The best team to be a fan of is not the one with the most trophies

We scored all 32 NFL franchises on six measured dimensions — out of 7,276 games, the 2024 Fan Cost Index and ESPN attendance back to 2013 — with every dimension counting exactly the same. New England sits at the 100th percentile on both winning and playoff payoff and finishes 15th. The Ravens win, and not for the trophy case.

Ask which NFL team is best and you get an argument about rings. Ask which team is best to be a fan of and it is a different question, because you are not asking what the franchise has won. You are asking what it gives back to the people who show up.

Those are not the same question, and the gap between them is the whole finding here.

The ruler

Six dimensions. Each one is a thing you can measure and point at, not a vibe with a number stapled to it.

dimension	what it measures	unit
Success	how often they win	win %, 1999–2025
Payoff	what the winning is worth	playoff wins per season
Drought	how long you wait	seasons, vs expected
Cost	what a Sunday costs	$ / family of four
Loyalty	whether the building fills	vs the club’s own peak crowd
Now	is there something to believe in	form index, 2021–2025

Each club is ranked against the other 31 on each dimension, converted to a percentile, then combined. All six count equally.

Equal weights are a declared value judgement, not a neutral default. Nobody publishes a weight vector over these six things, and inventing one would make the answer an opinion with a chart on it. So the weighting is stated up front and then tested rather than defended — see the last section, where it turns out to matter.

Two of the six are adjusted before they are scored. Good teams have short droughts and full stadiums because they are good, so counting those straight would be counting winning three times. Drought and loyalty are each regressed on win percentage and scored on what is left over.

The loyalty denominator is the choice that changed the most. Measuring attendance against listed stadium capacity puts six clubs above 100% — Dallas at 116.6%, Buffalo at 115.6% — and a share that routinely exceeds one is not a share. Worse, it hides the thing you most want to see: Washington removed seats as the crowds fell, 82,000 down to 64,000, so the denominator chased the collapse and the collapse vanished. Measured against their own record crowd, Washington is drawing 77.3% — 22.7% off their own peak.

What it found

The famous teams are not bad. They are spiky.

#	club	score	succ	pay	drgt	cost	loyal	now
15	Patriots	53.8	100	100	29	16	16	61
12	49ers	57.4	45	77	84	3	42	93
19	Steelers	52.3	97	77	23	35	39	43
20	Cowboys	50.8	65	15	52	26	84	65

The four most decorated franchises in football, by percentile on each dimension. New England is at the top of the league on two of the six and near the bottom on two others.

New England is the clearest case. 100th percentile on winning. 100th on playoff payoff. Six Super Bowls. And 16th percentile on cost — $765 for a family of four, the fourth most expensive game day in the league — and 16th on loyalty. Add it up and the Patriots finish 15th of 32.

Dallas is stranger still. The Cowboys have the fullest building in football relative to their own peak and have not reached a conference final in any of the twenty-seven seasons measured. Zero. They finish 20th. Pittsburgh, 97th percentile at winning, finishes 19th.

The top of the table

#	club	score	succ	pay	drgt	cost	loyal	now
1	Ravens	73.1	90	92	61	55	65	76
2	Seahawks	71.1	81	87	65	39	81	75
3	Colts	69.9	87	71	48	94	90	29
4	Vikings	65.5	69	39	68	87	77	52
5	Eagles	63.3	84	97	42	10	55	93
6	Rams	62.9	55	84	35	45	74	84
7	Chargers	61.8	61	27	74	77	94	37
8	Packers	60.8	94	77	71	0	58	65

The eight highest-scoring franchises. Note that none of them is spiky — and note the Packers’ zero.

Baltimore wins with 73.1 / 100, +1.60σ, two points clear of Seattle. They are not the best at anything on this table. They are 90th percentile at winning, 92nd at payoff, and merely respectable at everything else — and that is the point.

The Ravens are the only club in the league above the median on all six measures. Their worst dimension is the 55th percentile. No other club in football has a worst dimension that clears the 39th. Green Bay scores 94th at winning and zero at cost; Philadelphia scores 97th at payoff and ten at cost. Everyone else has a hole somewhere.

Two Super Bowls, not six. Fifteenth-cheapest game day rather than the cheapest. Nothing about Baltimore is a league record. They simply have no weakness, and over a lifetime of Sundays that is what a franchise actually owes you.

And the bottom

#	club	score	succ	pay	drgt	cost	loyal	now
29	Raiders	27.3	3	15	16	13	100	17
30	Commanders	23.3	19	15	26	32	0	48
31	Lions	21.5	6	5	10	29	13	66
32	Browns	16.9	0	0	32	23	26	21

The Raiders are the interesting one: 100th percentile on loyalty — the most faithful crowd in football relative to its own peak — and 3rd percentile at winning.

The part that argues against us

The weighting is a choice, and a different choice moves the answer. That is not a caveat hidden at the bottom of a methodology file; it is a result, and it is this:

reading	winner	where Baltimore finishes
All six equal (published)	Ravens	1st
Trophies first	Ravens	1st
Value first	Colts	2nd
The Sundays first	Cardinals	6th

And across 100,000 weightings drawn uniformly from the whole simplex — not four opinions, all of them — Baltimore is #1 in 32.7%, against 25.1% for Indianapolis. Best mean rank in the league at 3.2.

Nineteen of the thirty-two clubs finish first under some defensible weighting. Baltimore is the most robust answer in football by a clear margin — it wins a third of every possible weighting, half again as often as the next club — and it still loses two weightings in three.

So the honest claim is not “the Ravens are the best team to support” full stop. It is that under the stated definition, and under most other defensible ones, no franchise gives back more per season of belief. If you weight the trophy case higher, you get the same answer. If you weight the price of a ticket higher, you get Indianapolis. If what you care about is the Sunday itself, you get Arizona.

All three of those are in the file.

Get the data

All 32 franchises, scored and ranked, with every input column in its raw units, every percentile, both attendance denominators, the ranks under all four named weightings, and each club’s share of the 100,000 random ones.

the data — everything above is reproducible from these

nfl-fan-ruler-2026.csv — 32 franchises × 47 columns: the composite score, the overall rank, all six percentiles, win rate, playoff wins, droughts, the Fan Cost Index, attendance against both denominators, and the full sensitivity sweep.
nfl-fan-ruler-methodology.md — the sourcing, why two dimensions are residualised on winning, why the attendance denominator changed, and everything the score deliberately does not measure.

Sources: nflverse/nfldata games.csv — every NFL game 1999–2025, 7,276 of them; Team Marketing Report Fan Cost Index 2024; ESPN NFL attendance 2013–2025. Home-field advantage was measured (league mean +4.48 points) and then dropped from the model: after shrinkage the between-club variance is indistinguishable from sampling noise, so scoring it would have been fitting noise. It ships in the CSV as an unused column.