Ranking Defenses with the Negative NEPA

EPA/NEPA attaches a point value to every play based on whether that play increases or decreases a team's expected point total. Plays with a positive NEPA are good for the team on offense, plays with a negative NEPA are bad.

That being said, good defenses should force a lot of negative NEPA plays and prevent positive NEPA plays. The chart below aggregates just that. For each defense I have listed the percent positive NEPA on run plays, pass plays, and total. - I want to emphasize that this is not the same thing as a negative play or a tackle for a loss. For example, 5 yards on 3rd and 12 would earn a team a negative NEPA because their expected point total would go down. A majority of plays are negative (even when Johnny Football has the ball), but there are more large and positive NEPA plays than negative (for the distributionally inclined, the distribution is positively skewed); the average play has a NEPA of 0.



Michigan State and Florida State can each boast that fewer than a third of plays against them were positive this season. Four teams from the SEC - Alabama, Florida, LSU and Vanderbilt - are in the top 15. Don't expect to see a lot of fireworks from Virginia Tech and Rutgers in the Russell Athletic Bowl; neither team is offensively gifted and they both have a penchant for hurting offenses.

The teams at the other end are bad, bad football teams. But at the very bottom is Navy, a significantly better team than UMass, New Mexico and their ilk. But Navy allowed 68% completions and 4.5 yards per carry - not a lot of negative NEPA play potential in those stats. That being said, their 53.5% positive NEPA play percentage is utterly terrible.