The Jets Won Three Games and Outscored Nobody

The Jets went 3-14 in 2025. That part you probably knew. Here's the part that matters more: our fitted points-per-win model — structural wins, built from point differential rather than the scoreboard's coin flips — put them at 2.8 wins. The actual total was 3. The luck delta is +0.2. In other words, the Jets played almost exactly as badly as their record says they did. There's no hidden competence waiting to be unlocked, no close-game variance to blame. They were a 2.8-win team that won three.

The one-score record was 3-5, which is roughly neutral variance — the universe neither punished them nor rescued them in tight games. When a team goes 3-14 and their close-game record isn't dramatically worse than you'd expect, that's the ledger telling you something uncomfortable: the losses weren't flukes. They were earned.

The point differential was -203, which works out to about twelve points a game on the wrong side. That's not a team that needs a tweak. That's a team that needed a rebuild, and at least the front office appears to have read the same ledger we did.

The honest answer is: not much, and the EPA profile confirms it at every level. On offense, the Jets generated -0.20 expected points added per pass attempt and -0.07 per rush. Both figures are below replacement — the passing number significantly so. A quarterback situation that consumed 1.6% of the cap while producing deeply negative EPA per attempt is about the starkest summary of why this team lost twelve games.

The defense was worse. Opponents generated +0.28 EPA per pass attempt against New York's secondary — that's a number that puts you in competition for the worst pass defense in the league — and +0.01 per rush. The run defense was essentially average, which means the one functional piece of this team was its ability to hold up against the ground game. That's a thin foundation.

The situational numbers don't offer much shelter either. A 37.7% third- and fourth-down conversion rate and -0.21 EPA per play in the red zone tell you this offense wasn't just struggling to move the ball — it was specifically bad at the moments that decide games. Aaron Glenn, in his first season as head coach, agreed with our decision model on 63.4% of go/kick calls, went for it on fourth down at a 13.9% rate against a model-recommended 41.6%, and cost the team an estimated 0.328 wins on decision-making alone. That's a meaningful gap, and it's the kind of thing a first-year coach can close quickly once the data is in front of him. The roster problems underneath those decisions are harder to close.

The departures are worth reading carefully because three of them were offensive linemen. Alijah Vera-Tucker left for New England at $14M APY (1.11 market wins), Connor McGovern went to Buffalo at $13.1M APY (1.03 market wins), and John Simpson signed with Baltimore at $10M APY (0.77 market wins). Add Quincy Williams to Cleveland at $6.5M APY and Micheal Clemons to Indianapolis at $5.67M APY, and the Jets shipped out 3.76 combined market wins. Their cap profile already showed offensive line as one of the three most underfunded position groups by spending z-score (-1.13). Losing three starters doesn't improve that math.

The arrivals are defensive-heavy. Joseph Ossai (ED, Bengals, $11.5M APY, 0.9 market wins), Demario Davis (LB, Saints, $11M APY, 0.85 market wins), David Onyemata (IDL, Falcons, $10.5M APY, 0.81 market wins), and Kingsley Enagbare (ED, Packers, $9M APY, 0.68 market wins) give the front seven legitimate pieces to work with. Dylan Parham arrives from Denver at $8M APY (0.59 market wins) to help patch the interior line — a Band-Aid on a wound that just got larger. The net offseason move on a market-wins basis is +0.07. Essentially flat.

The draft class leads with David Bailey, OLB out of Texas Tech, at pick 2 — 2.52 chart wins, the most valuable asset the Jets have added in this cycle by a significant margin. Kenyon Sadiq, TE from Oregon at pick 16, adds 1.41 chart wins into the most underfunded position group on the roster (TE sits at -1.11 spending z-score). Omar Cooper Jr., WR from Indiana at pick 30, chips in 1.07. D'Angelo Ponds, CB from Indiana in round two, adds 0.82. The later rounds bring Darrell Jackson Jr. (DT, Florida St., 0.44), Cade Klubnik (QB, Clemson, 0.38), Anez Cooper (G, Miami, 0.20), and VJ Payne (S, Kansas St., 0.14).

Three contract verdicts from the new book. Joe Tippmann's four-year, $15.5M APY extension at the right guard spot sits at the 94th pay percentile with production data not yet available — verdict is WATCH, with an exit in 2028 at $4.8M dead. It's fair price discipline for a 25-year-old if the market moves, but it's top-of-book money right now with nothing proven yet. Breece Hall's three-year deal at $14.5M APY is the 98th pay percentile at the position with 75th-percentile production — FAIR PRICE, and the aging multiplier of 0.99 at age 26 says the curve isn't a concern yet. The falsifier is sharp: top-5 production makes it a bargain; falling out of the top 30 makes it an overpay. Minkah Fitzpatrick at $13.3M APY over three years is the 96th pay percentile for the safety position, age 30, verdict WATCH — and the exit clause matters here: $13M dead in 2027 means the Jets don't have a clean way out if the production declines, which the aging curve suggests it will.

The argument for this offseason is simple: the Jets correctly identified that the defense needed bodies, and they went and got some. Davis and Ossai are legitimate starters, Onyemata is a serviceable interior presence, and Bailey at pick two gives the pass rush a potential centerpiece for the next half-decade. If the defensive rebuild is the first chapter of a two-year plan, the moves make sense.

The argument against it is harder to dismiss. The offensive line shed three starters and added one replacement in Parham. The cap profile already flagged OL as a critical spending gap, and the line situation is now more fragile than it was in a 3-14 season. The quarterback position consumed 1.6% of the cap — a number that signals either a bridge or a placeholder, not a franchise answer — while the offense posted deeply negative EPA in every context that matters. You cannot fix a -203 point differential by improving the pass rush without also fixing the side of the ball that generates points.

The Fitzpatrick deal is the exposure question. Thirty years old, $13M dead in year two, at a position where our aging curves say decline is the expected outcome rather than the surprise. That's not an albatross yet, but it's a contract that needs to be right from day one because there's no painless exit if it isn't. The Hall deal is defensible — the aging curve is flat at 26 and the production percentile justifies the spend — but committing top-of-market money to a running back in a rebuild has a track record that the data doesn't love.

Here's the arithmetic, plainly stated. The 2025 point differential puts the structural base at 2.8 wins. The offseason moves net to zero on a production-wins basis — arrivals and departures essentially cancel, and the draft class adds chart value that won't convert to production wins for a year or two. The coaching decision delta is also zero in the model; Glenn's first-year numbers leave room for improvement, but we don't project improvement, we record it. Structural wins for 2026: 2.8.

That is an accounting projection from the ledger — not a betting line, not a prediction, not a pick. It's what the math says when you hold everything we don't yet know equal to zero. The falsifier: if the Jets win six or more games in 2026, this read is wrong and something meaningful changed that the ledger didn't capture — the quarterback situation most likely. If they finish at four or five, the model was in the right neighborhood and the rebuild is still in its early stages. If they finish at two or three, nothing surprising happened either.

The brief shows no vacated opportunity figures or arrival production data to work from, so the target-share arithmetic isn't available here — we don't invent numbers. What the structure of the roster does tell you is where the opportunity will concentrate.

Garrett Wilson sits at 3.1% of the cap — a legitimate market-rate commitment to a receiver — on an offense that generated -0.20 EPA per pass attempt last year. The efficiency floor matters: Wilson's value in any format is almost entirely dependent on the quarterback situation resolving into something functional. Without that, the volume may be there and the production won't follow.

Breece Hall at 3.6% of the cap is the clearest usage-floor argument on the roster. The Jets will run the ball because they have to — the passing game isn't built yet, and Hall is the most expensive skill player on the team. His 75th-percentile production grade means the efficiency is real. In formats that reward volume, that combination is what a usage floor looks like. Kenyon Sadiq, the Oregon tight end drafted at pick 16, enters the most underfunded position group on the team by spending z-score. If he contributes early, he's doing it against a low bar of established competition at the position — that's the rookie sleeper case, and it starts with opportunity, not projection.

Everything above traces to a live instrument, and the receipts stay public: the GM boards carry Aaron Glenn's decision card, the team's cap posture ($84.2M already charged to future caps), and every contract verdict with its falsifier and review date — including Joe Tippmann (WATCH), Breece Hall (FAIR PRICE), Minkah Fitzpatrick (WATCH). When the season grades these reads, the grades post whether they flatter us or not.