Carolina Won Eight Games on Six-Win Talent

The Panthers went 8-9 last season, which sounds like modest progress. The point differential tells a different story. Carolina was outscored by 69 points on the year — a margin that, run through the points-per-win conversion our model uses, produces a structural win total of 6.6. The actual record sits 1.4 wins above that. That gap is the luck delta, and it's the single most predictive number in this entire piece.

The mechanism was the close games. Carolina went 7-3 in one-score decisions, and one-score records are the part of a football season that regresses hardest toward the mean. Teams that run hot in tight games one year tend to play those games closer to .500 the next. The Panthers didn't build a 7-3 record in those spots through some repeatable late-game process the numbers can identify — the EPA profile we'll get to in a moment makes that clear. They built it the way most teams do: a few bounces that went right.

None of this means 2025 was a fraud. Eight wins is eight wins, and Dave Canales has his team trending in a direction. But the honest starting point for 2026 is 6.6 structural wins, not 8, and the offseason work needs to be evaluated against that baseline.

The per-play EPA profile is not flattering. Carolina's passing offense produced -0.05 EPA per attempt — below average — and the rushing attack came in at -0.04 EPA per carry. Both units were net-negative on a per-play basis, which makes an 8-9 record look more impressive than the underlying performance warrants. The defense was a mixed picture: opponents generated +0.11 EPA per pass attempt against the Panthers, which ranks among the softer pass defenses in the league, while the run defense held roughly even at 0.00 EPA per carry allowed. You can stop the run or you can stop the pass; Carolina managed one of those.

The situational numbers were the quiet positive. A 43.8% third- and fourth-down conversion rate is a respectable figure, and the red zone EPA per play came in at essentially neutral — +0.01. Neither number is a strength you can hang your hat on, but they suggest the offense wasn't collapsing in the spots that matter most. A team with negative per-play EPA across both phases that still converts at 43.8% on third down is getting something right in terms of play design or personnel usage in critical moments, even if the broader production doesn't warrant celebration.

On the coaching side, Canales made correct decisions 76.5% of the time against our model's recommendations, which is a reasonable mark. His go-for-it rate on fourth down — 29.6% — ran well below the model's suggested 45.9%, costing Carolina roughly 0.28 expected wins over the course of the season. That's not a large number in isolation, but it's worth noting: the roster's ceiling is constrained enough that leaving a quarter-win on the field in decision-making matters more than it would for a team with a dominant quarterback.

The two headline arrivals are edge defender Jaelan Phillips (four years, $30M APY, from Miami) and linebacker Devin Lloyd (three years, $15M APY, from Jacksonville). Both carry WATCH verdicts — graded on price discipline — because both deals land at the 98th percentile of their respective position markets. Phillips is 27 and the market is paying for upside that injury history has so far borrowed against. Lloyd is 26 and the number is aggressive for a linebacker in a league that increasingly treats the position as replaceable. The exits on both deals are manageable — $17.5M dead on Phillips in 2028, $5.1M on Lloyd — so the Panthers aren't trapped if either struggles, but the entry prices reflect optimism the production hasn't yet confirmed.

The departures are quieter than the arrivals. Interior lineman A'Shawn Robinson took his 0.77 market-wins to Tampa Bay ($10M APY). Center Cade Mays went to Detroit ($8.3M APY, 0.62 market-wins). Running back Rico Dowdle signed with Pittsburgh for $6.1M per year; he produced 0.04 wins last season against a market value of 0.43, which is the definition of overpaying for the position. His departure alone moves the needle slightly in Carolina's favor. The net offseason on a production basis is -0.04 wins. On a market basis it's +1.88 — the gap between those two numbers reflects how much of this offseason is a bet on future production rather than confirmed value.

The draft class is led by Georgia offensive tackle Monroe Freeling at pick 19 (1.21 chart wins), Texas Tech defensive tackle Lee Hunter at 49 (0.83), and Tennessee wide receiver Chris Brazzell II at 83 (0.54). Rounds four through seven added cornerback Will Lee III (0.35), center Sam Hecht (0.33), safety Zakee Wheatley (0.30), and linebacker Jackson Kuwatch (0.14). A projected Round 6 compensatory pick in 2027 adds another 0.16 chart wins to the pipeline. The class totals 3.70 chart wins — a respectable haul, with Freeling as the centerpiece of what is clearly an offensive-line-first building philosophy.

The cap picture is where Carolina's situation gets complicated. The team carries $154.6 million in future proration — dead money and signing-bonus spread that will constrain flexibility in the years ahead. The quarterback cap percentage is 5.2%, which is the tell: you don't spend a nickel at the most important position on the field and expect to contend. The biggest cap hits belong to right guard Robert Hunt (7.8% of the cap), right tackle Taylor Moton (6.8%), and safety Trevon Moehrig (6.8%). Carolina is spending like an offensive-line-first team — the OL group sits 2.6 standard deviations above league average in cap allocation — which is a coherent philosophy if the quarterback situation resolves itself. If it doesn't, the Panthers will have built a beautiful frame around an empty room.

The albatross on the books is running back Chuba Hubbard, who sits at the 82nd percentile of cap cost for his position while producing at the 51st percentile. That's not catastrophic, but it's the kind of inefficiency that compounds when you're trying to build around a young roster. On the other side of the ledger, wide receiver Tetairoa McMillan is the clearest bright spot: 85th-percentile production on a 65th-percentile contract. He's the rare player where the market hasn't caught up to the output yet, and that gap has real value in a salary-cap environment this constrained.

Net the whole thing out and the offseason is mildly additive on paper, primarily defensive. Phillips and Lloyd upgrade a pass defense that gave up +0.11 EPA per attempt — that number needed addressing. But the production basis of the moves is essentially flat (-0.04 wins), and the market-basis optimism (+1.88 wins) is contingent on two players at the 98th percentile of their position markets living up to prices that imply elite performance. The aging curves don't yet have production percentiles to work with on either arrival, which means the honest answer is: wait and see. The help is possible. The hurt — if both deals underperform at top-of-market prices — is also possible.

Here is the arithmetic, laid out plainly. The 2025 point differential generates a structural base of 6.6 wins. The offseason moves add -0.04 wins on a production basis. The coaching staff is unchanged, so the decision-making delta is 0.0. That produces a structural win projection of 6.5 for 2026. This is an accounting projection from the ledger — it is not a betting recommendation, not a prediction, and not a claim about how the season will unfold. It is what the numbers say before a ball is snapped.

The falsifier is straightforward: if Carolina wins nine or more games in 2026, this read is wrong. That outcome would require either a material improvement in per-play EPA — particularly in the passing game, where the team was negative last year — or another strong run in one-score games, or both. The former would be genuine progress and would warrant revising the structural outlook upward. The latter would be a second consecutive year of positive luck variance, which happens but shouldn't be the base case. A season in the 7-8 win range would be consistent with regression toward the structural number and wouldn't represent failure so much as honesty.

The most meaningful opportunity shift on this roster follows Rico Dowdle to Pittsburgh. He logged 236 carries, 50 targets, and 216.3 PPR points in 2025. That is a substantial workload sitting unclaimed, and the question — the only question that matters for fantasy purposes — is who absorbs it. The departures leave the backfield thin enough that Hubbard, despite his albatross contract situation, becomes the default incumbent. The case against leaning on him heavily in fantasy construction is the 51st-percentile production number against an 82nd-percentile price tag: the market doesn't expect him to be particularly good, and there's no efficiency edge hiding in the data to argue otherwise.

The sleeper case on this roster belongs to McMillan. The efficiency-to-price gap — 85th-percentile production against a 65th-percentile contract — is exactly what the word sleeper is supposed to mean. He's already producing at a level the market hasn't priced in. The caveat is that Dowdle's vacated usage is concentrated in the backfield, not at receiver, so McMillan's upside is more about the passing offense taking a step forward generally than about a specific target redistribution. If the quarterback situation improves and the offense's -0.05 EPA per attempt moves toward neutral, McMillan is the player most positioned to capture the gains. That's the thesis — and the passing EPA number is also the falsifier. If it doesn't improve, neither does the ceiling.

Everything above traces to a live instrument, and the receipts stay public: the GM boards carry Dave Canales's decision card, the team's cap posture ($154.6M already charged to future caps), and every contract verdict with its falsifier and review date — including Jaelan Phillips (WATCH), Devin Lloyd (WATCH). When the season grades these reads, the grades post whether they flatter us or not.