Muffed.ai 2025 QB Rankings — A Data-First Take on All 32 Starters

A 22-year-old quarterback in his second NFL season led the league in CPOE at +9.14, produced +12.88 QB-only Total EPA per game, and led the league in success rate at 55.2%. He took 47 sacks (a 7.8% raw sack rate) but his pressure-to-sack rate of 25.0% is league-average — the volume is a Patriots O-line problem, not him. The Patriots earned a playoff spot. This is the cleanest #1 we could draw up.

Quiet excellence. 15 games, +9.92 QB-only Total EPA per game, +0.299 EPA per dropback (second only to Maye), +3.82 CPOE, the league's #2 late-close EPA per dropback at +0.377, and a strong 15.2% pressure-to-sack rate. Green Bay went to the playoffs as a wild card. Love had a top-2 QB season and almost no one is talking about him.

Only 9 starts but elite efficiency: +0.236 EPA/dropback, +5.07 CPOE (2nd in the league), 12.2% pressure-to-sack rate (tied for best in the league with Caleb Williams), and a +0.393 late-close EPA per dropback that's the best in the league among the 32. Sample-size caveat: 316 dropbacks is on the lighter side. The rate stats are robust.

Threw 46 touchdowns (most in the league) with only 8 interceptions. His CPOE of +1.48 isn't elite, but his +0.260 QB-only EPA/dropback and 52.7% success rate are. The Rams' offense is a deep-throw machine — his average intended air yards of 9.10 is tied for fourth in the league with Drake Maye (behind Mariota, McCarthy, and Jackson) and his +1.02 air yards to sticks is second-largest (behind only Mariota's +1.20). He played all 17 games at age 38. Worth it.

30 TDs, 10 INTs, +4.45 CPOE on 656 dropbacks (second only to Bo Nix's 674). Quietly had a top-5 QB season even though Dallas finished 7-9-1 and missed the playoffs — the Cowboys' defense and special teams were what dragged the season down, not Dak. He didn't have peak Dak's deep-ball variance — his intended air yards (8.28) are more conservative than Stafford's — but he was efficient where it mattered.

14 rushing TDs, 599 rushing yards on 94 carries. His rushing alone added +48 EPA. The passing efficiency was good not great this year (+1.55 CPOE). His pressure-to-sack rate of 26.9% is mid-pack — he holds the ball some, but the volume of value he adds with his legs compensates.

Career reset in Indianapolis. 13 games, +7.48 QB-only Total EPA per game, +3.38 CPOE (6th), 100+ passer rating, and 38 carries for 168 yards and 5 rushing TDs. His late-close EPA was the worst part of his profile (-0.035 on 100 plays). Still a clear top-10 QB by 2025 production.

34 TDs, 8 INTs, played all 17, and put up +0.236 QB-only EPA per dropback. The only concerning number is late-close at -0.175, which is a small sample but a real pattern this year — Detroit went 9-8 and missed the playoffs after losing tight games down the stretch. Note: GB (9-7-1) took the last NFC wild card on win percentage tiebreaker (.559 vs .529), so DET is out.

Did not regress from his 2024 breakout the way many predicted. +0.191 QB-only EPA per dropback, +4.31 CPOE (4th in the league) on 477 attempts. Threw 25 TDs to 14 picks, played all 17 games, and Seattle won the NFC West. His +0.207 late-close EPA per dropback is also very good. The Vikings probably regret letting him go.

The CPOE collapse to -2.90 is real and matters; his total production stayed top-10 because he still adds value with his legs (56 carries, 429 yards, 5 TDs) and because his floor is so much higher than most. KC's offense as a whole was 13th-best by team pass EPA per attempt — not what we're used to. Note: he missed the final three regular season games with an ACL tear, hence 14 games played.

Only 8 games before injury. His +2.06 CPOE and 17:5 TD-to-INT ratio say he was on track for a normal Joe Burrow season; his +0.129 QB-only EPA per dropback is below where his ceiling sits on these accuracy numbers, and Cincinnati's offense ranked 24th in team pass EPA per attempt overall (Flacco started most of the rest). Hard to read Burrow's small-sample line with any confidence. He's ranked where the production was; the talent says higher.

14 games, +0.152 QB-only EPA per dropback, +0.77 CPOE, and a more involved running game than in years past (38 carries, 214 yards). The Texans' offense was 15th in team pass EPA per attempt — squarely middle of the pack. Reasonable bounce-back season after a noisy 2024.

Pressured 36.3% of the time — the second-highest rate in the league behind only Shedeur Sanders. When pressured, he gets sacked 24.2% of the time, which is league-average — so the 54 sacks he took were overwhelmingly because the LAC O-line lets him get pressured constantly, not because he was holding the ball. He had +3.31 CPOE (8th) and 510 rushing yards (more than Lamar) on 74 carries. With a competent O-line his composite is a clean top-10. The pressure-adjusted rank of 13th already builds in partial credit — without it, he'd be in the high teens.

Threw fewer TDs (25) and ran for more (8 rushing TDs, 94 carries, 436 yards) than the narrative around him suggests. His +3.09 CPOE is genuinely good. But his +0.131 QB-only EPA/dropback is middling and his 44.3% success rate is below average — he's a high-leverage QB whose value is concentrated in red-zone runs and a few explosive throws. The Eagles' offense was 14th in team pass EPA per attempt this season, so it's not a 2024 juggernaut anymore. This is roughly what we'd expect Jalen Hurts to be in a healthy year.

Denver's offense ranked 9th in team pass EPA per attempt, helped by Nix's 12.3% pressure-to-sack rate (3rd-best in the league, behind only Purdy and Caleb Williams at ~12.2%) — when he's pressured, he gets out of trouble. CPOE was -2.07 (28th in the NGS sample), which is the main thing keeping him out of the top 10. Most dropbacks in the league at 674.

29 passing TDs, 9 rushing TDs (tied with Jaxson Dart, behind only Josh Allen's 14). CPOE of -2.73 says the underlying accuracy isn't there yet. QB-only EPA per dropback of +0.129 is replacement-plus, not star. He played all 17 games. Coen's offense let him cook on designed runs.

Baltimore went 8-9 and missed the playoffs; Lamar started 13 of those games. The model is not “wrong” about Lamar — his per-game and per-dropback numbers really were below average this year, dragged primarily by a 29.0% pressure-to-sack rate (4th-worst) and his -2.23 CPOE. The pressure adjustment helped him a meaningful amount (he was #20 before we credited part of his sack volume to the OL); it didn't elevate him into the top 10. His ceiling and his floor are still higher than this in his career. 2025 just wasn't the year.

This is the v1.1 model's biggest gainer. +3.37 CPOE (7th in the league) on 552 dropbacks behind a chaotic Cardinals OL that pressured him 32.8% of the time (7th-highest pressure rate). When pressured, he converted to sack at a league-average 23.8%. He completes throws above expectation; the throws just don't generate enough value because Arizona's offense around him was stuck. Murray missed half the year.

Stepped in for Jayden Daniels (188 attempts to Mariota's 227) and was perfectly fine in a difficult spot. His +3.29 CPOE on 227 attempts is real and his late-close EPA was decent. The composite is dragged by a small +0.085 EPA per dropback. Career backup who didn't sink the season.

26 TDs and 11 INTs, but his per-dropback efficiency dropped to +0.108 and CPOE to -1.61 from his 2024 highs. His rushing line is interesting: 42 of his 47 carries were scrambles, and his total rush EPA of +41.3 (3rd among QBs) is doing a lot of the lifting. He plays all 17 games. He's exactly what he is.

Here's where the rankings get controversial. Williams threw for 27 TDs and only 7 INTs and posted +0.140 QB-only EPA per dropback — those are real numbers. But his CPOE of -6.88 is the worst of any QB in our sample with at least 200 attempts. His pressure-to-sack rate of 12.2% is tied with Purdy for best in the league — meaning when he's pressured he almost never gets sacked. That's good pocket movement, but it's also why his sack rate was deceptively low. The Bears' offense was 10th in team pass EPA per attempt, and Williams' time to throw of 3.20s was the second-highest in the league behind only Shedeur Sanders — he is holding the ball forever, hunting big plays. It's working some of the time. It is not a sustainable shape for a top-10 QB.

Rookie. 9 rushing TDs on 41 designed carries (3rd-most in the league, tied with Justin Fields and behind Hurts and Allen), and a +41.4 total rush EPA that essentially ties Mayfield for 2nd among QBs. But -2.03 CPOE, 28.2% pressure-to-sack rate (5th-worst), and a brutal -0.140 late-close EPA per dropback. The Giants' offensive scheme was clearly tilted around his legs because the throwing wasn't there yet. He'll be more in the conversation next year.

Took over from Kirk Cousins (Penix 290 attempts to Cousins' 269) and was learning on the job. 9 TDs, 3 INTs, -3.20 CPOE. The intended air yards (8.82) and aggressiveness (13.8%) say he was being asked to push the ball downfield, which is part of why CPOE is rough. Reasonable rookie growth profile.

Rookie, threw 10 TDs to 6 INTs in 11 games. +0.075 QB-only EPA per dropback and a 31.6% pressure-to-sack rate that's near the bottom. Saints are rebuilding. Shough is fine as a development arc.

20 TDs to 15 INTs is a brutal ratio. -0.32 CPOE — actually below the NFL average for the first time as a starter. 14 games. Miami's offense collapsed around him and our previous Muffed Short corrected a hot take that his CPOE was positive — it was not. This is what it looks like when the YAC machine breaks down.

24 TDs, 7 INTs, but +0.079 QB-only EPA per dropback and a 43.0% success rate that's bottom-10. His 5.95 average intended air yards is the shortest among the 32 starters — he is not pushing the ball downfield at all, and his -3.08 air yards to sticks is the most “below-the-marker” in the sample. He was pressured only 21.0% of the time (the lowest in the league — Pittsburgh's O-line protected him well), and he still couldn't generate value. End-of-career season.

Year three is supposed to be the leap year. Young's +0.041 QB-only EPA per dropback and 43.8% success rate suggest the leap is still pending. -0.29 CPOE is right at league average; the rest of his profile is below. His 16.3% pressure-to-sack rate is actually good — pocket movement isn't the problem. He's just not generating enough yards per attempt.

Started 9 games before being benched. 7 TDs, only 1 INT (good!), but +0.033 QB-only EPA per dropback, 29.3% pressure-to-sack rate (3rd-worst), and a brutal -0.229 late-close EPA per dropback. His designed runs were the only thing keeping him from being deeper in this list — 41 designed runs for 387 yards and 4 TDs. The Jets cycled through three QBs (Fields, Cook, Taylor).

This is the steepest year-over-year decline in the league. 19 TDs to 17 INTs. -0.032 QB-only EPA per dropback. The Raiders pressured him on 33.2% of dropbacks (6th-highest); when pressured, he converted to sack at 31.8% — the worst rate in the league among 200+ dropback QBs. So even with the OL bailout, this is a Geno problem too. CPOE was still positive (+0.77) — he was throwing accurately into hopeless situations. This is what playing on a 2025 Raiders team does to a QB profile.

Rookie (after the lost 2024 to injury). -0.066 QB-only EPA per dropback, -5.16 CPOE (35th in the NGS sample), 27.6% pressure-to-sack rate. Threw 11 TDs to 12 INTs. The Vikings' offense was 29th in team pass EPA per attempt. McCarthy was given the keys; the keys didn't fit.

Hard rookie year for the #1 overall pick. -2.42 QB-only Total EPA per game across all 17 starts means Tennessee played him through the pain. 15 TDs, 7 INTs, -3.05 CPOE, 37.9% success rate (2nd-worst). He took 55 sacks. The good news: he played every game and the tape is there to coach off of for year 2.

The worst per-dropback efficiency of any 200+ dropback QB in 2025: -0.157 QB-only EPA per dropback, 36.8% success rate, -3.63 CPOE, 7 TDs to 10 INTs in 8 games. He was pressured on 45.5% of his dropbacks — the highest rate in the league by 9 percentage points — and his -0.476 late-close EPA per dropback is in a class of its own (the next worst is McCarthy at -0.265). His 3.23s time to throw was the slowest in the league. Cleveland gave him a real opportunity and the production was the worst in the sample.