EPA splits
Expected points added per play, cut by situation. A play is a dropback (sacks and scrambles count) or a designed run. On offense higher is better; on defense it is EPA allowed, so lower is better. Cells are tinted against the league in that same column, so a team is compared with what everyone does in that situation, not with its own average. Small numbers under each value are plays; faded cells have fewer than 30.
TB, every split
EPA/play with the league average in the same situation and the team's rank among 32 (1 = best; on defense, fewest EPA allowed). Offense and defense situations are each from TB's side.
| Situation | Offense | League | Rank | Defense | League | Rank |
|---|---|---|---|---|---|---|
| All plays Overall | ||||||
| All plays | -0.092322 | +0.006 | 27/32 | -0.056294 | +0.006 | 6/32 |
| Home / away Setting | ||||||
| Home | -0.181196 | +0.004 | 30/32 | -0.091172 | +0.010 | 8/32 |
| Away | +0.046126 | +0.010 | 13/32 | -0.006122 | +0.004 | 16/32 |
| Rest days Setting | ||||||
| Short week (≤5 days) | +0.14870 | +0.186 | 4/8 | -0.01459 | +0.186 | 1/8 |
| Normal (6–8 days) | -0.181196 | -0.021 | 30/32 | -0.091172 | +0.000 | 7/32 |
| Indoors / outdoors Weather | ||||||
| Indoors | +0.14870 | -0.013 | 5/25 | -0.01459 | -0.013 | 12/25 |
| Outdoors | -0.159252 | +0.013 | 31/32 | -0.066235 | +0.013 | 9/32 |
| Temperature (outdoors) Weather | ||||||
| 80°F+ | -0.159252 | +0.019 | 18/20 | -0.066235 | +0.019 | 6/20 |
| Wind (outdoors) Weather | ||||||
| 0–9 mph | -0.159252 | +0.020 | 27/29 | -0.066235 | +0.020 | 10/29 |
| Rain or snow (outdoors) Weather | ||||||
| Dry | -0.159252 | +0.029 | 31/32 | -0.066235 | +0.029 | 7/32 |
| Score Game state | ||||||
| Trailing 9+ | +0.29059 | +0.035 | 6/29 | -0.08942 | +0.072 | 7/27 |
| Trailing 1–8 | -0.170129 | -0.001 | 26/32 | -0.20795 | -0.028 | 8/32 |
| Tied | -0.19782 | -0.010 | 25/32 | -0.01184 | -0.010 | 15/32 |
| Leading 1–8 | -0.20531 | -0.028 | 27/31 | +0.07048 | -0.001 | 19/31 |
| Leading 9+ | -0.11521 | +0.072 | 18/27 | +0.18125 | +0.035 | 23/29 |
| Win probability Game state | ||||||
| Win prob under 20% | +0.24445 | +0.044 | 9/29 | -0.17167 | +0.028 | 7/30 |
| Win prob 20–80% | -0.146246 | -0.010 | 29/32 | -0.055197 | -0.010 | 11/32 |
| Win prob over 80% | -0.15131 | +0.028 | 20/30 | +0.19630 | +0.044 | 23/30 |
| Quarter Game state | ||||||
| Q1 | -0.21870 | -0.038 | 24/32 | +0.02769 | -0.038 | 20/32 |
| Q2 | -0.07695 | +0.021 | 25/32 | -0.08376 | +0.021 | 8/32 |
| Q3 | +0.02874 | +0.016 | 16/32 | -0.19466 | +0.016 | 4/32 |
| Q4 | -0.11283 | +0.013 | 28/32 | +0.01083 | +0.013 | 15/32 |
| Two-minute Game state | ||||||
| Last 2 min of half | -0.24147 | -0.007 | 29/32 | -0.02528 | -0.007 | 15/32 |
| Rest of game | -0.067275 | +0.007 | 25/32 | -0.059266 | +0.007 | 8/32 |
| Down Situation | ||||||
| 1st down | -0.107139 | +0.002 | 28/32 | +0.075133 | +0.002 | 24/32 |
| 2nd down | -0.032101 | -0.003 | 17/32 | -0.13088 | -0.003 | 5/32 |
| 3rd down | -0.10374 | +0.009 | 24/32 | -0.28467 | +0.009 | 5/32 |
| 4th down | -0.4828 | +0.184 | 21/32 | +0.6876 | +0.184 | 22/32 |
| 3rd-down distance Situation | ||||||
| 3rd & 1–3 | +0.37022 | +0.031 | 5/32 | +0.39220 | +0.031 | 27/32 |
| 3rd & 4–6 | -0.27914 | +0.050 | 23/32 | -1.04017 | +0.050 | 1/32 |
| 3rd & 7+ | -0.31338 | -0.026 | 28/32 | -0.30530 | -0.026 | 7/32 |
| Field position Situation | ||||||
| Own 20 and in | +0.06634 | -0.014 | 13/32 | -0.03527 | -0.014 | 13/32 |
| Own 50–21 | -0.217144 | -0.029 | 31/32 | -0.093129 | -0.029 | 5/32 |
| Opp 21–49 | +0.03192 | +0.036 | 16/32 | -0.07994 | +0.036 | 10/32 |
| Red zone (opp 20 in) | -0.06752 | +0.053 | 22/32 | +0.09144 | +0.053 | 20/32 |
| Formation Style | ||||||
| Shotgun | -0.119204 | +0.029 | 27/32 | -0.056192 | +0.029 | 7/32 |
| Under center | -0.046118 | -0.031 | 14/32 | -0.056102 | -0.031 | 13/32 |
| Tempo Style | ||||||
| Huddle | -0.053302 | +0.001 | 22/32 | -0.066262 | +0.001 | 8/32 |
| No huddle | -0.69220 | +0.079 | 30/32 | +0.02532 | +0.079 | 12/32 |
Reading it. EPA/play is noisy: its standard deviation is about 1.4 points a play, so 100 plays carry roughly ±0.14 of noise and a single season's split of a few hundred plays still moves ±0.07–0.10 by chance. Situations also overlap with team quality: good offenses spend more time leading, and teams trailing late face defenses playing soft. Treat a split as a description of what happened, not a trait, until it holds across seasons. Source: nflverse play-by-play, rebuilt daily.