Tested as a player trait across 365 players at four positions. A first-half man-versus-zone split does not predict the second half anywhere — and at quarterback it inverts at −0.416.
Do some receivers really beat man coverage?
That specific receivers have a durable edge against man coverage is stated every Sunday and used as a start/sit tiebreaker.
Split each player’s games at his own median opponent man-rate, compute his man-minus-zone split over the first half of his sample, then check whether it predicts the second half. It does not, at any position. Wide receivers came in at +0.080 across 161 players, running backs +0.027 across 97, tight ends −0.133 across 69, and quarterbacks −0.416 across 38 — the quarterback figure inverting hard. A passer who crushed man coverage in the first half of his sample tends to do the opposite in the second.
| Position | Players | First half predicts second half | Verdict |
|---|---|---|---|
| WR | 161 | +0.080 | Does not persist |
| RB | 97 | +0.027 | Does not persist |
| TE | 69 | −0.133 | Reverses |
| QB | 38 | −0.416 | Actively reverses |
An apparent man-coverage advantage in a completed sample is mostly the residue of which defenses a player happened to face and what those defenses were doing on those particular snaps. If it were a skill it would have to persist within the same player across the same season. It does not — and the quarterback figure goes further than failing to persist. At −0.416 it is close to a pure mean-reversion signal: the splits that look most impressive are the ones most likely to regress hardest.
Each player is split at his own median opponent man-rate rather than a league-wide threshold, so a receiver who rarely faces man-heavy defenses is not compared against one who always does. That control is what makes this a test of the trait rather than of the schedule.
Figures on this page come from the GWTTKB methodology report, where the full testing record is published.
This says the split does not persist as a forward-looking player trait. It does not say the historical split is fabricated — it happened. It says you cannot bet on it happening again.