NHL Prediction Model Performance & Calibration

See how our NHL prediction model performs: accuracy, Brier scores, calibration, and error metrics across thousands of games.

To understand each metric, read Understanding Performance Metrics . To apply this in practice, view today's NHL predictions, our NHL playoff odds, and in-game win probability charts.

How to Read These Metrics

Accuracy: The percentage of games where the predicted winner (team with >50% win probability) actually won. Simple but incomplete — it ignores how confident the model was.
Brier Score: Measures the mean squared error of probability predictions (0–1 scale, lower is better). A coin-flip baseline yields 0.25; our model targets values below 0.24. Brier score rewards well-calibrated confidence levels, not just picking the right side.
Log Loss: A logarithmic scoring rule that heavily penalises confident wrong predictions. Assigning 90% to a team that loses costs far more than assigning 55%. This keeps the model honest about uncertainty.
Calibration: Shows whether stated probabilities match real outcomes. In a well-calibrated model, games given a 70% win probability should be won about 70% of the time. The calibration tables below group predictions into decile bins so you can verify this directly.
RMSE (Root Mean Squared Error): For goal total predictions, RMSE measures how far predicted totals are from actual totals on average, in goal units.

Why calibration matters most: For probabilistic predictions, calibration is more important than raw accuracy. A model that says "55%" every game can be 55% accurate but useless for decision-making. A well-calibrated model tells you how much to trust each prediction. Learn more in our analytics guide and methodology.

Game Predictions (Multi-Window)

Window	Start	End	Games	Accuracy	Brier	Log Loss	Avg Winner Prob	RMSE Total
last 30	2026-03-17	2026-04-16	245	55.9%	0.2362	0.6632	53.2%	2.639
season to date	2025-10-01	2026-04-16	1312	56.2%	0.2439	0.6807	52.2%	2.588
multi season	2023-10-10	2026-04-16	3936	61.8%	0.2271	0.6449	53.9%	2.202

Totals (Over 5.5)

Window	Games	Accuracy	Brier	Log Loss	Avg Outcome Prob
last 30	245	50.6%	0.2693	0.7363	50.6%
season to date	1312	54.4%	0.2616	0.7214	51.5%
multi season	3936	62.9%	0.2257	0.6417	55.2%

Playoff Game Performance

Start	End	Games	Accuracy	Brier	Log Loss
2024-04-20	2026-05-24	263	63.1%	0.2396	0.6732

Current Matchups

MTL vs CAR model odds VGK vs COL model odds MTL vs CAR model odds

Daily Performance

Show:

Date	Games	Accuracy	Brier	Log Loss

Prediction Recap Highlights

Definition: High confidence means the model assigned the predicted team a win probability well above 50%. Edges that hit are the highest-confidence correct calls, misses are the highest-confidence incorrect calls, and surprise results show the largest absolute gap between win probability and the actual outcome.

Show:

Biggest Model Edges That Hit

Date	Matchup	Win Prob	Outcome

Biggest Misses (High Confidence)

Date	Matchup	Win Prob	Outcome

Surprise Results

Date	Matchup	Surprise	Outcome

Calibration (Win Prob Deciles) — Last 30

Bin	Count	Mean Pred	Observed
2	2	25.2%	0.0%
3	22	37.3%	31.8%
4	66	45.4%	50.0%
5	71	54.5%	43.7%
6	49	65.2%	57.1%
7	27	73.6%	77.8%
8	8	83.5%	87.5%

Calibration (Win Prob Deciles) — Season To Date

Bin	Count	Mean Pred	Observed
2	10	26.8%	30.0%
3	104	36.8%	42.3%
4	313	45.7%	44.7%
5	447	54.7%	49.7%
6	289	64.6%	59.2%
7	132	73.5%	69.7%
8	17	83.6%	76.5%

Calibration (Win Prob Deciles) — Multi Season

Bin	Count	Mean Pred	Observed
1	1	18.3%	0.0%
2	56	26.6%	7.1%
3	368	36.3%	31.8%
4	969	45.5%	44.0%
5	1272	54.9%	54.0%
6	835	64.4%	66.0%
7	368	73.7%	77.7%
8	67	83.3%	92.5%

Calibration (Over 5.5) — Last 30

Bin	Count	Mean Pred	Observed
1	2	16.8%	50.0%
2	7	26.5%	71.4%
3	49	35.4%	59.2%
4	31	44.9%	58.1%
5	78	55.1%	55.1%
6	39	63.9%	56.4%
7	29	75.5%	58.6%
8	8	84.3%	50.0%
9	2	91.3%	100.0%

Calibration (Over 5.5) — Season To Date

Bin	Count	Mean Pred	Observed
1	11	18.5%	72.7%
2	48	25.9%	47.9%
3	189	35.6%	52.9%
4	146	44.5%	58.2%
5	445	55.4%	58.2%
6	203	63.8%	55.2%
7	205	74.5%	60.0%
8	56	84.8%	60.7%
9	9	94.3%	88.9%

Calibration (Over 5.5) — Multi Season

Bin	Count	Mean Pred	Observed
1	34	18.2%	29.4%
2	157	26.0%	23.6%
3	584	35.7%	38.2%
4	459	44.4%	46.4%
5	1254	55.3%	56.5%
6	617	63.6%	62.7%
7	614	74.5%	73.1%
8	178	84.9%	80.3%
9	39	93.5%	97.4%

Team Calibration (Home, Top 15 by Volume)

Team	Count	Mean Pred	Observed	Bias
EDM	149	61.0%	61.7%	-0.7%
DAL	147	56.0%	62.6%	-6.6%
CAR	147	66.4%	68.7%	-2.3%
FLA	146	60.9%	62.3%	-1.4%
COL	141	63.7%	65.2%	-1.5%
VGK	141	58.5%	59.6%	-1.1%
WPG	133	56.4%	62.4%	-6.0%
BOS	133	50.9%	56.4%	-5.5%
TOR	133	53.5%	54.1%	-0.6%
TBL	132	58.4%	62.1%	-3.7%
MIN	132	54.5%	52.3%	+2.2%
LAK	131	58.9%	53.4%	+5.4%
WSH	131	54.8%	58.8%	-4.0%
NYR	131	53.7%	51.9%	+1.8%
MTL	131	48.7%	50.4%	-1.7%

Team Calibration (Pred vs Observed)

Mean Pred Observed

Starter Calibration (Home)

Window	Starter Status	Games	Accuracy	Brier	Log Loss
last 30	Starter	245	55.9%	0.2362	0.6632
season to date	Starter	1312	56.2%	0.2439	0.6807
multi season	Unknown	18	72.2%	0.2279	0.6480
multi season	Starter	3918	61.8%	0.2271	0.6449

Cross-Validation (Expanding Window)

Summary: 3 folds | Brier: 0.2495 | Log Loss: 0.6925 | RMSE Total: 2.394

Show fold details

Fold	Train N	Val N	Brier	Log Loss	RMSE
Fold 1	701	2,097	0.2542	0.7024	2.432
Fold 2	1,399	1,399	0.2491	0.6915	2.376
Fold 3	2,103	695	0.2453	0.6838	2.373

In-Game Checkpoints — Last 30

Checkpoint	Games	Accuracy	Brier	Log Loss
end_p1	51	68.6%	0.1947	0.5889
end_p2	51	78.4%	0.1677	0.5093
ot_start	14	71.4%	0.2148	0.6131
p3_10	51	78.4%	0.1114	0.3390
p3_5	51	84.3%	0.0851	0.2730
pregame	51	51.0%	0.2527	0.6987

In-Game Checkpoints — Season To Date

Checkpoint	Games	Accuracy	Brier	Log Loss
end_p1	1384	67.0%	0.2065	0.5979
end_p2	1384	78.7%	0.1450	0.4423
ot_start	345	64.1%	0.1942	0.5547
p3_10	1384	83.5%	0.1027	0.3212
p3_5	1384	85.5%	0.0883	0.2767
pregame	1384	53.4%	0.2477	0.6886

In-Game Calibration — Pregame (Last 30 Days)

Bin	Count	Mean Pred	Observed
4	6	48.4%	33.3%
5	43	54.9%	48.8%
6	2	64.5%	50.0%

In-Game Calibration — End P2 (Last 30 Days)

Bin	Count	Mean Pred	Observed
0	6	4.5%	16.7%
1	6	14.4%	16.7%
2	4	26.5%	25.0%
3	7	34.9%	28.6%
4	3	47.9%	0.0%
5	3	54.5%	66.7%
6	7	64.0%	57.1%
7	5	75.5%	60.0%
8	4	84.4%	100.0%
9	6	94.3%	100.0%

In-Game Calibration — P3 10 (Last 30 Days)

Bin	Count	Mean Pred	Observed
0	15	2.9%	6.7%
1	7	15.4%	14.3%
4	2	43.7%	100.0%
5	7	53.9%	28.6%
6	3	62.6%	66.7%
8	4	84.8%	75.0%
9	13	97.3%	100.0%

xG Holdout — Contextual

Train: 2023-10-10 – 2025-12-11 | Test: 2025-12-12 – 2026-05-14

Games (test): 885 | Shots (test): 76090 | ROC AUC: 0.785 | Log Loss: 0.2214 | Brier: 0.0601

xG Splits — Contextual Strength State

Split	Shots	Goal Rate	AUC	Log Loss	Brier
Even	60771	6.2%	0.778	0.2023	0.0539
PP	12929	10.6%	0.727	0.2915	0.0809
PK	1690	7.2%	0.841	0.2112	0.0602
EmptyNet	700	51.1%	0.743	0.6079	0.2114

xG Splits — Contextual Shot Type

Split	Shots	Goal Rate	AUC	Log Loss	Brier
wrist	32222	7.2%	0.812	0.2072	0.0564
snap	19339	8.4%	0.775	0.2483	0.0701
slap	9096	4.8%	0.720	0.1796	0.0446
tip-in	7349	6.3%	0.663	0.2264	0.0579
backhand	5643	8.9%	0.823	0.2357	0.0658
deflected	1234	11.8%	0.697	0.3301	0.0958
wrap-around	461	5.2%	0.738	0.1821	0.0455
bat	398	8.5%	0.789	0.2426	0.0661
poke	248	8.9%	0.658	0.2861	0.0747
between-legs	52	9.6%	0.783	0.2554	0.0678
nan	41	61.0%	0.732	1.6486	0.2915
cradle	7	14.3%	1.000	0.2286	0.0656

xG Holdout — Neutral

Train: 2023-10-10 – 2025-11-16 | Test: 2025-11-17 – 2026-04-13

Games (test): 991 | Shots (test): 85192 | ROC AUC: 0.783 | Log Loss: 0.2247 | Brier: 0.0613

xG Splits — Neutral Strength State

Split	Shots	Goal Rate	AUC	Log Loss	Brier
Even	68274	6.2%	0.781	0.2021	0.0540
PP	14330	10.6%	0.705	0.3133	0.0885
PK	1823	7.1%	0.822	0.2146	0.0605
EmptyNet	765	50.7%	0.744	0.6043	0.2089

xG Splits — Neutral Shot Type

Split	Shots	Goal Rate	AUC	Log Loss	Brier
wrist	36458	7.1%	0.809	0.2083	0.0569
snap	21423	8.3%	0.774	0.2486	0.0703
slap	10084	5.0%	0.713	0.1850	0.0461
tip-in	8203	6.6%	0.669	0.2331	0.0602
backhand	6300	9.5%	0.812	0.2558	0.0741
deflected	1372	11.7%	0.712	0.3249	0.0946
wrap-around	542	4.4%	0.655	0.1758	0.0415
bat	436	8.9%	0.835	0.2328	0.0639
poke	269	9.7%	0.747	0.2677	0.0705
between-legs	56	12.5%	0.777	0.3116	0.0909
nan	42	61.9%	0.713	2.3896	0.2752
cradle	7	14.3%	1.000	0.2025	0.0577

Monthly Performance Trends

Track how model performance varies month-to-month across the season.

Month	Games	Accuracy	Brier	Log Loss
2023-10	140	62.1%	0.2210	0.6297
2023-11	213	66.2%	0.2159	0.6207
2023-12	219	66.7%	0.2185	0.6271
2024-01	208	62.5%	0.2168	0.6222
2024-02	172	67.4%	0.2064	0.6014
2024-03	228	71.1%	0.2036	0.5951
2024-04	132	62.1%	0.2266	0.6436
2024-10	166	69.9%	0.2040	0.5965
2024-11	220	66.8%	0.2145	0.6181
2024-12	214	69.2%	0.2053	0.6002
2025-01	224	60.7%	0.2275	0.6459
2025-02	122	55.7%	0.2473	0.6864
2025-03	234	62.0%	0.2280	0.6475
2025-04	132	53.8%	0.2467	0.6861
2025-10	180	57.2%	0.2494	0.6939
2025-11	225	53.8%	0.2439	0.6795
2025-12	226	54.0%	0.2499	0.6936
2026-01	240	56.7%	0.2439	0.6809
2026-02	74	67.6%	0.2234	0.6378
2026-03	242	54.5%	0.2493	0.6913
2026-04	125	59.2%	0.2273	0.6450

Playoff Model Performance

Game-level and series-level accuracy across playoff rounds.

Playoff Games

Round	Games	Accuracy	Brier	Log Loss
All Rounds	72	58.3%	0.2441	0.6814
Round 1	45	60.0%	0.2394	0.6718
Round 2	22	59.1%	0.2423	0.6777
Round 3	5	40.0%	0.2943	0.7840

Playoff Series

Round	Series	Accuracy	Brier	Log Loss
All Rounds	12	66.7%	0.2210	0.6346
Round 1	8	62.5%	0.2134	0.6189
Round 2	4	75.0%	0.2361	0.6660

Playoff Calibration (Pred vs Observed)

Mean Pred Observed