NFL · state-space ratings · diagnostic

What we knew, versus what was true

A Kalman filter rates each team using only the games played to date. An RTS smoother re-rates the same weeks knowing how the season ended. The gap between those two lines is how badly the model misjudged a team at the time — and it is the view that answers whether these ratings pass a smell test, rather than whether their arithmetic is self-consistent.

Mean misjudgment
1.39
points, over 32 teams × 22 weeks
Worst single week
7.76
points, the largest gap any team opened
Home-field miss
2.18
points in 2019 — the known defect, still visible

That last tile is a positive control, and it is supposed to fail. Home-field advantage is pinned flat by construction, so the model cannot track the 2019 collapse — it estimated +2.3 while the league actually played to +0.1. A diagnostic that cannot surface a defect its author already knows about is not evidence about the ones they don’t.

Every team’s season, both ways

Solid is the causal filter — the rating through that week, using nothing later. Dashed is hindsight over the full season. They necessarily converge in the final week: at the last step there is no future left to condition on.

Causal — what we knew thenSmoothed — hindsight
Seattle Seahawks weekly causal versus smoothed rating, 2025
SEA · week 22Causal +11.1Smoothed +11.1Misjudged by 0.00
+0+5+101357911131517192122WEEK+11.1

Conditions on future games. Display only — these numbers may never inform a prediction.

Who the filter misread

Mean absolute gap between the two lines across 2025. Select a bar to put that team in the chart above.

Mean absolute gap between causal and smoothed rating, by team
0.01.02.03.0MEAN |CAUSAL − SMOOTHED|, POINTSSEA3.51NE3.04JAX2.83NYJ2.74ARI2.25LA2.04WAS1.97KC1.81LV1.71HOU1.59TB1.50CHI1.48SF1.35DET1.29GB1.26CIN1.23IND1.22BAL1.13DAL1.11MIN1.09NO0.95MIA0.87NYG0.87PHI0.85TEN0.74LAC0.72ATL0.71CAR0.66BUF0.55CLE0.51PIT0.50DEN0.46

The control: home-field advantage

The model’s home-field estimate against what actually happened, 19992025. The model line barely moves — by construction. The real one collapses, then recovers.

Model estimateActual home margin
Model home-field advantage versus actual home margin, by season
+0+1+2+3+419992002200520082011201420172020202320252019 · +0.1

The season that fails worst is 2019, not 2020. That is the more uncomfortable result: 2020 has an explanation anyone would accept — empty stadiums — while 2019 was played in front of full crowds and still ran a home margin of +0.09 points. A model that cannot move its home-field term cannot see either one, and only one of them has a story attached.

Table view — all 32 teams
2025 NFL causal versus smoothed ratings, by team
TeamMean gapMax gapFinal causalFinal smoothed
SEASeattle Seahawks3.517.76+11.1+11.1
NENew England Patriots3.046.93+5.5+5.5
JAXJacksonville Jaguars2.835.99+5.5+5.5
NYJNew York Jets2.744.83−10.1−10.1
ARIArizona Cardinals2.254.52−5.3−5.3
LALos Angeles Rams2.045.33+8.1+8.1
WASWashington Commanders1.975.81−3.8−3.8
KCKansas City Chiefs1.814.20+0.4+0.4
LVLas Vegas Raiders1.715.02−8.6−8.6
HOUHouston Texans1.595.03+5.5+5.5
TBTampa Bay Buccaneers1.503.52−1.1−1.1
CHIChicago Bears1.483.85+1.5+1.5
SFSan Francisco 49ers1.352.85+2.8+2.8
DETDetroit Lions1.292.84+3.6+3.6
GBGreen Bay Packers1.262.50+1.5+1.5
CINCincinnati Bengals1.233.36−1.7−1.7
INDIndianapolis Colts1.222.82+1.0+1.0
BALBaltimore Ravens1.133.58+3.5+3.5
DALDallas Cowboys1.112.99−3.5−3.5
MINMinnesota Vikings1.093.85+1.5+1.5
NONew Orleans Saints0.952.56−3.6−3.6
MIAMiami Dolphins0.872.98−4.3−4.3
NYGNew York Giants0.872.40−2.6−2.6
PHIPhiladelphia Eagles0.852.01+3.1+3.1
TENTennessee Titans0.743.26−7.5−7.5
LACLos Angeles Chargers0.721.93+0.8+0.8
ATLAtlanta Falcons0.712.69−2.4−2.4
CARCarolina Panthers0.661.68−3.4−3.4
BUFBuffalo Bills0.551.75+5.5+5.5
CLECleveland Browns0.511.66−5.5−5.5
PITPittsburgh Steelers0.501.68−0.6−0.6
DENDenver Broncos0.461.20+3.3+3.3

How this is built

Two passes over one filter

A state-space model carries each team’s strength as a distribution that drifts week to week and reverts across an offseason. The forward pass is a Kalman filter: at week k it has seen weeks 1–k and nothing else. The backward pass is Rauch-Tung-Striebel, smoothed within the season, which revisits every week with the whole season in hand. Ratings are taken through each week — the state after that week’s games.

Why the dashed line is not a forecast

A smoothed week-4 rating is a function of weeks 5 onward. It is strictly better on every in-sample metric and means nothing as a prediction, which is exactly what makes it easy to misuse. Four tested barriers in the rating engine keep the smoother off every prediction path, and the payload behind this page carries a machine-readable non-causal stamp that the reader refuses to serve without. Nothing on this page is a projection.

Source: nflverse schedules, 19992025. Computed deterministically — never modeled by a language model. Built by scripts/build_nfl_hindsight.py (byte-reproducible), whose build gate reproduces the published 2024 validation cells before it writes.