NFL Power Ratings vs. the Market — 2026

Two ratings for all 32 teams, on the same points scale. One is ours. The other is the rating the prediction market is implying, solved backwards out of live contract prices. The table is sorted by where they disagree, because the disagreements are the product.

How both numbers are built

PWR — points per game above an average team on a neutral field. PWR = Off PR + Def PR + ST PR. It comes from opponent-adjusted play-by-play, not from a vote.

Market PWR is inverted from Kalshi win-total contract prices. We read the expected wins implied by a team's 17-rung ladder, then solve for the single rating that reproduces exactly that many wins across the team's real 2026 schedule, through the identical win-probability function our own model uses. We do not read anyone else's published power score.

Opponents are held at our PWR. So market PWR is precisely “the rating the market implies for this team, holding our read of its opponents fixed” — which is what puts both columns on one scale. A team with no listed win-total market shows an em-dash, never an estimated rating.

#TeamPWRMarketGapRead

MIABelow Avg-3.3-9.7+6.4Market low

ARIBelow Avg-3.7-9.4+5.7Market low

CLEBelow Avg-2.3-7.3+4.9Market low

KCAverage-0.6+3.4-4.0Market high

LACAverage-0.6+3.2-3.8Market high

MINBelow Avg-2.6+1.1-3.7Market high

ATLAverage-0.4-3.9+3.5Market low

CINAverage-0.6+2.6-3.2Market high

LARContender+5.1+8.3-3.2Market high

BALPlayoff+1.2+4.1-2.9Market high

HOUPlayoff+2.0+4.2-2.3We lean down

NEContender+5.5+3.3+2.2We lean up

DETPlayoff+2.5+4.5-2.0We lean down

NYJBelow Avg-4.0-5.9+1.9We lean up

DENPlayoff+1.2+3.1-1.9We lean down

DALPlayoff+1.4+3.2-1.8We lean down

PHIPlayoff+2.4+3.7-1.4We lean down

LVRebuild-6.0-4.7-1.3We lean down

TENBelow Avg-3.7-5.0+1.3We lean up

WASAverage-1.5-2.8+1.3We lean up

SFPlayoff+1.0+2.2-1.2We lean down

BUFPlayoff+3.8+4.8-1.1We lean down

CHIPlayoff+1.4+2.4-1.0We lean down

GBPlayoff+3.0+2.1+0.9Agree

CARAverage-1.8-2.7+0.9Agree

PITAverage-0.5+0.2-0.7Agree

INDAverage+0.0-0.6+0.6Agree

TBAverage-0.6+0.0-0.6Agree

SEAContender+5.5+5.0+0.5Agree

NOBelow Avg-2.3-1.8-0.5Agree

NYGAverage-1.3-1.5+0.2Agree

JAXPlayoff+2.6+2.7-0.1Agree

32 of 32 teams carry a listed win-total market. Ratings as of 2026-09-01; prices read 2026-09-01 21:17 UTC and refreshed hourly. Full PWR board · Win totals vs the ladder · What this board does across 10,000 simulated seasons

Power ratings vs. the market — questions

What is an NFL power rating in points?

PWR — points per game above an average team on a neutral field. PWR = Off PR + Def PR + ST PR. A team at +6.0 is six points better than a league-average opponent before home field, so two teams four points apart project to a four-point spread on a neutral field. That is what separates a points-scale rating from a 1-32 media ranking: a ranking tells you the order, a points rating produces a number you can compare directly to a spread or a contract price.

How do market-implied power ratings differ from a media power ranking?

A media ranking is an ordering, usually re-shuffled each week by whoever writes it. A market-implied rating is solved backwards out of money: we take the live Kalshi win-total ladder for a team, read off the expected wins it implies, then find the single rating that reproduces exactly that many wins across the team’s real 2026 schedule, using the same win-probability function our own model uses. Opponents are held at our rating throughout, which is what puts both numbers on one scale.

Where does our model most disagree with the prediction market right now?

The table on this page is sorted by the size of the disagreement, largest first, so the top rows are the answer and they change as prices move. A positive gap means our rating is above the market-implied one — we think the market is underrating that team. A negative gap means the opposite. The gap is in the same points-per-game units as the ratings themselves.

How accurate is the model?

The DAEPA engine behind these ratings grades at 59.6% straight-up across 3,266 games (2013–2024), with a Brier score of 0.236 and log loss of 0.664. Those figures grade DAEPA, the underlying opponent-adjusted play-by-play model — PWR is one aggregation of it and is not separately backtested. Average error on a season win total is 2.32 wins, which is the honest scale of the disagreements below.

More NFL Markets

Grok fantasy football — connect the draft desk → — One URL, no key: the grok.com/connectors walkthrough and six prompts
NFL 2026 Market Guide — the 61-page book → — All 32 teams priced against the market, plus a 153-player draft board
NFL Player Props vs Kalshi — the edge board → — Our projection vs the live Kalshi prop line, biggest gap first
Madden 27 ratings vs our model vs the market → — All 32 team ratings, our power ratings, and live win-total prices on one board
NFL Predictions — every game vs the market → — Free model win probability vs Kalshi on all 18 weeks
2026 NFL Win Totals — model vs Kalshi → — All 32 teams: projected wins vs the KXNFLWINS ladder
NFL Game Edges — where the model disagrees → — Moneyline, spread & total edges vs Kalshi prices
NFL MVP Edges → — Live MVP market mispricings vs Kalshi KXNFLMVP
Win Total Futures → — Monte Carlo season wins vs the Kalshi line
Championship Edges → — Playoff / conference / Super Bowl probabilities
Kalshi NFL Markets → — How Kalshi prices every NFL market + live edges
Power Rankings — the model methodology anchor → — All 32 teams on the points-per-game scale
Fantasy Draft Assistant — the same model, in your AI agent → — Free MCP draft desk: board, best available, sleepers vs ADP, who-do-I-draft
NFL Model Calibration Ledger — how accurate is it, actually → — The graded record: 59.6% straight up, 0.236 Brier, and hit rate by confidence bucket across 3,266 games
NFL Power Ratings tool — build your own PWR board → — Adjust the model weights and re-rank all 32 teams on the points-per-game scale
NFL Win Probability calculator → — Turn any rating gap, spread or moneyline into a calibrated win probability

Model win probabilities from the DAEPA power-rating engine. Prediction markets carry risk; contracts can settle at $0. Informational only — trade responsibly.