What the numbers say

The rule of thumb gets the outcome right in 46% of matches. No model goes beyond that; the best (Anthropics beste KI) reaches 45%.

Season 2025/26 · after matchday 34

The AI table

Models, panel and rule of thumb scored by the same system. Tap a row to highlight it everywhere on this page.

Pos.	Model	Picks	Exact	Outcome	Avg points	Index
1	gpt-4o	92	10 · 11%	40%	-0.28	37.4after matchday 34
2	grok-3	83	8 · 10%	42%	-0.37	36.6after matchday 34
3	claude-sonnet-4-5-20250929	71	11 · 15%	45%	+0.06	36after matchday 32
4	gemini-3.1-flash-lite-preview	63	9 · 14%	44%	-0.17	34.3after matchday 34
–	Table favourite wins 1:0	90	7 · 8%	46%	+0.45	–no index

Form = average points per matchday, most recent on the right.strong (+1.5 and up)positivenegativeweak (below −1.5)no pick

The rule of thumb always picks 1:0 for the table favourite. It is the zero line and gets no index, because it states neither confidence nor reasons.

Profiles

Every participant checked against the same questions, so they compare directly.

1.

37Index

stable

after matchday 34

Points form30 · 50 %
Calibration68 · 25 %
Reasoning22 · 25 %

Gets the outcome right in 40% of matches and is overconfident doing so (states 69% confidence on average). Backs the table favourite in 72% of matches. No underdog pick came off.

Outcome right40% · 37/92

Exact scores10 · 11%

Wrong outcomes41%

Stated confidence69%

Bar = actually right (40%), tick = stated confidence (69%).

Reasons confirmed23%

At least one factor could be checked in 71% of predictions.

Most cited reasons

Heimvorteil
cited 28× · 68% confirmed
Starke Form von RB Leipzig
cited 3× · not checkable
Starke Form von Dortmund
cited 3× · not checkable

2.

37Index

stable

after matchday 34

Points form29 · 50 %
Calibration73 · 25 %
Reasoning15 · 25 %

Gets the outcome right in 42% of matches and is overconfident doing so (states 69% confidence on average). Backs the table favourite in 83% of matches. No underdog pick came off.

Outcome right42% · 35/83

Exact scores8 · 10%

Wrong outcomes41%

Stated confidence69%

Bar = actually right (42%), tick = stated confidence (69%).

Reasons confirmed14%

At least one factor could be checked in 60% of predictions.

Most cited reasons

Heimvorteil
cited 5× · 80% confirmed
Schwache Form von St. Pauli
cited 3× · not checkable
Schwache Defensiven beider Teams
cited 3× · not checkable

3.

36Index

stable

after matchday 32

Points form30 · 50 %
Calibration66 · 25 %
Reasoning18 · 25 %

Gets the outcome right in 45% of matches and is overconfident doing so (states 62% confidence on average). Backs the table favourite in 63% of matches. No underdog pick came off.

Outcome right45% · 32/71

Exact scores11 · 15%

Wrong outcomes39%

Stated confidence62%

Bar = actually right (45%), tick = stated confidence (62%).

Reasons confirmed21%

At least one factor could be checked in 76% of predictions.

Most cited reasons

Starker Heimvorteil im Signal Iduna Park
cited 2× · 100% confirmed
Heimvorteil Frankfurt
cited 2× · 50% confirmed
Enormer Qualitätsunterschied im Kader
cited 2× · not checkable

4.

34Index

stable

after matchday 34

Points form28 · 50 %
Calibration69 · 25 %
Reasoning12 · 25 %

Gets the outcome right in 44% of matches and is overconfident doing so (states 70% confidence on average). Backs the table favourite in 63% of matches. No underdog pick came off.

Outcome right44% · 28/63

Exact scores9 · 14%

Wrong outcomes44%

Stated confidence70%

Bar = actually right (44%), tick = stated confidence (70%).

Reasons confirmed16%

At least one factor could be checked in 57% of predictions.

Most cited reasons

Heimvorteil im Signal Iduna Park
cited 3× · 67% confirmed
Ausgeglichene Tabellensituation
cited 3× · not checkable
Klassenunterschied in der Tabelle
cited 3× · not checkable

Season view

Four angles on the same picks.

Index over time

The index (0–100) weights the last 3–8 matchdays, recent ones count more. Higher is better.

Hit profile

Share of each rung of the scoring ladder across all scored picks of a model.

OpenAI
11%
10%
14%
18%
38%
92
xAI
10%
12%
12%
17%
36%
83
Anthropic
15%
10%
15%
15%
39%
71
Google
14%
16%
11%
43%
63
Rule
13%
10%
14%
17%
30%
90

Exact score+5
Goal-difference hit+4
Partial hit+3
Outcome hit+2
Near miss-1
Miss-2
Bad miss-3

Self-assessment

How often was a model right when it said it was X% sure? Above the line: more cautious than needed. Below: overconfident.

↔ stated confidence · ↕ actually right perfectly calibrated

Easy or hard-earned?

Hit rate when backing the table favourite – and when going against it (underdog or draw).

OpenAIs beste KI

Favourite picked42% · 66 picks

Against the favourite35% · 26 picks

xAI Grok

Favourite picked43% · 69 picks

Against the favourite36% · 14 picks

Anthropics beste KI

Favourite picked42% · 45 picks

Against the favourite50% · 26 picks

Googles beste KI

Favourite picked48% · 40 picks

Against the favourite39% · 23 picks

Rule of thumb

Favourite picked47% · 81 picks

Against the favourite33% · 9 picks

Matchday log

Every match, every pick, every score – ready to check. Tap a pick for the breakdown.

Day ranking
1.OpenAIs beste KIØ -1.11 · 3/9
2.Googles beste KIØ -1.60 · 2/9
3.xAI GrokØ -1.73 · 2/9

Match

OpenAI

xAI

Anthropic

Google

Frankfurt
2:2Sat, 05/16, 01:30 PM
Stuttgart
no pick
Frankfurt
2:2
Stuttgart
OpenAI
xAI
Anthropic
no pick
Google
Bayern
5:1Sat, 05/16, 01:30 PM
Köln
no pick
Bayern
5:1
Köln
OpenAI
xAI
Anthropic
no pick
Google
Heidenheim
0:2Sat, 05/16, 01:30 PM
Mainz
no pick
Heidenheim
0:2
Mainz
OpenAI
xAI
Anthropic
no pick
Google
Leverkusen
1:1Sat, 05/16, 01:30 PM
HSV
no pick
Leverkusen
1:1
HSV
OpenAI
xAI
Anthropic
no pick
Google
Union Berlin
4:0Sat, 05/16, 01:30 PM
Augsburg
no pick
Union Berlin
4:0
Augsburg
OpenAI
xAI
Anthropic
no pick
Google
Bremen
0:2Sat, 05/16, 01:30 PM
Dortmund
no pick
Bremen
0:2
Dortmund
OpenAI
xAI
Anthropic
no pick
Google
Freiburg
4:1Sat, 05/16, 01:30 PM
Leipzig
no pick
Freiburg
4:1
Leipzig
OpenAI
xAI
Anthropic
no pick
Google
Gladbach
4:0Sat, 05/16, 01:30 PM
Hoffenheim
no pick
Gladbach
4:0
Hoffenheim
OpenAI
xAI
Anthropic
no pick
Google
St. Pauli
1:3Sat, 05/16, 01:30 PM
Wolfsburg
no pick
St. Pauli
1:3
Wolfsburg
OpenAI
xAI
Anthropic
no pick
Google