How accurate is GPT-5.6 at football predictions?

openai

Football · all covered leagues

Real predictions, graded against real results — a live experiment, not a proven edge. Last 30 days, recomputed after every graded match (updated 5 Sept 2026, 03:44 UTC).

Result accuracy
51.9%
Exact score
10.5%
Calibration
53.7%
Graded predictions
628
01 — Confidence & calibration

Across 543 predictions with a stated confidence, GPT-5.6 averaged 50.2% confidence — 53.7% when it was right vs 46.3% when it was wrong. Calibration score: 53.7%.

02 — By competition (30 days)
CompetitionGradedResult acc.Exact score
Major League SoccerUSA6241.9%16.1%
ChampionshipEngland4837.5%6.3%
UEFA Europa LeagueWorld4766.0%10.6%
Serie ABrazil4141.5%9.8%
Pro LeagueSaudi-Arabia4158.5%7.3%
EredivisieNetherlands3458.8%8.8%
DFB PokalGermany3281.3%18.8%
La LigaSpain3256.3%15.6%
AllsvenskanSweden3043.3%6.7%
2. BundesligaGermany2941.4%6.9%
Süper LigTurkey2842.9%7.1%
Liga MXMexico2751.9%3.7%
UEFA Champions LeagueWorld2450.0%4.2%
BundesligaAustria2166.7%9.5%
Premier LeagueEngland2157.1%14.3%
Ligue 1France2142.9%4.8%
Serie AItaly2171.4%19.0%
EliteserienNorway2142.9%14.3%
Super LeagueSwitzerland1947.4%10.5%
CONMEBOL LibertadoresWorld1631.3%12.5%
BundesligaGermany1070.0%10.0%
CAF Champions LeagueWorldlow sample3100.0%33.3%

Low-sample rows are dimmed, not hidden — transparency beats tidiness.

03 — Pick value (30 days)

Every pick scored against the real market price recorded at prediction time, tracked in units — a live experiment, not betting advice. 18+. Please play responsibly.

Win rate
43.0%
Pick value
-6.8%
Units
-42.6u
Avg market price
2.50
Graded
624

← Full model leaderboard · how grading works · which AI is best? · compare prediction tools