New England Won Fourteen Games and Still Has Room to Grow

Start with the most important number from New England's 2025 season, and it isn't 14. It's 13.2 — the structural win total implied by a +170 point differential, using our fitted formula of 8.5 wins plus point differential divided by 35.8. The Patriots won 14 games. The math says they deserved 13.2. That's a luck delta of +0.8, which means nearly all of what happened last year was earned. This wasn't a mirage team that ran hot in close games and needs to be faded. It was a legitimately dominant football team.

The one-score record — 7-3 in games decided by eight points or fewer — tells the same story. A 7-3 mark in close games is good, not unsustainable. Teams that go something like 2-8 in one-score games while outscoring opponents by 170 points are the ones that should scare you in regression analysis. New England wasn't one of those. The wins and the underlying numbers pointed in the same direction all season. That alignment is the foundation for everything that follows.

The honest takeaway: the Patriots were as good as their record suggested, the luck adjustment is minimal, and there's no meaningful regression tax sitting in the ledger waiting to be collected. When a team this strong heads into an offseason with flexibility and a franchise quarterback on a rookie contract, the conversation shifts from survival to extension.

The passing game was the engine. An offensive passing EPA of +0.32 per attempt is a genuine separator — that's the kind of number that shows up on playoff rosters, not wild-card hopefuls. The run game was essentially a push at -0.03 EPA per carry, which matters less when your quarterback is making defenses pay through the air on nearly every meaningful down. Drake Maye doesn't appear in the EPA table by name, but a +0.32 passing efficiency on a team where the quarterback sits at 96th percentile production and 3.2% of the cap doesn't happen by accident. The market hasn't priced him yet, and the ledger reflects that reality.

The defense was balanced and quietly excellent. Opponents produced -0.04 EPA per pass attempt and -0.04 EPA per carry against New England — equal-opportunity stinginess on both sides of the ball. A defense that limits opponents in the run game and the passing game at the same clip doesn't have an exploitable soft spot. That kind of profile tends to hold up across opponent types, which is part of why the point differential reached 170.

Third and fourth downs came in at 48.2% conversion, which is efficient without being extraordinary. The red zone EPA per play was exactly zero — neither an asset nor a liability. There's room there, but you don't fix a +170 differential by chasing red zone tinkering. The real work this team needs to do is modest.

Mike Vrabel's decision index landed at 71.8% agreement with our model — a competent number, but not a top-shelf one. His go-for-it rate of 23.1% on fourth down trailed a model recommendation of 41%, meaning he left roughly 0.268 expected wins on the field last season through conservative play-calling. That gap exists, and it's real, but it's also the kind of correctable habit that doesn't define a 14-3 team. A coach who wins 14 games with room to improve his decision rate is a problem most front offices would happily inherit.

The arrivals list is a mix of upside and caution. Romeo Doubs comes over from Green Bay at $17M per year — 96th percentile pay for 62nd percentile production, and our aging curves attach a 0.75 multiplier to a player who's 26 and has yet to post an elite season. The contract verdict is OVERPAY. The falsifier is straightforward: if Doubs finishes top-five in wide receiver production in 2026, this price is fair and the verdict is wrong. Until then, the market paid for a projection rather than a track record. The exit cost sits at $7.8M dead in 2028, which is manageable if the read is right.

Alijah Vera-Tucker arrives from the Jets at $14M per year — 93rd percentile pay for a guard whose production data is incomplete enough to earn a WATCH verdict rather than a grade. At 27, the aging picture is neutral. The exit in 2027 carries $12M dead, which is the detail worth monitoring; that's a meaningful number if the line doesn't gel. Dre'Mont Jones comes from Baltimore at $12.17M per year, also 93rd percentile pay, also a WATCH on incomplete production data. He's 29, which the aging curves will start to notice, and his exit in 2028 costs only $3.7M dead — the most team-friendly escape hatch in the class. Kevin Byard at $7M per year and Julian Hill at $5M round out the additions at safety and tight end, both at prices that won't haunt the cap sheet.

The departures were K'Lavon Chaisson (to Washington, $11M market value), Khyiris Tonga (to Kansas City, $7M), and Jaylinn Hawkins (to Baltimore, $5M) — a combined $23M in market value walking out the door. The net offseason ledger on a market basis is +2.58 wins. On a production basis, it's nearly flat at +0.01 wins, because the arrivals haven't yet proven they produce at the prices they're being paid. That gap between market and production — 2.57 wins — is the offseason's honest uncertainty. The draft class data wasn't supplied, so no class wins are in the model.

The core of this roster is in genuinely strong shape, and the cap posture backs that up. The quarterback situation is the most important number on any team's balance sheet: Maye sits at 3.2% of the cap while producing at the 96th percentile. That's the definition of a rookie windfall, and it's the structural advantage that makes everything else possible. A.J. Brown reinforces it — 2% of the cap, 88th percentile production, 65th percentile pay. Two of the most important offensive pieces are being paid like role players. Teams that enjoy that kind of imbalance can afford to be patient with the pieces around them.

The cap's biggest concentration sits in the interior defensive line — an IDL group spending at 1.71 standard deviations above league average, led by Milton Williams at 8.1% of the cap and Christian Barmore at 4.9%. Carlton Davis owns 6.2% at corner, and Michael Onwenu takes 5% at right guard. That's a lot of cap tied to premium positions that aren't the quarterback, which is fine as long as the quarterback remains on a rookie deal. The moment that changes, the math on some of these commitments gets harder. There are no albatrosses on the ledger today, but the future proration of $96.4M is a number to watch as the next contract cycle approaches.

The aging flag on Doubs is the honest concern. A 0.75 aging multiplier on a $17M-per-year wide receiver means the production case for this deal depends almost entirely on a player having the best seasons of his career on a contract he signed before proving he could deliver them. Special teams spending sits at -1.32 standard deviations below league average — the single biggest relative underinvestment on the roster. That's a place where modest dollars could produce outsized field-position returns for a team that already wins the point differential battle.

The arithmetic here is about as clean as it gets. The 2025 point differential projects to 13.2 structural wins. The offseason adds 0.01 production wins — essentially nothing, because the arrivals haven't demonstrated they produce at their market prices. The coaching staff is unchanged, so the decision delta is zero. The structural win projection for 2026 is 13.3.

To be direct about what that number is: it's an accounting projection from the ledger, built from point differential and offseason movement. It is not a betting model output and it is not a pick. It is the honest answer to the question of what New England's résumé earns them heading into 2026, before a single game is played.

The falsifier: if the Patriots win fewer than eleven games in 2026, this read is wrong. That outcome would suggest the defensive performance wasn't as sustainable as the EPA profile implied, or the Doubs addition set back the receiving room rather than upgrading it, or Vrabel's fourth-down conservatism compounded into something more damaging than a quarter-win drag. Any of those things can happen. Eleven wins is where this projection earns its skepticism.

The departures left no vacated opportunity data in the brief — the players who left were primarily defensive contributors, not offensive targets — so the fantasy case here is built around what arrived rather than what was left behind. Romeo Doubs saw 85 targets in 2025 and posted 165.4 PPR points, a per-target rate of 1.95. That's the efficiency number to hold onto. He steps into an offense with a quarterback producing at the 96th percentile, which is the single best thing you can say about any receiver's landing spot. The market may not have fully priced that context into his 2026 cost, particularly early in the offseason.

The aging flag on Doubs — that 0.75 multiplier — is the counterargument, and it's real. A receiver whose contract was graded an overpay at signing, with an aging curve already pulling against him, is exactly the kind of player to fade in formats where you're paying full price for projected upside. The efficiency is there. The volume is there. The question is whether 26-year-old Doubs can sustain the level that justifies the price, or whether the curve wins. That's a DFS decision made by the salary, not by the player.

Julian Hill drew 20 targets in 2025 with a 31 PPR total — modest numbers, but he's playing behind a quarterback who moves the chains through the air. If the tight end role expands in Vrabel's offense, Hill's 20-target baseline is a floor with upside, not a ceiling. No sleeper case without a stronger per-target number in the brief, but the situation is worth monitoring into training camp. The honest read on New England's fantasy ecosystem is simply this: Drake Maye makes everyone around him more interesting than they'd be anywhere else, and the market for his supporting cast hasn't fully caught up to that yet.

Everything above traces to a live instrument, and the receipts stay public: the GM boards carry Mike Vrabel's decision card, the team's cap posture ($96.4M already charged to future caps), and every contract verdict with its falsifier and review date — including Romeo Doubs (OVERPAY), Alijah Vera-Tucker (WATCH), Dre'Mont Jones (WATCH). When the season grades these reads, the grades post whether they flatter us or not.