Denver Won Fourteen Games and Outplayed Eleven of Them

Denver went 14-3 last season, and the instinct is to hand them a championship-contender badge and move on. But every record deserves a second look, and this one is interesting. The Broncos' +90 point differential — applied to our fitted points-per-win formula — projects to 11 structural wins. They actually won 14. That three-game luck delta is the gap between what they earned on a play-by-play basis and what the scoreboard handed them, and it's the most important number in this piece.

A lot of that gap lives in the one-score column. Denver went 11-2 in games decided by eight points or fewer. That's a remarkable record, and it reflects real things — situational discipline, good moments from the quarterback, a defense that doesn't flinch late. But it also reflects variance. Coin flips don't always land the same way twice, and an 11-2 mark in close games is the kind of thing that tends to drift back toward the middle. The structural read isn't that Denver was lucky, exactly. It's that a genuinely good team won some tight ones that could have gone either way, and next year the math won't spot them three wins in advance.

The defense is the foundation, and it's a good one. Opponents produced -0.13 EPA per passing attempt and -0.04 EPA per carry against Denver last season — both figures well into negative territory, meaning the Broncos were actively taking value away from offenses on both sides of the ball. That's a complete defensive profile. Interior linemen Zach Allen and D.J. Jones anchor the run defense, and they didn't need the pass rush to carry the whole load.

The offense is the more complicated story. At +0.10 EPA per passing attempt, Bo Nix and this passing game were genuinely positive — not elite, but comfortably above neutral. The run game at -0.02 EPA per carry is essentially break-even, which is fine as long as the pass carries the weight. Third-down conversion at 45.1% and a red-zone EPA of +0.10 per play suggest an offense that does its job when the field shrinks. None of this is flashy. Most of it held up.

Sean Payton's decision index deserves a mention here because it cuts against the otherwise clean story. Payton agreed with our model's recommendations on 76.6% of fourth-down and game-management decisions, went for it on 4th down at a 12.6% rate against a model-recommended 34.2%, and cost the team an estimated 0.313 expected wins through those choices. That's not catastrophic — roughly a third of a win over seventeen games — but it's a consistent leak from a coaching staff that wins in almost every other dimension. An aggressive defense paired with a conservative fourth-down philosophy is an odd combination, and it's the one place the numbers say Payton leaves something on the field.

The offseason ledger is short, and that's the story. Denver didn't add anyone. They lost three players: edge defender John Franklin-Myers departed for Tennessee on a deal the market values at 1.72 wins per year, guard Dylan Parham went to the Jets at 0.59 market wins, and linebacker Dre Greenlaw returned to San Francisco at 0.42 market wins. None of the three have production-win numbers attached — the brief doesn't have them — so we're working from market price alone. The total outflow is 2.73 market wins with nothing coming back.

The draft class is modest in scale, which is typical for a team that wins fourteen games and doesn't pick until Round 3. Third-round defensive tackle Tyler Onyedim out of Texas A&M carries 0.76 chart wins — the most valuable pick in the class by a wide margin. Running back Jonah Coleman (Washington, Round 4) comes in at 0.39 chart wins, tackle Kage Casey (Boise State, Round 4) at 0.38, tight end Justin Joly (NC State, Round 5) at 0.29, and three seventh-rounders — safety Miles Scott, tight end Dallen Bentley, and linebacker Red Murdock — at 0.13 each. The class totals roughly 2.08 chart wins, almost entirely dependent on Onyedim delivering at or above expectation.

Denver also has a modest compensation-pick projection worth 0.64 wins in expected 2027 draft value — a Round 4, a Round 6, and a Round 7. That's future capital, not present-tense help. Net it all together and the production-basis offseason change is essentially zero: no new arrivals to count, no quantified production leaving. The market-basis number is -2.73 wins — a real number that reflects real players gone — but the production basis is what drives the structural model forward.

The honest argument here is that Denver didn't hurt themselves badly, but they didn't help themselves either. The roster that won 14 games is mostly intact in terms of production accounting. The concern is what's underneath the surface of the cap sheet. Right tackle Mike McGlinchey is eating 7.8% of the cap. Allen and Jones together account for 10.2%. Tight end Evan Engram sits at 4.7% of the cap and is the most glaring albatross on the roster: 23rd percentile in production, 98th percentile in pay. That's a contract that looks worse every time you open the spreadsheet.

Running back J.K. Dobbins is a softer version of the same problem — 41st percentile production against 74th percentile pay. Neither is a cap emergency on its own, but together they represent money that isn't returning value, and that money can't be spent elsewhere. Denver carries $152.6 million in future proration — dead cap and contract obligations rolling forward — which constrains how aggressively they can address the holes Franklin-Myers and company left behind.

The one genuinely bright spot in the cap picture is Bo Nix. He's producing at the 81st percentile for quarterbacks while sitting at the 44th percentile in pay, with a cap hit of just 1.7%. A quarterback on a rookie deal who plays well is the single best financial advantage in football. Denver's window is as long as that advantage lasts. The question is whether the surrounding cast — minus Franklin-Myers on the edge, with Engram's contract parked on the books — is good enough to take full advantage of it.

Here's the arithmetic, stated plainly. Denver's 2025 point differential projects to 11 structural wins. The offseason adds zero production wins — no meaningful arrivals, and the production basis on the departures isn't quantified in our data. The coaching staff is unchanged, and Payton's decision index doesn't move the needle up or down relative to last year. Eleven plus zero plus zero equals 11. That is the structural projection for 2026.

This is an accounting projection from the ledger — not a pick, not a forecast in the probabilistic sense. It's what the numbers say the team earned, adjusted for what they added and subtracted. The falsifier is straightforward: if Denver wins fewer than nine games, the defense regressed beyond what the departures explain, or Nix took a step back, or both. If they win 13 or more again, the close-game luck ran hot a second consecutive year, which happens — it's just not the base case. The number that would prove this read wrong in the other direction is 13 wins without a meaningfully improved point differential. Nine wins or fewer, and the defense or quarterback story changed in ways the offseason ledger didn't predict.

The brief doesn't surface vacated opportunity numbers for Denver — no departed receiver or tight end left a clean target share on the table that we can quantify with confidence. What the data does show is an offense that produced +0.10 EPA per pass attempt with a quarterback on a rookie deal, which means the surrounding cast is worth paying attention to as contracts get set.

The sleeper case the numbers point toward is Nix himself, in formats where quarterback value is priced on contract efficiency rather than name recognition. At 1.7% of the cap and 81st-percentile production, the market still hasn't fully caught up to what he is. That's not a point projection — it's the structure of an efficiency gap. Engram is the flip side: 23rd-percentile production at 98th-percentile pay is a fade on principle, and you don't need to love the matchup to have a reason to look elsewhere. The aging curves don't flag anyone specifically in the brief, but a tight end at top-of-market money producing at the bottom quarter of the position is its own kind of warning.

Everything above traces to a live instrument, and the receipts stay public: the GM boards carry Sean Payton's decision card, the team's cap posture ($152.6M already charged to future caps), and every contract verdict with its falsifier and review date. When the season grades these reads, the grades post whether they flatter us or not.