Do NFL Deep-Target Rates and Efficiency Fall in 15+ MPH Wind Through the 2024 NFL Season

If you're banking on a deep-threat receiver in a windy game, the data through 2024 says the targets should still come — but the production on those targets is where the risk lives.

The fantasy question here is whether wind kills the deep ball by shrinking volume or by degrading efficiency. Using 101 NFL regular-season games with observed winds of at least 15 mph against 807 games below 10 mph through the 2024 season, the answer is clearly the second: offenses kept attempting deep passes at the same rate, but those passes connected less often and produced less value when they did.

Deep-target rate — the share of pass attempts thrown deep — averaged 10.8% in 15+ mph wind games and 10.8% in below-10-mph games. The adjusted change was -0.0 points with a 95% interval from -0.8 to +0.8 points and a Holm-adjusted p-value of 0.975. That's a clean null result: wind does not appear to reduce how often teams try the deep ball.

Efficiency is a different story. Deep completion rate averaged 32.4% in windy games versus 36.1% in calm games, an adjusted drop of 3.8 points with a 95% interval from -7.6 to -0.1 points and a Holm-adjusted p-value of 0.094. EPA per deep target — a per-play value measure that credits expected points gained or lost on each throw — fell from 0.399 to 0.218, an adjusted change of -0.189 with a 95% interval from -0.362 to -0.017 and a Holm-adjusted p-value of 0.094. Both intervals exclude zero, though after correcting for testing three outcomes, neither clears the conventional 0.05 bar — so treat these as suggestive, not settled.

The pattern matters for how you weigh risk. A volume-based wind penalty would show up as fewer deep targets for your field-stretching receivers; that didn't happen. Instead, the penalty lands on outcomes — the same targets are being thrown, but fewer are completed and each is worth less on average. That's a hit to the ceiling of boom-or-bust deep threats specifically, not to their opportunity.

Two boundaries on this evidence. First, these are retrospective associations from observed game-window conditions — they don't establish causality and don't prove that a pregame wind forecast improves fantasy projections. Second, this is a game-level pattern, not a sit/start rule: it shouldn't override matchup, role, or usage considerations for any individual player.