The Los Angeles Rams went 12-5 last season, and the most honest thing you can say about that record is that it undersells them. Run the point differential — +172 for the year — through the fitted points-per-win formula and you get a structural win total of 13.3. The actual ledger reads 12. That gap of 1.3 wins isn't bad luck in the casual sense; it's a measurable residue, and it has a name: the Rams were 4-5 in one-score games. Nine games decided by eight points or fewer, and they lost five of them. The league's coin flip in those situations is closer to even, which means the most predictive thing about Los Angeles heading into 2026 isn't that they were a twelve-win team — it's that they played like a thirteen-win team and the scoreboard didn't always agree.

That distinction matters more than it sounds. Teams that outperform on point differential and underperform on one-score record tend to regress toward their underlying quality, not away from it. The Rams' structural read doesn't require a bounce-back story, a new coach, or a healthy roster. It just requires the close games to break a little more evenly, which is what the evidence says they do over time.

The short version: the Rams were an above-average football team on both sides of the ball, and the passing game was the engine. Their offensive EPA per pass attempt came in at +0.25 — genuinely good, not just functional. The rushing attack was essentially a wash at -0.01 EPA per carry, which is fine as long as you're not building an identity around it, and the Rams aren't. The offense lived and died through the air, and mostly it lived.

Defensively, the picture is quieter but real. Opponents produced -0.03 EPA per pass attempt against this defense and -0.04 EPA per carry — both below zero, both pointing toward a unit that was making offenses work for yards rather than giving them away. A defense that holds opponents below zero on both dimensions in the same season is doing something right. The 46.2% third- and fourth-down conversion rate on offense and a red-zone EPA of +0.09 per play round out a team that was efficient when the field shrank, which is where good teams separate themselves from average ones.

Sean McVay's decision index tells a familiar story for one of the league's more accomplished coaches: 79% agreement with the model, 0.218 expected wins left on the field, a fourth-down go-rate of 28.4% against a model recommendation of 40.7%. McVay is aggressive by reputation and more conservative by the numbers than that reputation suggests. The gap between his actual go-rate and the model's recommendation is worth about two-tenths of a win a season — not ruinous, but not nothing either. A coach who closes that gap even partially over a full year hands himself a meaningful edge before the roster does anything.

The Rams didn't lose anyone of consequence in the offseason — the departures column is empty — and they went out and rebuilt their secondary. Trent McDuffie arrives from Kansas City on a four-year deal at $31 million per year, a contract that sits at the 100th percentile of the cornerback market. Jaylen Watson, also from the Chiefs, comes in behind him at $17 million per year over three years, landing at the 97th percentile for his position. The market priced these two together at 3.96 net wins added. No production data yet in a Rams uniform, so the production-basis number stays at zero — that's honesty, not pessimism.

On the contract verdicts: McDuffie and Watson both carry WATCH flags, and the reasoning is the same for both — they're graded on price discipline, not on talent, because we simply don't have their production numbers in this system yet. The falsifier for both is straightforward: if the cornerback market's ceiling moves significantly past these deals by 2027, they'll look like bargains. If it doesn't, the Rams paid top of market for players the numbers haven't confirmed yet. Matt Stafford's one-year extension at $55 million per year gets a FAIR PRICE verdict — he's at the 96th pay percentile and 88th production percentile for the position at age 39. One year, $9.6 million in dead money on exit. The team priced in the age and left themselves an off-ramp. That's competent underwriting.

Kamren Curl rounds out the new deals at safety — three years, $12 million per year, a WATCH verdict at the 95th pay percentile. The same price-discipline caveat applies. No draft class data is present in this brief. Compensatory pick projections for 2027 are not available.

Inside the existing roster, the cap picture has some texture worth noting. Colby Parkinson at tight end is the one clear albatross: 2.9% of the cap, 35th-percentile production, 84th-percentile pay. That's a meaningful mismatch at a position the Rams are already spending above average on — their tight end group sits at a +1.01 cap z-score relative to the league. Against that, Puka Nacua at wide receiver is the clearest bargain on the roster: 1.9% of the cap, 97th-percentile production, 59th-percentile pay. He's doing elite work on a price the market hasn't caught up with yet.

On balance, the offseason helps — or at least doesn't hurt — and the honest accounting explains why both halves of that sentence matter. The Rams added two cornerbacks to a defense that was already producing negative EPA per pass attempt allowed. If McDuffie and Watson play to their market price, the secondary improves. If they play to their production history from Kansas City, the secondary improves more. The floor here is a competent secondary getting expensive; the ceiling is a genuinely elite one. The uncertainty is real, and the WATCH flags exist for a reason, but the direction of the addition is correct.

The albatross exposure at tight end is the quiet drag on the roster. The Rams are paying above market at the position group level, and their best tight end by the production numbers — Parkinson — is a 35th-percentile producer on 84th-percentile pay. That's money that could be doing something else, and in a cap environment where the Rams are carrying $311.2 million in future proration — a number that constrains flexibility more than current-year hits do — every dollar misallocated inside the roster costs twice. Once now, once in cap flexibility later.

Stafford at 39 on one year is the right structure for the right player at the right moment. The aging curve on quarterbacks at his age is steep and well-documented, but the one-year deal means the Rams don't get buried if it bends. What would concern a neutral observer is a scenario where Stafford falls from the 88th percentile toward the 30th — the cap hit stays, the production doesn't, and a team built around his arm has a problem. The deal acknowledges that risk without pretending it doesn't exist. That's the best you can do.

Here's the arithmetic, stated plainly. The structural base coming out of 2025 is 13.3 wins, derived from a +172 point differential. The offseason adds zero wins on a production basis — the new arrivals carry market-basis value but no confirmed production in this system yet, so the honest ledger holds them at zero rather than count dollars as wins. The coaching staff is unchanged, so there's no decision-delta adjustment in either direction. The structural projection for 2026 is 13.3 wins. That is an accounting projection from the ledger, not a betting model output, not a pick, not a guarantee.

The falsifier: if the Rams win ten games or fewer in 2026, this read was wrong, and the most likely explanation would be Stafford's production falling sharply from its 2025 level, the new cornerbacks underperforming their contracts, or the one-score record breaking the wrong way again. Any of those could happen. The structural number says they shouldn't all happen at once — but the job of a falsifier is to name what would break the model, not to promise it won't.

The brief flags no vacated opportunity from departures — no one of consequence left — so there's no target redistribution story to tell here, and manufacturing one would be inventing data. What the brief does show is a receiver already doing the work: Nacua at the 97th percentile of production on 59th-percentile pay is the textbook definition of what a sleeper actually is in a mature market. The efficiency is real, the price hasn't caught up, and the opportunity is already his. That combination doesn't last forever, which is what makes it worth noting now.

The aging flag that matters most in this offense belongs to Stafford. He's at 88th-percentile production on a one-year deal, which means the team has already priced in the risk that age bends his output — and in fantasy contexts, a 39-year-old quarterback's floor is wider than the surface numbers suggest. That's not a fade recommendation; it's a range-of-outcomes flag. The upside is a top-ten QB season with a receiver room that can sustain it. The downside is a mid-season performance decline that the Rams can absorb contractually but fantasy managers cannot. Price him accordingly, and know the risk you're holding.

Everything above traces to a live instrument, and the receipts stay public: the GM boards carry Sean McVay's decision card, the team's cap posture ($311.2M already charged to future caps), and every contract verdict with its falsifier and review date — including Matt Stafford (FAIR PRICE), Trent McDuffie (WATCH), Jaylen Watson (WATCH), Kamren Curl (WATCH). When the season grades these reads, the grades post whether they flatter us or not.