounceBackRate": 30.9,
"Difference": -0.7
},
{
"Player": "John PATTERSON",
"Triggers": 407,
"BounceBacks": 110,
"BaselineParOrBetterRate": 25.4,
"BounceBackRate": 27.0,
"Difference": 1.6
},
{
"Player": "Stuart NEUMANN",
"Triggers": 607,
"BounceBacks": 149,
"BaselineParOrBetterRate": 23.8,
"BounceBackRate": 24.5,
"Difference": 0.7
},
{
"Player": "Gregg WILLIAMS",
"Triggers": 865,
"BounceBacks": 201,
"BaselineParOrBetterRate": 24.5,
"BounceBackRate": 23.2,
"Difference": -1.3
},
{
"Player": "Henry MELLER",
"Triggers": 347,
"BounceBacks": 78,
"BaselineParOrBetterRate": 24.0,
"BounceBackRate": 22.5,
"Difference": -1.5
},
{
"Player": "Alex BAKER",
"Triggers": 668,
"BounceBacks": 117,
"BaselineParOrBetterRate": 17.6,
"BounceBackRate": 17.5,
"Difference": -0.1
}
],
"definition": "Trigger: a hole at bogey or worse (gross vs par). Bounce-back: the next hole in the same round at par or better (gross). Rate = bounce-backs / triggers, as a %. Baseline = the player's par-or-better rate on all holes 2-18, for comparison; Difference = rate minus baseline in percentage points.",
"assumptions": [
"Hole 18 is never a trigger, because the next hole is in a different round.",
"Holes are taken in numbered order (1-18) within a round.",
"Includes any TEG currently in progress.",
"Small samples (under ~50 triggers) are noisy."
],
"page": null
}
Who has won the most Green Jackets?
David MULLIN has won the most Green Jackets, with 9. Next are Jon BAKER (4), Gregg WILLIAMS (3) and Stuart NEUMANN (1).
Mullin won seven of them in a row from TEG 6 to TEG 10, which is a fairly dominant run. See the Honours board for the full list.
How I worked this out: I used the official honours list, which counts completed TEGs only. It includes the manual override for TEG 5, where the Green Jacket went to Stuart NEUMANN for best Stableford round, even though Mullin had the best gross score. Without that override, Mullin would have 10.
Show the workings (1 step)
Lookup: get_honours
{
"competition": "jacket"
}
→ {
"winners_by_teg": [
{
"TEG": "TEG 2",
"Year": 2009,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 3",
"Year": 2010,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 4",
"Year": 2011,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 5",
"Year": 2012,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Stuart NEUMANN*",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 6",
"Year": 2013,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Henry MELLER"
},
{
"TEG": "TEG 7",
"Year": 2014,
"TEG Trophy": "Henry MELLER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 8",
"Year": 2015,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 9",
"Year": 2016,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Stuart NEUMANN"
},
{
"TEG": "TEG 10",
"Year": 2017,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "John PATTERSON"
},
{
"TEG": "TEG 11",
"Year": 2018,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 12",
"Year": 2019,
"TEG Trophy": "John PATTERSON",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 13",
"Year": 2020,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "David MULLIN"
},
{
"TEG": "TEG 14",
"Year": 2021,
"TEG Trophy": "David MULLIN",
"Green Jacket": "David MULLIN",
"HMM Wooden Spoon": "Jon BAKER"
},
{
"TEG": "TEG 15",
"Year": 2022,
"TEG Trophy": "Gregg WILLIAMS",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 16",
"Year": 2023,
"TEG Trophy": "Stuart NEUMANN",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Alex BAKER"
},
{
"TEG": "TEG 17",
"Year": 2024,
"TEG Trophy": "Jon BAKER",
"Green Jacket": "Jon BAKER",
"HMM Wooden Spoon": "Gregg WILLIAMS"
},
{
"TEG": "TEG 18",
"Year": 2025,
"TEG Trophy": "Alex BAKER",
"Green Jacket": "Gregg WILLIAMS",
"HMM Wooden Spoon": "Jon BAKER"
}
],
"counts": {
"Green Jacket": {
"David MULLIN": 9,
"Jon BAKER": 4,
"Gregg WILLIAMS": 3,
"Stuart NEUMANN": 1
}
},
"page": "/honours?tab=jacket",
"notes": [
"Only completed TEGs are counted.",
"TEG Trophy = net competition (net vs par up to TEG 7, Stableford points from TEG 8).",
"Green Jacket = best gross. Wooden Spoon = last in the net competition.",
"TEG 5 Green Jacket was awarded for best Stableford round; David MULLIN had the best gross score."
]
}