Track record
Serie A track record
Every Serie A prediction timestamped before kick-off and settled against the final score. Nothing is removed afterwards.
Since August 22, 2026 the model’s 1X2 pick came in 31 of 53 settled Serie A matches (58.5%), with a Brier score of 0.554.
Over/under 2.5 goals landed 56.6% of the time (53 matches) and both-teams-to-score 52.8% (53).
One unit on every 1X2 pick at the closing market price would have returned -24.3% over 18 matches.
By month: Aug 2026 75% (20) · Sep 2026 50% (24) · Oct 2026 44% (9).
Window Aug 22 – Oct 17 · How the numbers are defined · Brier score explained
- Settled
- 53
- 1X2 picks with a final score
- Hit rate
- 58.5%
- 31 of 53 picks
- Brier
- 0.554
- 1X2, 3-outcome; lower is better, 0.667 = uniform guess
- Return · Closing price
- -24.3%
- flat stake, closing odds, 18 bets
By market
| Market | Hit rate | W–L | Brier |
|---|---|---|---|
| 1X2 | 58.5% | 31–22 | 0.554 |
| Over/Under 2.5 | 56.6% | 30–23 | 0.253 |
| Both teams to score | 52.8% | 28–25 | 0.250 |
Over/under 2.5 and both-teams-to-score are scored on the raw model probability and settled from the final score. Their Brier is binary (0.25 = coin flip).
Monthly hit rate
| Month | n | Hit rate | W–L |
|---|---|---|---|
| Aug 2026 | 20 | 75.0% | 15–5 |
| Sep 2026 | 24 | 50.0% | 12–12 |
| Oct 2026 | 9 | 44.4% | 4–5 |
Small months swing wildly; read them with the sample size in view.
Latest settled Serie A matches
Score and the model’s pre-match pick, newest first. Open a match for the full overview.
| Date | Match | Score | Pick | Result |
|---|---|---|---|---|
| Torino – Udinese | 2–1 | Udinese · 42% | Lost | |
| Cagliari – Juventus | 1–1 | Juventus · 51% | Lost | |
| Sassuolo – Milan | 1–1 | Milan · 46% | Lost | |
| Lecce – Bologna | 1–2 | Bologna · 45% | Won | |
| Como – Roma | 0–0 | Roma · 37% | Lost | |
| Inter – Parma | 3–0 | Inter · 76% | Won | |
| Genoa – Fiorentina | 2–1 | Genoa · 36% | Won | |
| Milan – Lecce | 3–0 | Milan · 56% | Won | |
| Juventus – Atalanta | 2–0 | Juventus · 42% | Won | |
| Parma – Genoa | 2–1 | Draw · 37% | Lost | |
| Fiorentina – Napoli | 1–1 | Napoli · 59% | Lost | |
| Roma – Inter | 2–2 | Roma · 42% | Lost | |
| Bologna – Torino | 1–1 | Bologna · 45% | Lost | |
| Udinese – Cagliari | 0–1 | Udinese · 40% | Lost | |
| Inter – Udinese | 5–3 | Inter · 67% | Won | |
| Torino – Roma | 0–2 | Roma · 73% | Won | |
| Como – Parma | 2–1 | Como · 70% | Won | |
| Sassuolo – Juventus | 3–2 | Juventus · 55% | Lost | |
| Napoli – Bologna | 1–0 | Napoli · 52% | Won | |
| Atalanta – Cagliari | 1–2 | Atalanta · 51% | Lost |
Frequently asked
- How accurate are the Serie A predictions?
- Since August 22, 2026 the model’s 1X2 pick came in 31 of 53 settled Serie A matches (58.5%), with a Brier score of 0.554.
- What is a Brier score?
- The Brier score is the mean squared error of the published probabilities against what happened: 0 is perfect, 0.667 is a uniform guess across three outcomes. Lower is better.
- How is the Serie A record computed?
- Every Serie A prediction is stored with a timestamp before kick-off and settled automatically from the final score. The hit rate counts the 1X2 pick, the Brier score the full probability split, and the return assumes one unit on each pick at the last captured market price.