> ## Content Index
> Fetch the complete content index at: https://www.statsdigest.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# What Six Weeks of 2026 Actually Tells Us
- URL: https://www.statsdigest.com/what-six-weeks-of-2026-actually-tells-us/
- Published: 2026-10-21T14:38:22.000Z
- Updated: 2026-10-21T14:38:22.000Z
- Description: Six weeks in, EPA data flatters some teams and flags others—but small samples cut both ways.
- Author: Bill Seifert
- Tags: NFL, EPA, 2026 NFL season, 49ers, Chargers, Vikings

## Table of Contents

1. The sample, stated plainly
2. What we think we know (six readings across the league)
3. Three movers worth looking at twice
4. Believe it, or not yet

---

## 1\. The sample, stated plainly

Six weeks of pooled EPA is not a power ranking. It is a quarter-season of plays run against whichever five or six opponents the schedule happened to hand each team, with no opponent adjustment baked in. To show the size of that distortion once: through five games, Minnesota's defense is allowing −0.205 EPA/play across 279 plays — league best, by a lot. But the offenses the Vikings have faced (Green Bay, Chicago, Tampa Bay, Miami, New Orleans) have a plays-weighted average offensive EPA/play of roughly −0.05 themselves, across 1,838 combined plays. That is below league average (which is 0 by construction). The Vikings have been very good; they have also drawn a soft offensive menu. Both are true, and raw EPA cannot separate them. Keep that in mind for every ranking below.

## 2\. What we think we know

**(a) San Francisco's offense is operating at the top of the league.** The 49ers rank first in offense EPA/play through six weeks at **+0.209** across 337 plays, with pass-game EPA/play of **+0.316**. For context, their 2025 full-season offense finished ninth at **+0.061** across 1,180 plays, and their 2024 offense finished at **+0.069**. Brock Purdy's season-to-date line (176 dropbacks, +0.316 EPA/play, +5.1 CPOE, 14 pass TDs on 1,460 yards) is the second-best EPA/play among qualified passers behind Lamar Jackson. The sample is small; the gap to the Niners' recent baseline is not.

**(b) Buffalo's offense is holding its 2025 form, not surging past it.** Bills offense is second at **+0.204** EPA/play on 354 plays. Their 2025 full-season offense finished second at **+0.138** across 1,203 plays. Six weeks in, that's a continuation story: top-tier offense, same tier as last year, essentially the same quarterback output (Josh Allen, 186 dropbacks, +0.247 EPA/play, 1,504 yards, 9 TDs, 4 INTs).

**(c) The Chargers' offense has collapsed, and the collapse is league-worst.** Los Angeles is 32nd (last) in offense EPA/play at **−0.225** across 370 plays — pass EPA/play **−0.268**, run EPA/play **−0.174**. Both sides underwater is the ugly combination. For comparison, the 2025 Chargers offense finished 25th at **−0.035** (1,139 plays), and the 2024 Chargers were 14th at **+0.011** (1,052 plays). This isn't a bad-quarter trailing edge of a 2025 trend; it's a step down from where they ended last year.

**(d) The Vikings' defense is pacing the league in raw terms, with a schedule caveat — and its offense is the league's weakest at the same time.** Minnesota is first in defense EPA/play allowed at **−0.205** across 279 plays, and also allows the league's lowest success rate, 33.2%. The 2025 Vikings finished third at **−0.101** across 1,016 plays, so there's a real echo of last year's identity. But see §1 on opponent quality: this ranking is partly a schedule artifact and shouldn't be read as a power rating. Meanwhile Minnesota's own offense has the league's worst success rate (35.9%) — a defense-carries-everything profile, not a complete team.

**(e) Dallas is bad on defense in the same way they were bad on defense in 2025.** The Cowboys are 32nd (last) in defense EPA/play at **+0.164** across 371 plays, and also allow the league's highest success rate, 49.3% — the EPA number isn't a garbage-time mirage, a second metric points the same direction. The 2025 Cowboys finished dead last on defense at **+0.167** (1,027 plays). Six weeks of 2026 is doing nothing to update that read; it's confirming it.

**(f) Jared Goff has thrown zero interceptions through five games.** 202 dropbacks, +0.298 EPA/play, 1,485 yards, 12 pass TDs, 0 INTs, 51% success rate. He's third among qualified passers in EPA/play behind Jackson and Purdy. For a quarterback whose 2026 is a continuation of a 2025 Detroit offense that finished sixth in EPA/play at +0.082, that INT line is the eye-catcher. INT-free streaks regress; the underlying efficiency is where I'd anchor.

## 3\. Three movers worth looking at twice

**Jacksonville's offense, up.** Jaguars 2024 offense: 18th at −0.020 EPA/play across 991 plays. 2025 Jaguars offense: 14th at +0.031 across 1,128 plays. 2026 Jaguars offense through six weeks: **fourth at +0.155 across 355 plays.** Trevor Lawrence's QB line (194 dropbacks, +0.275 EPA/play, 14 TDs, 4 INTs, 1,480 yards) is the engine. Of the three movers in this section, this is the one where the trajectory has pointed the same direction for two straight baselines.

**Las Vegas's offense, up from a historic floor.** The 2025 Raiders finished 32nd on offense at **−0.215** EPA/play across 941 plays — the worst mark in the league. Through six weeks of 2026 they're 14th at **+0.019** on 378 plays. Kirk Cousins is the active starter in the data: 221 dropbacks, +0.183 EPA/play, 14 pass TDs. That's a very large pooled move from a very small base-rate expectation. I'd put more weight on "they are no longer the league's worst offense" than on "they are a league-average unit"; the latter still needs the second half of the season to defend.

**Tampa Bay's offense, down.** The 2024 Buccaneers offense finished fifth at +0.129 EPA/play across 1,128 plays. The 2025 Bucs slid to 20th at **−0.009** across 1,064 plays. Through six weeks of 2026 they're 29th at **−0.130** across 374 plays — pass EPA/play **−0.246**. That's three straight baselines pointing the same direction and the gap is now large. Of the trio in this section, this is the one I'd be slowest to dismiss as sample noise.

## 4\. Believe it, or not yet

A judgment on evidence strength, not a prediction on records.

- **San Francisco offense: believe it, with caveats.** The gap to the 2025 baseline (+0.061 vs. +0.209 now) is more than two-tenths of EPA per play — enormous for six games. Purdy's efficiency is unlikely to be entirely a schedule illusion. What would kill it: a comparable sample against a tougher defensive slate that compresses the number back toward +0.10.
- **Chargers offense: believe it as a problem, not yet as a floor.** Both pass and run underwater, a two-season slide from 14th to 25th to 32nd. 370 plays is enough to call this real. What would walk it back: a 100-play stretch against lesser defenses that pulls pass EPA out of the −0.20s.
- **Vikings defense: hold.** A strong number from an already-good unit (3rd in 2025), but the opponent slate has been genuinely soft. What would confirm it: holding sub-−0.10 through the next three or four weeks against stiffer offenses. At 279 plays, it's the smallest defensive sample in the league right now — and the companion fact that Minnesota's own offense is the league's least successful unit is a reminder the whole team picture is lopsided, not just efficient.
- **Jacksonville offense: tentatively believe it.** Two consecutive baselines pointing up, plus a QB efficiency number consistent with the team total. What would kill it: the pass game collapsing against a defense that pressures Lawrence — the 2025 full year was only +0.031, so there's a long way to fall back.
- **Cowboys defense: believe it as continuity.** Last in both EPA/play and success rate allowed in 2025, last in both again in 2026\. There's no update to make here; it's the same defense, corroborated two ways.
- **Tampa Bay offense: believe the direction, suspend judgment on the level.** Three straight seasons of decline — +0.129 → −0.009 → −0.130 — is a trajectory, not a bad week. But the step from −0.009 to −0.130 is the largest of the three and has had the least time to be tested. What would confirm it: another three weeks under −0.08\. What would kill it: a return to roughly league-average by Thanksgiving.

---

Six weeks is enough to see outlines. It isn't enough, by itself, to repaint a full prior season. Where the two agree — Dallas's defense, Buffalo's offense, Jacksonville's slow upward slope — the signal is strongest. Where they disagree sharply — the Chargers' collapse, the Niners' leap, the Raiders' rebound — the number is interesting precisely because it's unresolved. That's the honest state of the season through week 6: a lot of promising reads, almost all of them still inside the margin where another month of games can rewrite them.