Pittsburgh Bets on Old Bones and New Blood

Pittsburgh went 10-7 last season, which sounds like a team that figured something out. The point differential says something more modest. A +10 margin — plug that into our fitted points-per-win model and you get 8.8 structural wins. The Steelers won ten. The gap between those two numbers is 1.2 wins, which is the luck delta, and it came almost entirely from the one-score game record: 7-3 in games decided by eight points or fewer. Close games are roughly coin-flip territory for even good teams. Going 7-3 in them is a nice result; it is not a skill set you can invoice.

That isn't a criticism — it's an observation that matters enormously for what comes next. The most reliable predictor of a team's next-season win total isn't last year's record. It's the point differential underneath that record. Pittsburgh's differential says they were an 8.8-win team dressed in a 10-win suit. Factor in offseason moves and coaching changes and the structural baseline for 2026 barely moves. The Steelers are not in a collapse scenario, but the bounce-back narrative runs the other direction here: it's last year's result that needs explaining, not next year's projection.

The EPA profile from 2025 tells you exactly what kind of team this was: competent, unspectacular, and held together by the defense. On offense, Pittsburgh produced +0.05 EPA per pass attempt and +0.01 per rush — both in positive territory, which means they weren't actively hurting themselves, but neither number is going to make a coordinator jealous. The passing game was functional. The run game was essentially break-even. They converted 44.3 percent of third and fourth downs, which is useful, and they posted exactly 0.00 EPA per play in the red zone, which is the statistical equivalent of a shrug.

The defense is where the real work happened. Opponents generated +0.04 EPA per pass attempt against Pittsburgh — that's a positive number, meaning the passing defense gave up some value — but the run defense was genuinely good at -0.01 EPA per carry allowed. Stop the run, bend-don't-break passing coverage, let the offense not lose you the game. That is a coherent identity, even if it's not a blueprint for blowing anyone out, which the +10 point differential reflects.

Then-coach Mike Tomlin — who is no longer with the organization — posted a decision index of 0.422 expected wins lost per season. His go-for-it rate on fourth down was 17.2 percent against a model-recommended rate of 38.4 percent. He agreed with the optimal decision 75.8 percent of the time. That's not catastrophic, but it's a consistent, quantifiable drag — roughly half a win a season — that was baked into every Pittsburgh result during his tenure.

The most significant arrival isn't on the free-agent list — it's on the sideline. Mike McCarthy takes over as head coach with a career expected-wins-lost rate of 0.30 per season, compared to his predecessor's 0.49. The coaching change alone is worth an estimated 0.19 wins before anyone touches the ball, simply because McCarthy's historical decision-making rate is better than what Pittsburgh was running. That's the kind of quiet upgrade that doesn't generate a press conference but shows up in the standings.

In the secondary, Jamel Dean (CB, from Tampa Bay, $12.25M APY) is the biggest free-agent bet — the market is paying him for 0.96 wins of value. His production data wasn't available for this ledger, so that's a price-based projection only; the production will need to validate it. Rico Dowdle arrives from Carolina at $6.125M APY, where market value clocks at 0.43 wins but his production history supports only 0.04. That's a contract that leans heavily on upside. Sebastian Joseph-Day (IDL, $5.5M) and Jaquan Brisker (S, $5.5M) round out the arrivals at reasonable prices, though production data for both is absent from the ledger.

The notable departure is Isaac Seumalo, the left guard who heads to Arizona — a player commanding $10.5M APY on the open market. Pittsburgh replaced him with nothing identifiable in the free-agent ledger and instead drafted two offensive tackles: Max Iheanachor in the first round (pick 21, 1.2 chart wins out of Arizona State) and Gennings Dunker in the third (pick 96, 0.51 chart wins out of Iowa). The interior offensive line question becomes a rookie answer, which is a gamble. Also gone is Kenneth Gainwell (RB, to Tampa, $7M APY), who actually produced — 0.09 wins last season — more than Dowdle's 0.04.

The contract room has two entries worth watching closely. Nick Herbig (LB) signed a four-year, $25M APY deal that sits at the 100th percentile of linebacker pay — literally the top of the market. Without production percentile data in the brief, this is graded on price discipline alone: WATCH. The exit cost is $15.7M dead in 2027, so the commitment is real. Aaron Rodgers signed a one-year deal at $22.5M APY — 92nd percentile pay for a quarterback whose 2025 production ranked at the 55th percentile, at age 43. That spread is rated OVERPAY, and the falsifier is clear: a top-5 QB production season makes this price fair and this verdict wrong. Cameron Heyward (IDL) gets a one-year deal at $18M APY, 96th percentile money for an interior lineman who turns 37 in season — rated WATCH, with a $7.5M dead cap exit in 2027. Michael Pittman Jr. (WR) comes in at $17.5M APY on two years, rated FAIR PRICE at age 28 with a 0.97 aging multiplier — the math says he should hold most of his production through the deal.

The full draft class totals 4.47 chart wins across ten picks. Second-rounder Germie Bernard (WR, Alabama, pick 47, 0.85 chart wins) addresses a real need alongside Pittman. Third-rounder Drew Allar (QB, Penn State, pick 76, 0.63 chart wins) is clearly a developmental piece behind Rodgers — and worth watching as the longer-term plan at the position. The rest of the class — corners, receivers, a tight end, a fullback — fills depth at a reasonable draft-chart price.

The cap posture tells you where Pittsburgh decided to spend and where it didn't. EDGE is at a +2.32 standard deviation above the league mean — T.J. Watt at 13.1 percent of the cap and Alex Highsmith at 6.3 percent represent a massive bet on pass-rush as the franchise's structural advantage. The linebackers are nearly as top-heavy at +2.25, which is where the Herbig extension lands. Offensive line, meanwhile, sits at -1.7 standard deviations — the lowest-invested position group on the roster, in a season when the starting left guard just left and the replacements are rookies. That is a visible tension.

The net offseason ledger on a production basis is -0.05 wins — essentially flat, with losses and gains roughly canceling. On a market basis it's +0.82, which means Pittsburgh paid for more improvement than the production history suggests it received. That gap between market and production value is where teams get into trouble in year two of deals, and there are a few players here — Dean, Dowdle, Herbig — where the production validation is still outstanding.

The aging exposure is real and concentrated. Rodgers is 43, Heyward is 37, and our aging curves don't favor either at their current price points. The Pittman deal is the cleanest of the big contracts — a 28-year-old with a 0.97 aging multiplier at a price the model rates fair. The rest of the veteran core asks you to believe that age either won't matter or won't matter this year. Sometimes that's true. It's not the way to bet, and the ledger knows it.

Here is the arithmetic, stated plainly: Pittsburgh's 2025 point differential projects to 8.8 structural wins as a base. Add the offseason's net production impact — a slim -0.05 wins — and add the coaching decision upgrade worth an estimated +0.19 wins as McCarthy replaces a coach who left 0.49 expected wins per season on the field with a coach whose career rate is 0.30. The sum is 8.9 structural wins for 2026.

That is an accounting projection from the ledger — not a betting model output, not a pick. It's what the numbers on the page add up to before a single game is played. The falsifier: if Pittsburgh wins eleven or more games in 2026, this read is wrong, and the Rodgers-era offense and the Dean secondary both deserve credit. If they win fewer than seven, the aging exposure and the offensive line gamble were bigger liabilities than this projection assumed. Anywhere between seven and ten wins, the model did its job.

The departure of Kenneth Gainwell removes a receiver out of the backfield — 0.09 production wins on a $7M market contract, which means the market had noticed him. The player positioned to absorb that role is Rico Dowdle, who logged 50 targets and 216.3 PPR points in 2025, a rate of 4.33 PPR points per target. That per-target efficiency is genuinely useful, and at $6.125M APY he's priced below what a back with that kind of receiving production typically commands. The opportunity is real; the efficiency is real; the price hasn't fully caught up. That's the sleeper case, stated without decoration.

The receiver room added Michael Pittman Jr. and draft pick Germie Bernard, which gives Rodgers more credible targets than he had access to in the final chapter of his career in New York. Pittman's 75th-percentile production history at 96th-percentile pay is still a fair deal — but from a fantasy standpoint, the more interesting question is how much of the target share a 43-year-old quarterback distributes downfield versus checkdowns, and the data doesn't answer that yet. Bernard at pick 47 has the draft capital to earn early snaps; whether the offense uses him that way in year one is the variable.

The aging flags here are straightforward. Rodgers at 43 is a fade in formats where ceiling matters — the 92nd-percentile price on a 55th-percentile production history at his age is an uncomfortable combination. Heyward at 37 is similarly thin as a DFS play, though his context in the defensive line is more protected by the system around him. The safest entry point on this offense is Dowdle: known efficiency, vacated opportunity, and a price that hasn't priced in either.

Everything above traces to a live instrument, and the receipts stay public: the GM boards carry Mike McCarthy's decision card, the team's cap posture ($132.4M already charged to future caps), and every contract verdict with its falsifier and review date — including Nick Herbig (WATCH), Aaron Rodgers (OVERPAY), Cameron Heyward (WATCH), Michael Pittman, Jr. (FAIR PRICE). When the season grades these reads, the grades post whether they flatter us or not.