Public results · auto-settled · no betting advice

Prediction track record

Methodology Public prediction archive

Season settled: 1,958
Best win rate: 54.4%
Avg Brier: 0.2399
Calibration n: 1,928

Season trust signals

LOLlol-elo-v1n=5

Brier: 0.1943; Beats a coin flip
Current streak: W3; since Oct 16

MLBelo-v1n=1869

Brier: 0.2468; Near coin flip
Current streak: L1; since Oct 16

NBAelo-v1n=39

Brier: 0.2682; Worse than a coin flip (honest signal)
Current streak: L1; since Oct 17

NBLelo-v1n=1

Brier: 0.2373; Near coin flip
Current streak: W1; since Oct 16

NFLelo-v1n=1

Brier: 0.2887; Worse than a coin flip (honest signal)
Current streak: L1; since Oct 16

NHLelo-v1n=13

Brier: 0.2038; Near coin flip
Current streak: L1; since Oct 17

Week highlights (last 7 days)

Most confident hit

LOL

Karmine CorpvsMVK Esports

We predicted: Home win 81.7%
Actual: Home win 3 – 0 Away win

Most confident miss

LOL

Karmine CorpvsCloud9 Kia

We predicted: Away win 67.2%
Actual: Home win 3 – 2 Away win

Home vs away breakdown (moneyline)

Splits our home-side picks from our away-side picks. If one side wins markedly more than the other, the model is mis-pricing home-court advantage.

LOL

Away win

Win rate: 50.0%
Hits / attempts: 1 / 2

LOL

Home win

Win rate: 100.0%
Hits / attempts: 3 / 3

MLB

Away win

Win rate: 56.4%
Hits / attempts: 110 / 195

MLB

Home win

Win rate: 54.1%
Hits / attempts: 906 / 1674

NBA

Home win

Win rate: 48.7%
Hits / attempts: 19 / 39

NBL

Away win

Win rate: 100.0%
Hits / attempts: 1 / 1

NFL

Away win

Win rate: 0.0%
Hits / attempts: 0 / 1

NHL

Away win

Win rate: 100.0%
Hits / attempts: 1 / 1

NHL

Home win

Win rate: 75.0%
Hits / attempts: 9 / 12

Last 7 days

Recent trend

Moneyline · win rate

League	Model	Win rate	Hits / attempts	Settled	Avg confidence
MLB	elo-v1	Sample too small (n=16)	5 / 8	16	53.9%
NBA	elo-v1	Sample too small (n=14)	9 / 14	14	58.9%
NHL	elo-v1	Sample too small (n=13)	10 / 13	13	57.0%
LOL	lol-elo-v1	Sample too small (n=5)	4 / 5	5	65.9%
NBL	elo-v1	Sample too small (n=1)	1 / 1	1	51.3%
NFL	elo-v1	Sample too small (n=1)	0 / 1	1	53.7%

Total · MAE error

League	Model	MAE	Bias	Avg confidence
BUNDESLIGA	bundesliga-elo-v1	Sample too small (n=0)	—	53.2%
EREDIVISIE	eredivisie-elo-v1	Sample too small (n=0)	—	60.7%
EUROPA-LEAGUE	europa-league-elo-v1	Sample too small (n=0)	—	54.5%
LA-LIGA	laliga-elo-v1	Sample too small (n=0)	—	53.7%
LIGUE-1	ligue1-elo-v1	Sample too small (n=0)	—	52.1%
MLB	score-model-v1	Sample too small (n=0)	—	56.0%
NBA	score-model-v1	Sample too small (n=0)	—	54.0%
NFL	score-model-v1	Sample too small (n=0)	—	56.0%
NHL	score-model-v1	Sample too small (n=0)	—	51.8%
SERIE-A	seriea-elo-v1	Sample too small (n=0)	—	68.9%

Last 30 days

Monthly trend

Moneyline · win rate

League	Model	Win rate	Hits / attempts	Settled	Avg confidence
MLB	elo-v1	59.3%	105 / 177	186	54.0%
NBA	elo-v1	Sample too small (n=14)	9 / 14	14	58.9%
NHL	elo-v1	Sample too small (n=13)	10 / 13	13	57.0%
LOL	lol-elo-v1	Sample too small (n=5)	4 / 5	5	65.9%
NBL	elo-v1	Sample too small (n=1)	1 / 1	1	51.3%
NFL	elo-v1	Sample too small (n=1)	0 / 1	1	53.7%

Total · MAE error

League	Model	MAE	Bias	Avg confidence
BUNDESLIGA	bundesliga-elo-v1	Sample too small (n=0)	—	53.2%
EREDIVISIE	eredivisie-elo-v1	Sample too small (n=0)	—	60.7%
EUROPA-LEAGUE	europa-league-elo-v1	Sample too small (n=0)	—	54.5%
LA-LIGA	laliga-elo-v1	Sample too small (n=0)	—	53.7%
LIGUE-1	ligue1-elo-v1	Sample too small (n=0)	—	52.1%
MLB	score-model-v1	Sample too small (n=0)	—	56.0%
NBA	score-model-v1	Sample too small (n=0)	—	54.0%
NFL	score-model-v1	Sample too small (n=0)	—	56.0%
NHL	score-model-v1	Sample too small (n=0)	—	51.8%
SERIE-A	seriea-elo-v1	Sample too small (n=0)	—	68.9%

Season

All-time

Moneyline · win rate

League	Model	Win rate	Hits / attempts	Settled	Avg confidence
MLB	elo-v1	54.4%	1016 / 1869	1889	53.8%
NBA	elo-v1	48.7%	19 / 39	49	60.6%
NHL	elo-v1	Sample too small (n=13)	10 / 13	13	57.0%
LOL	lol-elo-v1	Sample too small (n=5)	4 / 5	5	65.9%
NBL	elo-v1	Sample too small (n=1)	1 / 1	1	51.3%
NFL	elo-v1	Sample too small (n=1)	0 / 1	1	53.7%

Total · MAE error

League	Model	MAE	Bias	Avg confidence
BUNDESLIGA	bundesliga-elo-v1	Sample too small (n=0)	—	53.2%
EREDIVISIE	eredivisie-elo-v1	Sample too small (n=0)	—	60.7%
EUROPA-LEAGUE	europa-league-elo-v1	Sample too small (n=0)	—	54.5%
LA-LIGA	laliga-elo-v1	Sample too small (n=0)	—	53.7%
LIGUE-1	ligue1-elo-v1	Sample too small (n=0)	—	52.1%
MLB	score-model-v1	Sample too small (n=0)	—	56.0%
NBA	score-model-v1	Sample too small (n=0)	—	54.0%
NFL	score-model-v1	Sample too small (n=0)	—	56.0%
NHL	score-model-v1	Sample too small (n=0)	—	51.8%
SERIE-A	seriea-elo-v1	Sample too small (n=0)	—	68.9%

World Cup 1X2 (90-minute win-draw-loss)

Brier is a three-class normalized score (0–1, lower is better); not yet benchmarked against a favourite baseline or market odds.

1X2 · win rate

League	Model	Win rate	Hits / attempts	Brier	Settled
FIFA World Cup	wc-elo-v1	62.0%	57 / 92	0.2567	92
LaLiga	laliga-elo-v1	51.4%	36 / 70	0.2800	70
Ligue 1	ligue1-elo-v1	56.4%	31 / 55	0.3033	55
Serie A	seriea-elo-v1	56.9%	29 / 51	0.2834	51
Premier League	epl-elo-v1	40.0%	20 / 50	0.3315	50
Bundesliga	bundesliga-elo-v1	54.4%	25 / 46	0.2860	46
UEFA Europa League	europa-league-elo-v1	Sample too small (n=18)	12 / 18	0.2663	18
Eredivisie	eredivisie-elo-v1	Sample too small (n=1)	1 / 1	0.1597	1

Secondary markets (season)

Spread Brier uses the model's cover probability (0.25 ≈ coin flip, lower is better). Same auto-settled data as above, published unfiltered.

Spread · win rate + Brier

League	Model	Win rate	Hits / attempts	Brier	Avg cover prob	Settled
EUROPA-LEAGUE	europa-league-elo-v1	Sample too small (n=18)	7 / 18	—	—	18
NBA	score-model-v1	Sample too small (n=14)	11 / 14	—	—	14
NHL	score-model-v1	Sample too small (n=13)	8 / 13	—	—	13
MLB	score-model-v1	Sample too small (n=3)	1 / 2	—	—	3
NFL	score-model-v1	Sample too small (n=1)	1 / 1	—	—	1
BUNDESLIGA	bundesliga-elo-v1	Sample too small (n=1)	1 / 1	—	—	1
SERIE-A	seriea-elo-v1	Sample too small (n=1)	0 / 1	—	—	1
EREDIVISIE	eredivisie-elo-v1	Sample too small (n=1)	1 / 1	—	—	1
LA-LIGA	laliga-elo-v1	Sample too small (n=1)	1 / 1	—	—	1
LIGUE-1	ligue1-elo-v1	Sample too small (n=1)	0 / 1	—	—	1

First-half winner · win rate

League	Model	Win rate	Hits / attempts	Settled	Avg confidence
NBA	score-model-v1	Sample too small (n=14)	12 / 14	14	50.8%
NBL	score-model-v1	Sample too small (n=1)	0 / 1	1	48.3%
NFL	score-model-v1	Sample too small (n=1)	0 / 1	1	50.1%

First-half + team totals · MAE error

League	Model	Market	MAE	Bias
BUNDESLIGA	bundesliga-elo-v1	First-half total	Sample too small (n=0)	—
EREDIVISIE	eredivisie-elo-v1	First-half total	Sample too small (n=0)	—
EUROPA-LEAGUE	europa-league-elo-v1	First-half total	Sample too small (n=0)	—
LA-LIGA	laliga-elo-v1	First-half total	Sample too small (n=0)	—
LIGUE-1	ligue1-elo-v1	First-half total	Sample too small (n=0)	—
NBA	score-model-v1	First-half total	Sample too small (n=0)	—
NFL	score-model-v1	First-half total	Sample too small (n=0)	—
SERIE-A	seriea-elo-v1	First-half total	Sample too small (n=0)	—
BUNDESLIGA	bundesliga-elo-v1	Away team total	Sample too small (n=0)	—
EREDIVISIE	eredivisie-elo-v1	Away team total	Sample too small (n=0)	—
EUROPA-LEAGUE	europa-league-elo-v1	Away team total	Sample too small (n=0)	—
LA-LIGA	laliga-elo-v1	Away team total	Sample too small (n=0)	—
LIGUE-1	ligue1-elo-v1	Away team total	Sample too small (n=0)	—
NBA	score-model-v1	Away team total	Sample too small (n=0)	—
NFL	score-model-v1	Away team total	Sample too small (n=0)	—
NHL	score-model-v1	Away team total	Sample too small (n=0)	—
SERIE-A	seriea-elo-v1	Away team total	Sample too small (n=0)	—
BUNDESLIGA	bundesliga-elo-v1	Home team total	Sample too small (n=0)	—
EREDIVISIE	eredivisie-elo-v1	Home team total	Sample too small (n=0)	—
EUROPA-LEAGUE	europa-league-elo-v1	Home team total	Sample too small (n=0)	—
LA-LIGA	laliga-elo-v1	Home team total	Sample too small (n=0)	—
LIGUE-1	ligue1-elo-v1	Home team total	Sample too small (n=0)	—
NBA	score-model-v1	Home team total	Sample too small (n=0)	—
NFL	score-model-v1	Home team total	Sample too small (n=0)	—
NHL	score-model-v1	Home team total	Sample too small (n=0)	—
SERIE-A	seriea-elo-v1	Home team total	Sample too small (n=0)	—

Perfect calibration

Confidence calibration

When we say '70% to win,' how often do those teams actually win? Points hugging the dashed line mean the model's confidence is honest. Blue = beats prediction (under-confident), red = worse than prediction (over-confident), green = within ±5%.

Moneyline model · 1928 settled games (voids excluded)