The model has called 63% of 338 2023/24 games right · always picking the home side: 59%
No scheduled games left for this selection. The sections below cover the season's completed games.
Predicted vs actual338 finished games
Predicted vs actual 338
212 hit · 126 missed · 17 exact| 19 Apr 2024, 19:30 |  KACvs RBS | 66% | 3.0 – 2.1 | 2:2 | 2:6 | Miss | +3.0 |
|---|---|---|---|---|---|---|---|
| 16 Apr 2024, 19:30 | RBSvsKAC | 50% | 2.5 – 2.5 | 2:2 | 1:4 | Miss | +0.0 |
| 14 Apr 2024, 17:30 | KACvsRBS | 65% | 3.0 – 2.1 | 2:2 | 2:3 OT | Miss | -0.1 |
| 12 Apr 2024, 19:30 | RBSvsKAC | 49% | 2.4 – 2.4 | 2:2 | 6:5 OT | Miss | +6.2 |
| 09 Apr 2024, 19:30 | KACvsRBS | 65% | 2.9 – 2.0 | 2:2 | 4:2 | Hit | +1.1 |
| 07 Apr 2024, 17:30 | RBSvsKAC | 52% | 2.4 – 2.3 | 2:2 | 2:5 | Miss | +2.2 |
| 05 Apr 2024, 19:30 | KACvsRBS | 64% | 2.9 – 2.1 | 2:2 | 0:1 OT | Miss | -3.9 |
| 02 Apr 2024, 19:30 | RBSvs HCB | 66% | 2.9 – 2.0 | 2:1 | 3:2 OT | Hit | +0.2 |
| 30 Mar 2024, 18:00 | HCBvsRBS | 51% | 2.4 – 2.4 | 2:2 | 1:4 | Miss | +0.3 |
| 28 Mar 2024, 19:30 | RBSvsHCB | 65% | 2.9 – 2.0 | 2:1 | 2:3 | Miss | +0.2 |
| 26 Mar 2024, 19:45 | HCBvsRBS | 50% | 2.4 – 2.4 | 2:2 | 3:2 | Miss | +0.2 |
| 26 Mar 2024, 19:30 |  HCPvsKAC | 35% | 2.3 – 3.3 | 2:3 | 3:5 | Hit | +2.4 |
| 24 Mar 2024, 18:30 | RBSvsHCB | 65% | 2.8 – 1.9 | 2:1 | 4:3 | Hit | +2.2 |
| 24 Mar 2024, 17:30 | KACvsHCP | 79% | 3.9 – 2.0 | 3:1 | 3:1 | Hit★ | -1.9 |
| 22 Mar 2024, 19:45 | HCBvsRBS | 51% | 2.4 – 2.3 | 2:2 | 1:3 | Miss | -0.7 |
| 22 Mar 2024, 19:30 | HCPvsKAC | 36% | 2.4 – 3.2 | 2:3 | 2:4 | Hit | +0.4 |
| 19 Mar 2024, 19:30 | KACvsHCP | 76% | 3.8 – 2.0 | 3:2 | 7:1 | Hit | +2.2 |
| 19 Mar 2024, 19:15 | RBSvsHCB | 66% | 2.9 – 2.0 | 2:1 | 0:2 | Miss | -2.9 |
| 16 Mar 2024, 17:30 |  AVSvsHCP | 65% | 3.4 – 2.5 | 3:2 | 1:4 | Miss | -0.9 |
| 13 Mar 2024, 19:45 | HCPvsAVS | 52% | 3.0 – 2.9 | 2:2 | 4:2 | Hit | +0.2 |
| 13 Mar 2024, 19:30 |  PIVvsKAC | 37% | 2.5 – 3.4 | 2:3 | 1:5 | Hit | +0.1 |
| 11 Mar 2024, 19:45 | HCBvs VSV | 64% | 3.3 – 2.4 | 3:2 | 2:1 | Hit | -2.8 |
| 11 Mar 2024, 19:30 | RBSvs BWL | 67% | 3.0 – 2.0 | 3:1 | 5:1 | Hit | +1.0 |
| 11 Mar 2024, 19:15 | AVSvsHCP | 65% | 3.4 – 2.5 | 3:2 | 5:3 | Hit | +2.2 |
| 11 Mar 2024, 19:15 | KACvsPIV | 81% | 4.2 – 2.0 | 4:2 | 1:6 | Miss | +0.8 |
| 09 Mar 2024, 19:30 | VSVvsHCB | 50% | 2.8 – 2.8 | 2:2 | 6:1 | Hit | +1.4 |
| 09 Mar 2024, 19:30 | PIVvsKAC | 35% | 2.5 – 3.5 | 2:3 | 1:3 | Hit | -2.0 |
| 09 Mar 2024, 18:00 | BWLvsRBS | 49% | 2.4 – 2.5 | 2:2 | 2:3 | Hit | +0.1 |
| 08 Mar 2024, 19:45 | HCPvsAVS | 52% | 3.0 – 2.8 | 2:2 | 2:1 | Hit | -2.8 |
| 07 Mar 2024, 19:45 | HCPvsAVS | 52% | 3.0 – 2.9 | 2:2 | 4:3 OT | Hit | +1.2 |
| 07 Mar 2024, 19:30 | HCBvsVSV | 64% | 3.3 – 2.4 | 3:2 | 6:1 | Hit | +1.3 |
| 07 Mar 2024, 19:15 | RBSvsBWL | 65% | 2.9 – 2.1 | 2:2 | 5:1 | Hit | +1.0 |
| 07 Mar 2024, 19:15 | KACvsPIV | 77% | 4.0 – 2.1 | 4:2 | 7:0 | Hit | +0.8 |
| 05 Mar 2024, 19:30 | PIVvsKAC | 38% | 2.6 – 3.4 | 2:3 | 0:1 | Hit | -5.1 |
| 05 Mar 2024, 19:15 | BWLvsRBS | 50% | 2.4 – 2.5 | 2:2 | 3:2 | Miss | +0.1 |
| 05 Mar 2024, 19:15 | VSVvsHCB | 53% | 2.9 – 2.7 | 2:2 | 2:4 | Miss | +0.4 |
| 04 Mar 2024, 19:15 | AVSvsHCP | 65% | 3.4 – 2.5 | 3:2 | 3:2 | Hit★ | -0.9 |
| 03 Mar 2024, 18:00 | HCBvsVSV | 61% | 3.1 – 2.4 | 3:2 | 5:2 | Hit | +1.4 |
| 03 Mar 2024, 17:30 | RBSvsBWL | 63% | 2.9 – 2.1 | 2:2 | 4:0 | Hit | -1.0 |
| 03 Mar 2024, 17:30 | AVSvsHCP | 64% | 3.4 – 2.5 | 3:2 | 3:2 OT | Hit★ | -0.9 |
| 03 Mar 2024, 17:30 | KACvsPIV | 79% | 4.2 – 2.2 | 4:2 | 0:2 | Miss | -4.5 |
| 29 Feb 2024, 19:30 | HCPvs OLL | 62% | 3.5 – 2.8 | 3:2 | 4:3 OT | Hit | +0.8 |
| 27 Feb 2024, 19:30 | PIVvs HCI | 59% | 3.4 – 2.8 | 3:2 | 5:2 | Hit | +0.8 |
| 27 Feb 2024, 19:15 | OLLvsHCP | 53% | 3.2 – 3.0 | 3:2 | 6:2 | Hit | +1.9 |
| 25 Feb 2024, 18:00 | HCPvsOLL | 62% | 3.5 – 2.8 | 3:2 | 3:0 | Hit | -3.3 |
| 25 Feb 2024, 17:30 | HCIvsPIV | 59% | 3.3 – 2.7 | 3:2 | 3:6 | Miss | +2.9 |
| 23 Feb 2024, 19:15 | HCPvsRBS | 44% | 2.4 – 2.8 | 2:2 | 3:2 | Miss | -0.2 |
| 23 Feb 2024, 19:15 | HCBvsKAC | 42% | 2.4 – 2.9 | 2:2 | 3:0 | Miss | -2.3 |
| 23 Feb 2024, 19:15 | HCIvs ASH | 78% | 4.2 – 2.2 | 4:2 | 5:0 | Hit | -1.4 |
| 23 Feb 2024, 19:15 | BWLvsOLL | 68% | 3.5 – 2.4 | 3:2 | 6:5 | Hit | +5.1 |
Showing 50 of 338 rows. Sort or filter to narrow the list.
Each row's prediction was built from the games played before that date only — nothing after it. Sort by Goals Δ or Call to find where the model went wrong. A ★ marks a game whose exact scoreline it also got right.
Biggest upsets
- RBSvs
VIC27 Oct 2023, 19:15 · predicted 90% home
1:2 OTaway won - PIVvs
G9917 Nov 2023, 19:30 · predicted 86% home
1:2 OTaway won - BWLvsG99
30 Dec 2023, 18:00 · predicted 85% home
1:4away won - BWLvsASH
12 Jan 2024, 19:15 · predicted 84% home
3:4 OTaway won - G99vsHCP
01 Nov 2023, 17:30 · predicted 16% home
3:2 OThome won
Games where the model was most confident in the side that lost.
Best calls
- RBSvsASH
20 Oct 2023, 19:15 · predicted 95% home
4:2as predicted - G99vsRBS
29 Oct 2023, 17:30 · predicted 9% home
1:2as predicted - VSVvsASH
26 Sept 2023, 19:15 · predicted 90% home
2:1 SOas predicted - KACvsG99
27 Oct 2023, 19:15 · predicted 89% home
4:1as predicted - VSVvsG99
03 Nov 2023, 19:15 · predicted 89% home
3:2as predicted
Confident calls that came in. Note these are cheap to collect — a good model should have many of them, which is why the upsets matter more.
Accuracy and calibrationBrier 0.227
How well does this model do?
338 games scored- Home-only baseline
- Model
Table view
| Games scored | Home-only baseline | Model |
|---|---|---|
| 12 | 58.3% | 66.7% |
| 24 | 58.3% | 66.7% |
| 36 | 63.9% | 58.3% |
| 48 | 58.3% | 58.3% |
| 60 | 58.3% | 63.3% |
| 72 | 55.6% | 61.1% |
| 84 | 57.1% | 60.7% |
| 96 | 58.3% | 60.4% |
| 108 | 59.3% | 60.2% |
| 120 | 62.5% | 60.8% |
| 132 | 62.1% | 59.8% |
| 144 | 61.8% | 59.0% |
| 156 | 60.9% | 58.3% |
| 168 | 58.9% | 57.7% |
| 180 | 58.9% | 57.8% |
| 192 | 58.3% | 57.8% |
| 204 | 58.8% | 58.3% |
| 216 | 57.4% | 57.9% |
| 228 | 57.9% | 58.3% |
| 240 | 57.9% | 59.2% |
| 252 | 57.9% | 59.9% |
| 264 | 58.3% | 60.6% |
| 276 | 58.3% | 61.6% |
| 288 | 58.3% | 62.8% |
| 300 | 59.3% | 63.0% |
| 312 | 59.6% | 63.8% |
| 324 | 59.3% | 63.9% |
| 336 | 58.9% | 63.1% |
| 338 | 58.6% | 62.7% |
Both lines are cumulative, so they settle as the sample grows. The gap between them is the model's edge over the naive pick.
- Predicted
- Actual
Table view
| Predicted home win % | Predicted | Actual | Games |
|---|---|---|---|
| 10–20 | 18% | 86% | 7 |
| 20–30 | 25% | 32% | 22 |
| 30–40 | 35% | 28% | 40 |
| 40–50 | 46% | 58% | 50 |
| 50–60 | 55% | 55% | 67 |
| 60–70 | 65% | 67% | 70 |
| 70–80 | 75% | 72% | 54 |
| 80–90 | 85% | 80% | 25 |
The two lines tracking each other means the probabilities can be taken at face value. Where "actual" sits above "predicted", the model is under-confident in the home side; below, it is over-confident. Check the game count before reading much into a bucket — the extremes hold far fewer games than the middle. Buckets with fewer than five games are left out.
Every game is predicted using only the games played before it — a walk-forward test, so these are the numbers the model would really have produced. The baseline is "always back the home side", priced at the home-win rate known at the time. Team strengths start from the 2022/23 final table.
How the prediction works
- Each team gets an attack and a defence multiplier from the goals it has scored and conceded, measured against the league average.
- A fixture's expected goals are the league's home (or away) scoring rate times the attacking team's attack times the defending team's defence — so home advantage comes out of fitting home and away separately, not from a bonus added on top.
- Both sides' goals are then treated as independent Poisson draws, which gives a probability for every scoreline rather than one guess, and from that the win, overtime and exact-score numbers.
- Rates are shrunk toward the prior with six games of weight, so an early 7-1 win does not make a team look unbeatable. This is why fixtures are marked low confidence until both teams have played a few games.
- Overtime and shootouts are treated as a coin flip. The feeds carry nothing that reliably predicts a shootout winner, and pretending otherwise would just be noise.
Goal models like this one capture scoring rates, not injuries, suspensions, travel, goalie rotation or roster changes. Treat the probabilities as a calibrated baseline, not a forecast.