niftiIQLab
niftiNFL — Backtest
How the method would have done, season by season, before it went live. Walk-forward: every game predicted using only games played before it. · predictions are published to show the model works. nothing here is betting advice.
65.0%
straight-up, 4,146 regular-season games 2010–2025
66.5%
Vegas closing favorite, same games
±1.5
95% margin on that accuracy (points)
0.219
Brier score · Vegas 0.211 (lower is better)
87%
of picks agree with Vegas
10.4
avg. margin error (pts)
What this says, plainly: over sixteen seasons the model picks the winner about 65 times in 100 without ever seeing a betting line; the market's favorite wins about 67 in 100. That gap is real but small, and no public model closes it entirely — the NFL is a coin with a heavy side, not a solved game. The useful claims are the calibration below (a 70% pick really does win about 70% of the time) and that the live record is locked before kickoff.

Season by season

Regular season only. One season is ~270 games, so ±6 points is normal noise around the long-run average.
50% 60% 70% 80% 2010: 61.7% (240 games) · Vegas 66.7% ’10 2011: 68.4% (256 games) · Vegas 66.8% ’11 2012: 63.1% (255 games) · Vegas 64.3% ’12 2013: 63.5% (255 games) · Vegas 71.4% ’13 2014: 69.4% (255 games) · Vegas 66.7% ’14 2015: 64.1% (256 games) · Vegas 62.5% ’15 2016: 65.0% (254 games) · Vegas 64.4% ’16 2017: 65.6% (256 games) · Vegas 70.8% ’17 2018: 68.9% (254 games) · Vegas 66.1% ’18 2019: 64.3% (255 games) · Vegas 64.3% ’19 2020: 67.1% (255 games) · Vegas 67.5% ’20 2021: 61.3% (271 games) · Vegas 62.4% ’21 2022: 62.5% (269 games) · Vegas 66.2% ’22 2023: 63.6% (272 games) · Vegas 68.0% ’23 2024: 70.2% (272 games) · Vegas 71.3% ’24 2025: 60.9% (271 games) · Vegas 65.3% ’25
Bars: our walk-forward accuracy. Dashed marks: the Vegas closing favorite the same season.
SeasonGamesOur accuracyVegas favoriteBrier
201024061.7%66.7%0.234
201125668.4%66.8%0.212
201225563.1%64.3%0.221
201325563.5%71.4%0.216
201425569.4%66.7%0.214
201525664.1%62.5%0.226
201625465.0%64.4%0.219
201725665.6%70.8%0.208
201825468.9%66.1%0.212
201925564.3%64.3%0.224
202025567.1%67.5%0.213
202127161.3%62.4%0.227
202226962.5%66.2%0.222
202327263.6%68.0%0.227
202427270.2%71.3%0.205
202527160.9%65.3%0.223

When we differ from Vegas

The honest cut. On the 3,613 games where our pick and the closing favorite agree, both are right 68.1% of the time. The 529 games where we differ (13% of games) are where the whole gap lives.
43.9%
our pick wins, when we differ (529 games)
56.1%
Vegas favorite wins, same games
41.5%
when we back the home team against Vegas (301)
46.9%
when we back the away team against Vegas (228)
Read this the right way round: a game where we differ from the market is not a hidden edge — it is usually a game where the market knows something the public feeds don't. The live version of this record is on the Record page.

Calibration over the backtest

Stated confidence bucket vs. actual hit rate. Bars should climb in step with the labels.
90–100% (42)
95%
80–90% (394)
82%
75–80% (376)
76%
70–75% (560)
72%
65–70% (578)
67%
60–65% (693)
60%
55–60% (752)
57%
50–55% (751)
55%

How it was run

The rule book lives in docs/BACKTEST_METHOD.md.

For every week from 2010 to 2025, the model was trained on every completed game before that week (history starts in 2009) and asked to pick that week's games. The same feature builder and model code make the live picks, so the backtest cannot be better than live by construction. A leak audit rebuilds features for sampled games from strictly-prior data and asserts they match. Ties are pushed. The Vegas column uses the closing spread from the nflverse schedule feed; the model never sees it.

Known limit: the historical starting quarterback is the one who actually started; live picks use the named starter from the feed at run time, so a late scratch can cost a live pick that the backtest would have got.