Week 7 of the 2026 season is in the books: 14 games, from Thursday's Chicago win over New England to Monday's Dallas blowout in Philadelphia. The league has now played 107 of 272 regular-season games — 39% of the schedule — enough of a baseline that a single week's results can start to be measured against something real. But one week is still roughly 60 plays per team, and most of Sunday's scorelines are just that: clean final numbers that haven't been pulled apart yet. Here is the one result from Week 7 we can say something rigorous about, and an honest accounting of what the rest of the slate does and doesn't tell us.

The one performance that moved the needle: Joe Burrow in Baltimore

Cincinnati beat Baltimore 35-13, and the passing-game split explains almost the entire margin. Burrow went 22-of-31 for 292 yards and four touchdowns, posting +0.587 EPA per dropback across 31 attempts — a total of roughly +18.2 EPA on those plays alone. For comparison, the 2026 league has averaged +0.042 EPA per pass play across 7,627 passes so far this season. Burrow's day was roughly fourteen times that league baseline, sustained over 31 attempts rather than a couple of explosive outliers — which is what makes it a genuinely informative single-game number rather than noise.

The context that sharpens it further: Baltimore's defense had actually been one of the better units in the league through six weeks, allowing -0.033 EPA per play across 372 snaps — stingier than the 2026 league average of +0.008 EPA/play overall. Burrow's afternoon (worth about +16.9 total EPA on 51 Cincinnati snaps) was large enough by itself to drag the Ravens' full-season mark up to +0.011 allowed per play across 423 snaps. That's not a defense that was already struggling getting picked on; that's a defense that had been performing well getting torched by one quarterback in one game.

Baltimore's Tyler Huntley, by contrast, needed 38 dropbacks to produce +0.091 EPA per play (23-of-38, 298 yards, one touchdown) — a middling day that did nothing to keep pace. Cincinnati's own run game was unremarkable (-0.065 EPA per carry on 20 carries), and Baltimore's run game was outright bad (-0.349 EPA per carry on 27 carries). This was a game Cincinnati won almost entirely through the air, against a defense that came in playing well.

What to watch in Week 8

• Does Burrow's day hold up as a game, or build into a stretch? +0.587 EPA/dropback on 31 attempts is one of the best single-game passing numbers the league will produce all season. One more month of it would be a real story; one more month of Baltimore's defense bouncing back toward its pre-Week-7 form (-0.033 EPA allowed per play) would be, too.
• Was Kansas City's shutout a defensive statement from Seattle or an off night from an offense that had been performing well? The six-week trend line (+0.155 EPA/play) says Kansas City was fine coming in. The scoreboard says something went wrong on Sunday. Only the play-by-play from that specific game will say which.
• How much of Week 7's margin around the league was passing-game variance versus real separation? With 39% of the season complete, this is close to the point where six-game and seven-game samples start to mean something — but a single week's scorelines, on their own, are not that evidence.

A closing caveat worth repeating: none of the EPA figures above are opponent-adjusted, and the 2026 league-wide baselines used for comparison (+0.008 EPA/play overall, +0.042 per pass, -0.038 per run, across 13,179 plays so far) are themselves still being built week by week. Treat everything here as what the numbers show happened — not as a verdict on what will.