Skip to content

Model Performance

Graded results — every prediction the model has made, whether or not you bet it
← Model picks
Moneyline accuracy
62.5%
Market accuracy
71.9%
Brier score
0.2457
Log loss
0.688 vs 0.693 baseline
Predictions graded
32
Scored against the benchmarks
MethodGamesBrierLog lossAccuracy
hamstats model320.24570.687662.5%
Always pick home320.24900.691253.1%
Betting marketbenchmark, not target320.22210.639371.9%

Lower Brier and log loss are better. The floor is what you'd score knowing nothing but home-field advantage; the market is the ceiling — it is measured against, deliberately not chased. Landing just below it with honest calibration is the goal.

By market
Moneyline
n=32
62.5%
straight-up hit rate
Brier 0.2457 vs 0.2490 flatmarket 71.9%
Anytime TD
n=941
46.9%
top pick per game scored
Brier 0.0980 vs 0.1136 flatany player 13.1%
First TD
n=941
12.5%
top pick scored first
Brier 0.0290 vs 0.0309 flatany player 3.2%
Calibration — moneyline
50-55
62.5% · n=8
55-65
64.3% · n=14
65-75
60.0% · n=5
75-85
60.0% · n=5
By week
1
2
Calibration — touchdown markets
Anytime TD
0-2
0.0% · n=83
2-5
5.1% · n=253
5-10
5.3% · n=189
10-20
15.4% · n=162
20-35
23.4% · n=197
35-50
46.0% · n=50
50+
85.7% · n=7
First TD
0-2
0.6% · n=517
2-5
2.3% · n=264
5-10
12.2% · n=139
10-20
19.0% · n=21

Buckets are the model's predicted probability that a player scores, against how often players in that bucket actually did. Graded over the active roster — a player who dressed and never touched the ball counts as a miss, while one who was inactive, on the practice squad or on IR is left out, because a touchdown prop voids rather than losing when its player never plays.

Last graded slate
GamePickModelBookResult
NYG @ LALA69.4%71.8%hit
CAR @ ATLCAR56.0%58.3%hit
IND @ KCKC54.0%69.6%hit
MIA @ SFSF77.7%86.3%hit
WAS @ DALDAL61.1%65.7%hit
SEA @ ARISEA62.3%65.7%hit
LV @ LACLAC62.5%73.4%miss
JAX @ DENJAX64.2%42.8%miss

The scoreboard above and the last-graded slate cover the moneyline market, measured on graded predictions only — ties excluded. The market comparison is the published closing moneyline with the vig removed. It is never a model input: the model is built without any market feature, so that whether it is accurate stays a question this page can actually answer. The touchdown markets are scored against real per-player results, and are headlined by the model's top-ranked player in each game rather than a hit rate: scoring a touchdown is rare enough across a full roster that “nobody scores” would look over 90% accurate.