One measure, in full · Next Gen Stats: what holds · Receiving
Separation
Definition
One definition, the column that binds its range, and what the figure is a share of.
lag-1 autocorrelation of the player's weekly deviation from his own season mean
ngs_stability.within_lag1.receiving.avg_separation
- Source
ngs; n counts independent players- Bound column
- starts 2016: ngs (first season on the release, measured in f_ngs_week) is not usable before it
- Denominator
- The subject's own total: no one else shares it.
- Season range
- Current 2016–2026. Current. Its inputs cover 2016 to 2026 and reach the season in progress.
- Refreshed
- This file was generated Tue, Oct 27, 9:32 AM ET. It does not state a refresh schedule.
Distribution
No distribution: the file is one figure for the whole league, not a value per player.
Does it carry?
No: the interval sits wholly left of zero, so a week above his own average is not followed by another.
Week-to-week carry-over
r = −0.07
[−0.09, −0.04] · n = 326 · 10,128 rows
Doesn’t carry over- Separationthis measure−0.07 [−0.09, −0.04]
- Share of team air yardsfor contrast−0.07 [−0.10, −0.05]
- Cushionfor contrast−0.11 [−0.14, −0.08]
Separation, among receiving stats
Separation: r −0.07 week to week (interval −0.09 to −0.04, 326 players). A week above his own average does not carry into the next.
Whether a player's week above his own season average is followed by another: the correlation of each week's deviation from his own mean with the next week's. An interval wholly above zero carries; one that reaches zero, or sits under it, does not.
- Cushion−0.11
- Expected yards after catch−0.08
- Separation−0.07
- Yards after catch above expectation−0.08
- Share of team air yards−0.07
analytics/nfl/ngs_stability.within_lag1.receiving.avg_separation.json · generated Tue, Oct 27, 9:32 AM ET
This stat, all three ways
| Measure | Estimate | Interval | n |
|---|---|---|---|
| Week to week | −0.07 | −0.09 to −0.04 | 326 players |
| Between players | 26% | 24% to 29% | 326 players |
| Naive | 0.20 | 0.17 to 0.23 | 326 players |
How it is measured
- Week to week
- lag-1 autocorrelation of the player's weekly deviation from his own season mean
- Between players
- share of variance between player-seasons
- Naive
- pooled lag-1 correlation of raw weekly values
- Interval
- Interval from resampling players with replacement, 2,000 times. Built at 95%.
- Sample
- n counts independent players, never weekly rows (10,128 weekly rows from those 326).
Leaderboard
No leaderboard: the file is one league-wide figure, not a value per player.
What the file does and does not say
Four limits, stated plainly.
- Interval from resampling players with replacement, 2,000 times. Built at 95%. That coverage is nominal: this page does not check it against outcomes.
- n counts 326 independent players, not the 10,128 rows behind them. Rows inside one player share a context, so counting them as the sample would draw the interval too narrow.
- It covers 2016 to 2026, and says why in the producer's words: starts 2016: ngs (first season on the release, measured in f_ngs_week) is not usable before it.
- It is one figure for the whole league, pooled. It does not say what any one player does, and it does not split by position.