That gives us a football story to chew on.

Safety had the biggest playoff-appearance lift. EDGE fell into the second slot. Quarterback stayed near the top. Tight end remained interesting. Cornerback was the clearest negative playoff-appearance room.

Running back/fullback, the position group that often gets the loudest argument, landed mildly positive but mostly quiet: +2.0 percentage points. Not a boom. Not a collapse. Mostly a shrug.

Top-quartile spend playoff appearance lift. Caption: Playoff-appearance rate gap between top-quartile and bottom-quartile spenders by position room.

Safety: the biggest playoff-appearance lift

The biggest playoff-appearance lift belonged to safety.

Teams in the top quartile of safety spending made the playoffs 52.0% of the time. Teams in the bottom quartile made it 34.2% of the time. That is the +17.8 percentage-point lift.

Safety also showed up well in the correlation table:

Safety vs. Win Percentage: r = +0.131

Safety vs. Playoff Appearance: r = +0.118

That is not the result most roster-building conversations would have predicted. Safety is often treated as a place to find value unless the player is truly special. But in this study, safety spending showed up first in the playoff-lift table. That should not be interpreted as “pay safeties.” It simply means high safety spending tended to appear on more playoff teams in this sample

The supporting examples are not hard to find.

2017 Patriots – Safety room Room cap: 14.0%        Result: 13 wins – Lost Super Bowl Key cap charges: • Devin McCourty, S – $10.94M – 6.5% of cap • Patrick Chung, S – $6.20M – 3.7% of cap • Duron Harmon, S – $3.50M – 2.1% of cap

2022 Bengals – Safety room Room cap: 12.0%        Result: 12 wins – Lost Conference Championship Key cap charges: • Jessie Bates III, S – $12.91M – 6.2% of cap • Vonn Bell, S – $7.49M – 3.6% of cap • Daxton Hill, S – $2.12M – 1.0% of cap

2021 Chiefs – Safety room Room cap: 12.9%        Result: 12 wins – Lost Conference Championship Key cap charges: • Tyrann Mathieu, S – $19.73M – 10.8% of cap • Juan Thornhill, S – $1.26M – 0.7% of cap • Daniel Sorensen, S – $1.21M – 0.7% of cap

2014 Seahawks – Safety room Room cap: 11.8%        Result: 12 wins – Lost Super Bowl Key cap charges: • Earl Thomas, S – $7.37M – 5.5% of cap • Kam Chancellor, S – $5.83M – 4.4% of cap • Jeron Johnson, S – $1.45M – 1.1% of cap

That is the version of the safety story that fits the aggregate result: top-quartile safety spend, playoff team, and in several cases a deep postseason run.

For fairness, high safety spend was not automatic. The 2023 Seahawks spent 12.6% of the cap at safety, led by Quandre Diggs, Jamal Adams, and Julian Love, and won nine games while missing the playoffs. That is a useful reminder, but it is not the headline. Across the full sample, safety had the biggest playoff-appearance lift.

Safety room playoff-lift result with supporting top-quartile playoff examples and a brief fairness note.

Quarterback becomes the cleanest overall signal

Quarterback remained exactly where football logic would expect it: near the top.

Teams in the top quartile of QB spending made the playoffs 47.4% of the time. Teams in the bottom quartile made it 32.9% of the time. That is a +14.5 percentage-point playoff-appearance lift.

Quarterback also produced the strongest room-level correlation anywhere in the matrix: QB vs. Win Percentage at r = +0.163.

QB spending also had a +1.30 regular-season win lift between top-quartile and bottom-quartile spenders.

That makes intuitive sense. Expensive quarterback rooms often mean a team has found someone worth paying. In the NFL, that is still roster-building oxygen.

The supporting examples look exactly like football logic says they should.

2009 Colts – Quarterback room Room cap: 20.2%        Result: 14 wins – Lost Super Bowl Key cap charges: • Peyton Manning, QB – $23.22M – 18.9% of cap • Jim Sorgi, QB – $1.30M – 1.1% of cap • Curtis Painter, QB – $0.33M – 0.3% of cap

2012 Broncos – Quarterback room Room cap: 16.4%        Result: 13 wins – Lost Divisional Round Key cap charges: • Peyton Manning, QB – $18.00M – 14.9% of cap • Caleb Hanie, QB – $1.12M – 0.9% of cap • Brock Osweiler, QB – $0.64M – 0.5% of cap

2022 Chiefs – Quarterback room Room cap: 18.6%        Result: 14 wins – Won Super Bowl Key cap charges: • Patrick Mahomes, QB – $35.79M – 17.2% of cap • Chad Henne, QB – $2.00M – 1.0% of cap • Shane Buechele, QB – $0.82M – 0.4% of cap

2006 Patriots – Quarterback room Room cap: 14.2%        Result: 12 wins – Lost Conference Championship Key cap charges: • Tom Brady, QB – $13.83M – 13.6% of cap • Matt Cassel, QB – $0.36M – 0.4% of cap • Vinny Testaverde, QB – $0.33M – 0.3% of cap

This is the cleanest positional story in the dataset. High-end quarterback spending often meant a real quarterback room, and those teams tended to win more.

For fairness, even quarterback spend is not magic. The 2015 Saints spent 17.7% of the cap at QB, led by Drew Brees at $23.80M, and won seven games. And the opposite path exists too: the 2014 Seahawks reached the Super Bowl while spending only 1.9% of the cap at quarterback, with Russell Wilson counting just 0.6%.

That is the quarterback complexity in one paragraph. Paying a quarterback can mean stability. It can mean you found a franchise player. Not paying one can be a problem, or it can be the best advantage in the sport if the player is good and still cheap.

Quarterback showed up. It should have.

EDGE looks positive, too

EDGE also landed in the positive group.

Top-quartile EDGE spending produced a +15.1 percentage-point playoff-appearance lift. EDGE spending also had positive correlations with regular-season win percentage and regular-season wins:

EDGE vs. Win Percentage: r = +0.157

EDGE vs. Regular-Season Wins: r = +0.157

That fits modern football logic. Pressure matters. Affecting the quarterback matters. Generating pass rush without blitzing matters. Expensive edge rooms are often expensive for a reason.

The supporting examples fit the premium-position argument.

2019 49ers – EDGE room Room cap: 20.6%        Result: 13 wins – Lost Super Bowl Key cap charges: • Dee Ford, EDGE – $14.37M – 7.6% of cap • Arik Armstead, EDGE – $9.05M – 4.8% of cap • Solomon Thomas, EDGE – $7.68M – 4.1% of cap

2021 Cowboys – EDGE room Room cap: 19.3%        Result: 12 wins – Lost Wild Card Key cap charges: • DeMarcus Lawrence, EDGE – $25.00M – 13.7% of cap • Tarell Basham, EDGE – $2.50M – 1.4% of cap • Randy Gregory, EDGE – $2.20M – 1.2% of cap

2024 Chargers – EDGE room Room cap: 22.7%        Result: 11 wins – Lost Wild Card Key cap charges: • Joey Bosa, EDGE – $26.11M – 10.2% of cap • Khalil Mack, EDGE – $25.59M – 10.0% of cap • Bud Dupree, EDGE – $2.36M – 0.9% of cap

2016 Packers – EDGE room Room cap: 19.5%        Result: 10 wins – Lost Conference Championship Key cap charges: • Clay Matthews Jr., EDGE – $13.75M – 8.9% of cap • Julius Peppers, EDGE – $10.50M – 6.8% of cap • Nick Perry, EDGE – $4.88M – 3.1% of cap

That is the right football read. Pass rush is valuable. Teams that invest heavily in edge pressure tend to be investing in one of the sport’s premium levers.

For fairness, EDGE classification can get messy across eras because some rushers are coded as ends, linebackers, or edge players depending on season and source. The direction is still sensible, but the bucket is not perfectly clean.

As a first-pass position-room result, EDGE belongs in the positive tier.

Tight end stays interesting; linebacker fades toward the middle

Tight end produced a +10.5 percentage-point playoff-appearance lift for top-quartile spenders. It also had the strongest room-level relationship with playoff wins and Super Bowl wins:

TE vs. Playoff Wins: r = +0.104

TE vs. Super Bowl Win: r = +0.098

The supporting examples are stronger than the usual tight end discourse would probably expect.

2024 Chiefs – Tight end room Room cap: 9.8%        Result: 15 wins – Lost Super Bowl Key cap charges: • Travis Kelce, TE – $19.55M – 7.7% of cap • Noah Gray, TE – $2.64M – 1.0% of cap • Jared Wiley, TE – $0.97M – 0.4% of cap

2013 Seahawks – Tight end room Room cap: 10.4%        Result: 13 wins – Won Super Bowl Key cap charges: • Zach Miller, TE – $11.00M – 8.9% of cap • Kellen Davis, TE – $0.67M – 0.5% of cap • Anthony McCoy, TE – $0.66M – 0.5% of cap

2020 Buccaneers – Tight end room Room cap: 9.5%        Result: 11 wins – Won Super Bowl Key cap charges: • Rob Gronkowski, TE – $9.25M – 4.7% of cap • Cameron Brate, TE – $4.25M – 2.1% of cap • O.J. Howard, TE – $3.53M – 1.8% of cap

2016 Patriots – Tight end room Room cap: 8.1%        Result: 14 wins – Won Super Bowl Key cap charges: • Rob Gronkowski, TE – $6.62M – 4.3% of cap • Martellus Bennett, TE – $5.18M – 3.3% of cap • Greg Scruggs, TE – $0.56M – 0.4% of cap

That is a legitimately interesting cluster: two Super Bowl winners, two more Super Bowl appearances, and several tight end rooms with meaningful veteran money.

For fairness, tight end spend was not magic. The 2022 Patriots spent 11.4% of the cap at tight end, led by Hunter Henry and Jonnu Smith, and won eight games.

Linebacker is a good example of why the article has to follow the data instead of the nostalgia. There are plenty of memorable high-spend linebacker rooms on good teams, but the updated top-vs-bottom playoff lift was 0.0 points. The room no longer drives the story the way safety, QB, EDGE, and TE do.

The 2012 49ers spent 19.8% of the cap at linebacker and reached the Super Bowl, led by Patrick Willis at $17.78M and Navorro Bowman at $2.22M. The 2008 Ravens spent 18.7% at linebacker and reached the conference championship, led by Ray Lewis and Bart Scott. Those are real football examples, but the broader room-level pattern faded in the full sample.

Cornerback is the negative outlier; RB/FB is mostly flat

Cornerback had the clearest negative playoff-appearance lift in the study. Naturally – this is where the Jets show up in the examples. Teams in the top quartile of CB spending had a playoff-appearance rate 4.6 percentage points lower than teams in the bottom quartile.

The strongest cornerback relationship was against Super Bowl appearance: CB vs. Super Bowl Appearance at r = -0.085.

The examples that support the negative cornerback result are pretty loud.

2020 Patriots – Cornerback room Room cap: 20.2%        Result: 7 wins – Missed playoffs Key cap charges: • Stephon Gilmore, CB – $23.64M – 11.9% of cap • Jonathan Jones, CB – $6.02M – 3.0% of cap • Jason McCourty, CB – $5.55M – 2.8% of cap

2012 Jets – Cornerback room Room cap: 21.1%        Result: 6 wins – Missed playoffs Key cap charges: • Darrelle Revis, CB – $11.77M – 9.8% of cap • Antonio Cromartie, CB – $8.25M – 6.8% of cap • Kyle Wilson, CB – $2.83M – 2.4% of cap

2016 Jets – Cornerback room Room cap: 15.8%        Result: 5 wins – Missed playoffs Key cap charges: • Darrelle Revis, CB – $17.00M – 10.9% of cap • Buster Skrine, CB – $5.25M – 3.4% of cap • Marcus Williams, CB – $0.60M – 0.4% of cap

2017 Eagles – Cornerback room Room cap: 2.3%        Result: 13 wins – Won Super Bowl Key cap charges: • Sidney Jones, CB – $1.12M – 0.7% of cap • Ronald Darby, CB – $0.80M – 0.5% of cap • Patrick Robinson, CB – $0.78M – 0.5% of cap

2022 Chiefs – Cornerback room Room cap: 3.4%        Result: 14 wins – Won Super Bowl Key cap charges: • Trent McDuffie, CB – $2.54M – 1.2% of cap • L’Jarius Sneed, CB – $1.05M – 0.5% of cap • Chris Lammons, CB – $0.90M – 0.4% of cap

That does not mean corners do not matter. It means expensive cornerback rooms did not show up as a positive team-success signal in this test.

For fairness, coverage absolutely matters, and there are expensive cornerback rooms that worked. The 2019 Ravens spent 20.5% of the cap at cornerback, with Jimmy Smith, Marcus Peters, and Brandon Carr carrying major charges, and still won 14 games. The point is narrower: high cornerback spending, as a room-level cap allocation variable, did not track cleanly with winning in this dataset.

Running back/fullback did not produce the dramatic result the discourse often expects. It landed mildly positive in top-quartile playoff lift, but close enough to flat that it is hard to build much around it:

Top-quartile RB/FB playoff appearance lift: +2.0 pts

RB/FB vs. Playoff Appearance: r = +0.001

The examples lean more cautionary than celebratory.

2025 Colts – RB/FB room Room cap: 6.6%        Result: 8 wins – Missed playoffs Key cap charges: • Jonathan Taylor, RB – $15.56M – 5.6% of cap • Tyler Goodson, RB – $1.04M – 0.4% of cap • DJ Giddens, RB – $0.95M – 0.3% of cap

2014 Vikings – RB/FB room Room cap: 13.9%        Result: 7 wins – Missed playoffs Key cap charges: • Adrian Peterson, RB – $14.40M – 10.8% of cap • Jerome Felton, FB – $2.13M – 1.6% of cap • Matt Asiata, RB – $0.57M – 0.4% of cap

2007 Saints – RB/FB room Room cap: 11.5%        Result: 7 wins – Missed playoffs Key cap charges: • Deuce McAllister, RB – $5.90M – 5.4% of cap • Reggie Bush, RB – $3.67M – 3.4% of cap • Aaron Stecker, RB – $1.44M – 1.3% of cap

2023 Titans – RB/FB room Room cap: 8.4%        Result: 6 wins – Missed playoffs Key cap charges: • Derrick Henry, RB – $16.37M – 7.3% of cap • Tyjae Spears, RB – $1.00M – 0.4% of cap • Hassan Haskins, RB – $0.65M – 0.3% of cap

That is not a clean ‘never pay running backs’ argument because the overall lift is not sharply negative. But it is absolutely not a ‘pay the room and winning follows’ argument either.

Put the first pass together and the hierarchy looks plausible enough.

Safety has the biggest playoff-appearance lift. Quarterback is the cleanest win-percentage room. EDGE looks useful. Tight end is interesting. Cornerback is the clearest negative outlier. RB/FB is basically a shrug.

That is the positional-value story the first layer of the data gives us.

The curveball: the signal is barely there

After all of that – after sorting by position room, checking playoff lifts, reviewing team examples, and looking for a positional-spending edge – the strongest signal in the entire matrix was:

QB vs. Win Percentage: r = +0.163

That was the best number on the board. That was the winner, the champion. That was the mighty positional-spending signal after 20 years of cap allocation data.

And it is tiny.

Because r is a standard correlation metric, we are not talking about some homemade grade or custom model score. We are asking a basic statistical question:

Across 20 years of team-seasons, did positional spending actually move with winning?

The best answer we found was +0.163.

There is no universal cutoff for Pearson’s r; these ranges are simply a practical framework for interpreting effects in this study:

0.00 to 0.10: static/noise

0.10 to 0.25: faint signal

0.25 or higher: worth real attention

0.35 or higher: moderate

0.50 or higher: strong

Now put the best result back on that scale.

r = +0.163 is in the faint-signal range.

It is below 0.25, the point where we would start paying real attention.

It is nowhere near 0.35, where we might call something moderate.

It is nowhere near 0.50, where a signal starts to look strong.

The best room-level finding in the entire study is only about two-thirds of the way to the ‘worth real attention’ line, and about one-third of the way to a strong signal.

That is the curveball.

The first table gave us a story. The correlation strength tells us the story is flimsy.

Each room’s strongest absolute correlation across the six outcomes, compared with practical signal thresholds.

A real-world way to understand how small this is

Correlation is not always intuitive, so here is the clean translation:

When you square r, you get a rough sense of how much variation the signal explains.

For the best result in the study:

0.163 squared is about 0.027.

That means the strongest positional-spending relationship explains roughly 2.7% of the variation.

Not 27%. 2.7%.

That is not a roster-building law. That is not a market inefficiency. That is not ‘here is the position you must pay.’

That is barely a fingerprint on the glass.

A 0.50 signal would be loud. That would explain about 25% of the variation. You would have to take it seriously.

A 0.35 signal would be meaningful. That would explain about 12% of the variation.

A 0.25 signal would at least deserve real attention. That would explain about 6% of the variation.

But 0.163? That is roughly 2.7%.

Trying to build a roster philosophy around that is like judging a steakhouse by the color of the plates. Maybe there is a little information there. Maybe good restaurants tend to use nicer plates. But if someone built an entire restaurant-ranking model around plate color, you would start looking for the emergency exit.

Or think of it like guessing a movie’s quality from the poster font. You might catch a genre. You might catch a vibe. But you are not getting the plot.

That is what positional cap allocation looks like in this data.

There are hints. There are patterns. There are fun rankings.

There is no strong signal.

The weak-signal problem holds across every outcome

The bigger problem for positional spending is that the weakness does not disappear when we change the outcome.

We tested regular-season win percentage, regular-season wins, playoff appearance, playoff wins, Super Bowl appearance, and Super Bowl win.

Here was the strongest room-level result for each outcome:

Regular-season win percentage: QB, r = +0.163 – faint

Regular-season wins: QB, r = +0.160 – faint

Playoff appearance: EDGE, r = +0.122 – faint

Playoff wins: Tight end, r = +0.104 – faint

Super Bowl appearance: Tight end, r = +0.087 – static/noise

Super Bowl win: Tight end, r = +0.098 – static/noise

That is the real argument.

Even when every outcome gets to pick its strongest room, nothing reaches 0.17.

Regular-season win percentage? Best signal: quarterback at +0.163.

Regular-season wins? Quarterback again at +0.160.

Playoff appearance? EDGE at +0.122.

Playoff wins? Tight end at +0.104.

Super Bowl appearance? Tight end at +0.087, which is static.

Super Bowl win? Tight end at +0.098, also static.

So yes, safety had the biggest playoff-appearance lift. Yes, quarterback became the cleanest overall room-level signal. Yes, EDGE looked positive. Yes, cornerback looked negative. Yes, RB/FB looked mostly flat.

But the larger truth is that none of these relationships is strong enough to support a sweeping roster-building rule.

The position-room ranking is interesting.

The signal strength is embarrassing.

The strongest positional spending relationship for each team-success measure.

The cap sheet cannot see the things that decide seasons

The problem is not that spending does not matter. Spending absolutely matters.

The problem is that positional allocation is a blunt instrument.

A cap percentage does not know whether the player is still good. It does not know whether a team is paying for future production or past reputation. It does not know whether the quarterback is stable or the player is hurt. It does not know whether the coaching staff can cover up a weak room. It does not know whether a low-cost player is actually a star on a rookie deal. It does not know whether a big cap charge reflects elite play, a restructure, a franchise tag, an injury, or a front office trying to fix yesterday’s mistake.

The room label cannot see the actual quality of the investment.

That is why even the supporting examples need context. The Chiefs’ QB spend worked because the player was Patrick Mahomes. The Patriots’ QB spend worked because the player was Tom Brady. The Chiefs’ tight end spend worked because the tight end was Travis Kelce. The 49ers’ EDGE spend worked because the roster had a real defensive front. The same cap bucket can point to a strength, a sunk cost, a team identity, or a roster problem.

The cap room tells you where the money was booked.

It does not tell you whether the money was good.

So what actually matters?

The real finding is not ‘pay quarterbacks no matter what.’ It is not ‘pay safeties.’ It is not ‘EDGE fixes everything.’ It is not ‘corners are bad investments.’ It is not ‘running backs are evil.’

The real finding is that positional salary-cap allocation is a weak explanatory variable for NFL success.

The NFL is not won by having the prettiest positional pie chart. It is won by getting the right players at the right prices.

The room label matters far less than the efficiency inside the room. Good player on a good contract is a premium position. Bad player on a bad contract is a non-premium position.

That is the positional value chart that actually matters.

For the people in the back

After 20 years of cap data, after ranking teams within each season to avoid cap-era distortion, and after using a standard statistical measure, the strongest positional-spending signal we found was:

QB vs. Win Percentage: r = +0.163

That explains roughly 2.7% of the variation.

So congratulations to positional-value discourse. After all the arguments, all the pie charts, all the ‘never pay this position’ and ‘you have to invest in that position’ commandments, the grand prize is two point seven percent and a participation trophy.

Not zero. Not less than zero.

Just barely more than zero.

The kind of nothing that wears a name tag.

Everyone can keep screaming about positional value if they want. The offseason content machine needs fuel. But in the actual results, positional spending shows up like a whisper in a hurricane.

The strongest signal is not strong. It is not moderate. It is not even close to the ‘worth real attention’ line. It is faint.

That is the joke.

Not that roster building does not matter. Roster building obviously matters.

The joke is pretending the answer is hidden in the positional spending buckets.

It is not.

It is about good players.

Point blank.

Good players, good contracts, quarterback stability, coaching, health, development, and surplus value.

That is the meal.

The positional cap pie chart is the plate color.

And after 20 years of data, the plate color barely tells us a damn thing.

Data note

Historical NFL salary-cap and player cap-hit data came from Spotrac.com. Position-room grouping, cap-percentage calculations, same-season percentile comparisons, correlation testing, playoff-lift calculations, and outcome joins were produced independently for this study using Codex.

The analysis covers team seasons from 2006 through 2025. The uncapped 2010 season is included in the broader database but excluded from cap-percentage correlation and lift calculations.

Player examples are based on listed cap charges within the analysis dataset. A cap charge does not always mean a player drove that team’s on-field result; that is exactly why this study treats positional allocation as a signal test, not a causal claim.