← Statistics · updated 28 Aug 2026
How volatile is each fantasy football position, week to week?
The median week-to-week swing of every position, from 1,634 player-seasons: quarterbacks are the steadiest, defenses the wildest.
Key numbers
- 47%quarterback: a typical week swings this much around his average the steadiest position
- 84%defense / special teams the wildest
- 69% · 70% · 66%wide receiver · tight end · running back
- 1,634player-seasons measured 2021–2025 regular seasons
| Position | Weekly swing (sd ÷ mean) | Player-seasons |
|---|---|---|
| Quarterback (QB) | 0.466 (47%) | 178 |
| Kicker (K) | 0.526 (53%) | 162 |
| Running back (RB) | 0.658 (66%) | 334 |
| Wide receiver (WR) | 0.691 (69%) | 557 |
| Tight end (TE) | 0.703 (70%) | 244 |
| Defense / special teams (DST) | 0.837 (84%) | 159 |
Questions this page answers
Which fantasy football position is the most consistent week to week?
Quarterback. Over 2021–2025 regular seasons, the median quarterback's weekly PPR points had a standard deviation of 47% of his own average (178 player-seasons). Kickers were next at 53%.
Which position is the most volatile?
Defense / special teams, at 84% of the mean (159 player-seasons), ahead of tight ends (70%), wide receivers (69%) and running backs (66%).
How does weekly variance change a start/sit call?
Two players projected a point apart at a volatile position are a coin flip; the same gap at quarterback is a real edge. Ribby uses these per-position spreads to turn a projected margin into a win probability, and calls a swap inside the noise band a toss-up rather than a recommendation.
How was this measured?
For every player-season with at least 8 games and at least 4 points per game, the standard deviation of his weekly PPR points was divided by his mean; the number per position is the median of that ratio, which resists the long right tail a low-mean player creates.
How this was measured
Our own weekly stat rows for the 2021–2025 regular seasons, scored full PPR. A player-season qualifies with at least 8 games played and at least 4 points per game. For each one we took the standard deviation of his weekly points divided by his own mean, then the MEDIAN of that ratio across the position — median rather than mean because the ratio has a long right tail (a low-mean player divides by a small number) and a mean chases it.
Three limits, stated rather than hidden: players are treated as independent (a quarterback and his own receiver are correlated, so a stack is wider than this); it describes the past five seasons, not any one week's matchup; and it is a spread around a projection, not a projection. The product uses exactly these numbers to turn a projected weekly margin into a win probability, and reports that probability in whole percents because that is the precision it has earned.
Reproduce it with scripts/nfl-variance.mts, which re-runs the query against the tables and fails if these constants drift. Measured 28 Aug 2026.
Cite this
These numbers are free to quote with a link (CC BY 4.0). Suggested: Ribby, “Fantasy Football Weekly Variance by Position, 2021–2025”, ribby.ai, updated 28 Aug 2026. https://ribby.ai/stats/fantasy-football-weekly-variance
https://ribby.ai/stats/fantasy-football-weekly-variance