Is Jacksonville the best team in football?
Through week 10 of 2026, Jacksonville ranks first in the NFL in opponent-adjusted net EPA per play (+0.276), 5th on offense and 3rd on defense. But the lead over the San Francisco 49ers is 0.014 EPA per play: resample each team’s games and Jacksonville comes out ahead 50% of the time, and first overall only 42%.
- Treat Jacksonville and San Francisco as co-favorites, not a clear #1. 8 games isn’t enough to separate teams this close.
- Expect the Jaguars to stay near the top. The week-10 efficiency leader finished the regular season in the top three in 9 of the last 10 seasons and made the playoffs every time.
- Don’t read it as a Super Bowl pick. None of those 10 leaders won the title; 2 reached the game.
The race at the top, week by week
How Jacksonville got here
The Jaguars are 6–2 through week 10. Their offense ranks 5th in expected points added per play and their defense 3rd in fewest allowed, once each opponent’s strength is accounted for. The schedule hasn’t flattered them: the opponent adjustment moves their net EPA/play by 0.0005. San Francisco (7–1, +0.262) gets there with the 1st-ranked offense and the 6th-ranked defense.
How sure is “first”?
Not very. Jacksonville has played 8 games and San Francisco 8, and a single blowout moves a season-long average. To see how much, every team’s games were resampled with replacement 2,000 times and the rankings recomputed. Jacksonville finished ahead of San Francisco in 50% of those, which is a coin flip, and finished first in the whole league in 42%: less likely than not.
What history says about the week-10 leader
The NFL's #1 team in opponent-adjusted net EPA/play through week 10, each season, and how it finished. Rank is the same measure at the end of the regular season.
Across 2016–2025, the leader after week 10 finished the regular season first 3 times and in the top three 9 times; every one made the playoffs, 2 reached the Super Bowl, and none won it. One of them was the 2017 Jaguars, who finished first. Efficiency this far into a season is a good guide to the rest of it, about as good as the standings: across 320 team-seasons, net EPA/play through week 10 correlated with rest-of-season point differential at r = 0.49, and win percentage at 0.47. The playoffs are another matter: single-elimination football is short and noisy.
Method, data, and limits
Data. nflverse play-by-play (CC-BY 4.0), regular season, garbage time removed (4th-quarter plays with the offense’s win probability below 10% or above 90%). The article’s numbers are as of week 10 of 2026 and don’t change; the chart above keeps updating.
Net EPA per play is offense EPA/play minus defense EPA/play allowed. Opponent-adjusted numbers subtract the strength of the opponents each team actually faced, estimated with a ridge regression on every play (see the team efficiency tool for details and its validation). Model version: NFL opponent adjustment v1.0.0 (experimental: validated before the backtest harness existed; its harness port is pending).
The resampling draws each team’s games with replacement (same number of games, 2,000 times, fixed seed) and recomputes unadjusted net EPA/play; adjustment changes the top teams’ numbers by almost nothing this season, and resampling the full ridge fit would add little but run time.
What it can’t tell you. Injuries, quarterback changes, and what’s left on the schedule aren’t in these numbers. Ten seasons is a small history: “none of ten won the Super Bowl” says the playoffs are noisy, not that the leader is cursed.
Code:nfl/jacksonville/ · Data: nflverse (CC-BY 4.0). Sparvi is not affiliated with the NFL or any club.