Does Weather Actually Affect College Football Outcomes? 6 Hypotheses Tested

Analysis · by Rick's Picks Analytics

Weather is the great uncontrollable variable in college football. Every fan has opinions about it: "Wind kills the passing game." "Dome teams can't play in the cold." "SEC teams thrive in humidity." We formalized six weather hypotheses and tested them with real data.

How We Tested

Every test uses our standard framework: Welch's t-test for comparing group means, Cohen's d for effect size, binomial tests against 50% for cover/over rates, and Bonferroni correction across all six hypotheses (adjusted alpha = 0.05/6 = 0.00833). Results must pass Bonferroni on the training set (2015-2022) AND replicate at p < 0.05 on the test set (2023-2024) to be considered robust.

Hypothesis 1: Dome Games Score More

The question: Do games played in domes produce higher total scoring than outdoor games?

The logic: Domes eliminate wind, precipitation, and temperature extremes. Receivers run crisper routes on dry turf. Kickers don't fight crosswinds. Quarterbacks grip a dry ball.

Our test: Welch's t-test comparing mean combined scoring in dome games versus outdoor games, with Cohen's d measuring the practical magnitude.

What we find: Dome games do tend to produce slightly higher total scores. The effect is consistent across training and test periods, though the magnitude (Cohen's d) matters more than statistical significance alone. A "statistically significant" difference of 1.5 points per game is real but may not be large enough to exploit profitably after vig.

Hypothesis 2: Cold Weather Pushes Games Under

The question: Do games played below 40F go UNDER at a rate above 50%?

Why it matters: If cold games systematically score lower than oddsmakers expect, there's a consistent under bias to exploit. We test this with a binomial test against 50%.

The finding: This is one of our more actionable findings. Sub-40-degree games show a meaningful lean toward the under. Cold weather affects passing accuracy, ball handling, and kicking consistency in ways that compound to suppress scoring. The effect is strongest in truly frigid conditions (below 32F) and weakens as temperatures approach 50F.

How we use it: For games forecast below 40F, our algorithm applies a negative adjustment to the projected total. This is one of the more reliable weather signals in our system.

Hypothesis 3: Wind Reduces Scoring

The question: Do games with high winds (above 15 MPH) produce lower total scoring than calm conditions?

The test: Welch's t-test comparing high-wind games to calm (below 10 MPH) games.

The finding: Wind does reduce scoring, particularly when sustained gusts exceed 15 MPH. The effect is intuitive -- deep passing becomes unreliable, field goal range shrinks, and punting becomes less predictable. However, the effect size varies. Games with 15-20 MPH winds show a moderate effect; games above 25 MPH show a stronger effect but are relatively rare.

Hypothesis 4: Precipitation and Spread Impact

The question: Does rain or snow affect point margins and home team cover rates?

The logic: Precipitation should create sloppier games with more turnovers, potentially equalizing talent gaps and helping underdogs cover.

Our analysis: We test both the ATS margin difference (is the home team's margin different in rain/snow?) and the home cover rate (do home teams cover more or less in precipitation?).

The finding: Precipitation effects are present but modest. Rain games show slightly more variance in outcomes, which can benefit underdogs. Snow games, while dramatic, are too rare in our dataset to draw firm conclusions with Bonferroni-level confidence.

Hypothesis 5: Extreme Heat

The question: Do games above 85F produce different scoring patterns than moderate-temperature games (65-75F)?

The finding: Extreme heat shows minimal scoring impact in our data. This might be because most 85F+ games occur in September when teams are conditioned for heat, or because both teams face the same conditions (unlike cold, where dome teams traveling to northern stadiums face an unfamiliar environment).

Hypothesis 6: SEC Humidity Advantage

The question: Do SEC home teams playing in high humidity (above 70%) win at a rate above 50% against non-SEC opponents?

The theory: SEC teams in the Deep South are acclimated to oppressive humidity that saps visiting teams from drier climates. This could amplify home-field advantage beyond the normal crowd effect.

The finding: This is a compelling theory with limited statistical support. SEC home teams do perform well in humid conditions, but separating the humidity effect from the general SEC home-field advantage (large stadiums, passionate crowds, recruiting advantages) is difficult. The sample of SEC-vs-non-SEC games in high humidity that also have spread data is relatively small, making it hard to achieve Bonferroni-level significance.

What Makes It Into Our Algorithm

Of the six hypotheses, cold weather (H2) and wind (H3) produce the most reliable, replicable effects. Dome scoring (H1) is directionally correct but small. Precipitation (H4) adds noise. Heat (H5) and humidity (H6) don't show strong enough effects to justify algorithmic weight.

Our prediction system incorporates weather through a tiered adjustment: temperature below 40F shifts projected totals down, sustained winds above 15 MPH provide an additional reduction, and dome games receive a small upward adjustment. These are calibrated from the training data and validated on the test set.

All results are Bonferroni-corrected across the six-hypothesis family, with replication required on 2023-2024 holdout data.

Rick's Picks publishes statistical analysis of college football for informational and entertainment purposes. Nothing here is betting advice. 21+. If gambling is affecting you or someone you know, call 1-800-GAMBLER.