🤖 AI benchmark: hit-rate of 7 models
Seven external AI models (Hermes contour) independently analyze the same partite — predicting the esito (1X2), total (Over/Under), both squadre to punteggio (BTTS) and the exact punteggio. Here we honestly compare their pronostici against the real result after the final whistle and combine everything into a single accuracy rating. An informational and analytical snapshot, not betting advice.
Leaderboard · Hockey
| Model | N (settled) | 1X2 | Doppia chance (1X) | Total gol | BTTS | Exact punteggio | Composite accuracy |
|---|---|---|---|---|---|---|---|
 Google AI
|
13 | 69.2%(9/13) | 69.2%(9/13) | 46.2%(6/13) | 76.9%(10/13) | 0.0%(0/13) | 48.1%(25/52) |
 Claude
|
12 | 50.0%(6/12) | 50.0%(6/12) | 58.3%(7/12) | 75.0%(9/12) | 0.0%(0/12) | 45.8%(22/48) |
 DeepSeek
|
0 | — | — | — | — | — | — |
 ChatGPT
|
0 | — | — | — | — | — | — |
 Qwen
|
0 | — | — | — | — | — | — |
 Kimi
|
0 | — | — | — | — | — | — |
 GLM 5.2
|
0 | — | — | — | — | — | — |
grey — sample <5, not representative; «—» — the modello has not made a settled pronostico yet.
Doppia chance (1X) — the same pronostico counts as a vittoria if the chosen side won or the partita drew. Of the 1X2 losses in football/hockey: 0 draws, 10 underdog (total settled 1X2 pronostici in these sports: 25, double chance combined 60.0% (15/25)). The models almost always take the favorite and don't bet on a pareggio — double chance shows how many bets are eaten specifically by draws.
Composite accuracy — the share of correct pronostici su all mostrato mercati together: (sum of correct pronostici) ÷ (sum of all settled pronostici) su the mercati 1X2 + Total gol + BTTS + Exact punteggio. Each mercato-pronostico weighs equally. This is hit-rate, not profitability — for money/ROI by modello see /ai-agent. Total: a push (punteggio exactly sulla quota) is excluded from the denominator. BTTS is checked against whether both squadre scored. «Exact punteggio» — the full final punteggio was guessed correctly (H and A matched); pronostici with no recognized punteggio do not count toward the denominator. Doppia chance (1X): a pronostico counts as a vittoria if the chosen side won OR it was a pareggio — it accounts for frequent draws that «eat» bets on the favorite. This metric is informational and is not included in composite accuracy.
Composite modello rating · all mercati · Hockey
Bar height = the modello's composite accuracy su all applicable mercati on the current sample. Sorted from best to worst.
Bars are AI models by version; grey/dimmed — sample <5, not representative. The snapshot is informational, not betting advice.
Accuracy by mercato · Hockey
Where each modello is strong: one mini-bar per applicable mercato, with the percentage and (hits/sample).
Google AI
Composite 48.1%
Claude
Composite 45.8%
DeepSeek
Composite —
ChatGPT
Composite —
Qwen
Composite —
Kimi
Composite —
GLM 5.2
Composite —
The modello's favorite by 1X2 = the max of P1/X/P2 in its probabilities; for sports without a pareggio (tennis, volleyball, etc.) the «X» option doesn't participate. grey — sample <5, not representative. Not betting advice.