Every August, NFL fans ask the same question:
“Did the best players make the 53-man roster?”
It’s a perfectly reasonable question. It’s also the wrong one.
Because the job of an NFL front office isn’t to reward the best performances from training camp or the preseason. Its job is to assemble the roster that gives the organization the greatest chance to succeed – not just this season, but over the next three or four years.
In management, there’s a simple framework that captures this dilemma: the Performance–Potential-Matrix,
better known as the 9-Box Grid. Organizations evaluate employees on two separate dimensions:
- Performance: What has this person accomplished so far?
- Potential: What could this person become?
The famous 9-box grid looks like this:
body .sbnu-legacy-content-table td, body .sbnu-legacy-content-table th, body .sbnu-legacy-content-table { border: 1px solid #000 !important; border-collapse: collapse !important; } body .sbnu-legacy-content-table td, body .sbnu-legacy-content-table th { padding: 4px 6px !important; }| Low Potential | Medium Potential | High Potential | |
|---|---|---|---|
| High Performance | Expert High performer Limited upside |
Key Contributor Strong performer Still developing |
Future Leader High performer High potential |
| Medium Performance | Steady Contributor | Core Employee | Emerging Talent |
| Low Performance | Poor Fit | Needs Development | High-Risk Bet High upside |
The key insight is that high performers aren’t automatically high potentials, and vice versa. Great organizations try to identify both.
NFL roster construction is not fundamentally different.
Every player fighting for one of the final roster spots sits somewhere on that same matrix. A veteran may have consistently demonstrated he can execute his assignments on gameday. A rookie may be less reliable today but possess physical traits that suggest a much higher ceiling.
The challenge isn’t identifying which player performed better in August. The challenge is deciding which player creates more value going forward.
Looking beyond current performance
This is where the philosophy behind Moneyball becomes surprisingly relevant. Most people remember Moneyball as the story of baseball’s statistical revolution – or an odd film about baseball with Brad Pitt. But the real lesson wasn’t about replacing scouts with spreadsheets. It was about understanding that observed performance isn’t the same thing as underlying talent – and that statistics are evidence about a player but not the player himself.
Billy Beane wasn’t asking who had the better season, he was asking why a particular player had a better season. Was it due to a repeatable skill? Or was it inflated by noisy information like opponent quality, teammate performance, or simple good fortune?
People often think the idea is about finding obscure, undervalued statistics. But that’s just the tactic. The philosophy goes much deeper and is about estimating the talent, not the performance.
The signal vs. noise problem
Imagine two hitters in baseball. Player A has a batting average of .340, Player B has a batting average of .285.
A traditional evaluation would hold that Player A is clearly the better player. A better evaluation asks why Player A has the better batting average.
- Did Player A hit an unusually high number of bloopers?
- Did Player B hit rockets directly at defenders?
- Did Player A benefit from an unusually high batting average on balls in play?
- Who is more likely to repeat next season?
Those are very different questions.
Player A may be .300 talent + .040 luck, while Player B may be .325 talent − .040 luck. The second player is likely the better future investment and the observed gap (.340 vs .285) exaggerates the real difference.
That’s exactly the same question NFL teams face every August. A productive preseason doesn’t automatically make someone the better long-term player. It merely provides another piece of evidence.
Separating signal from noise
Imagine two young edge rushers competing for the final roster spot. Player A finishes the preseason with three sacks. Player B recorded no sacks.
body .sbnu-legacy-content-table td, body .sbnu-legacy-content-table th, body .sbnu-legacy-content-table { border: 1px solid #000 !important; border-collapse: collapse !important; } body .sbnu-legacy-content-table td, body .sbnu-legacy-content-table th { padding: 4px 6px !important; }| Pressures | Sacks | |
|---|---|---|
| Player A | 8 | 3 |
| Player B | 15 | 0 |
But suppose Player B consistently beats offensive tackles, generating pressure on nearly every second pass rush, while Player A records his sacks after quarterbacks hold the ball too long or steps into pressure created by someone else.
Now the evaluation changes. Sacks are the outcome. Consistently beating offensive tackles is the skill. One is considerably noisier than the other. Plus, we know that pressure rate is generally much more stable from year to year.
Similar approach for cornerbacks, where one player might have more interceptions than another. But maybe the player with the higher INT number was targeted more often and benefited from tipped passes, overthrown balls, or late throws, while the player with the lower INT number was rarely targeted by the opposing offense. INTs fluctuate dramatically from year-to-year, coverage quality tends to be much more consistent. And again the evaluation changes.
At running back, instead of total yardage, you might want to look at yards before contact to understand which player created more yards on his own.
A receiver’s preseason statistics depend on which quarterback throws him the football, whether he’s facing starters or third-string defensive backs, how many opportunities he receives, and even which plays happen to be called.
The front office isn’t trying to identify who had the better preseason. It’s trying to identify which player is more likely to produce over the next four seasons. The same logic applies across the roster.
The numbers matter. But they don’t speak for themselves. Great personnel departments spend less time asking what happened and more time trying to figure out why it happened.
Different people. Different time horizons.
The production-versus-potential debate also exists because different people inside an NFL organization are solving different problems.
A position coach wants to maximize this season’s roster. His performance review begins in September. Winning games keeps him employed. Naturally, he’ll often prefer the veteran who already knows the system and can be trusted on third-and-six in September.
A scout is solving a different problem. His job is to identify the players with the highest long-term NFL ceiling. He isn’t evaluating September. He’s evaluating the next three years.
The front office has to reconcile both perspectives while also considering contracts, salary cap implications, future free agents, and the overall direction of the franchise.
And because of this, none of these people are asking exactly the same question.
The coach wants to know who’ll help him win now. The scout is trying to figure out who can become the better player. And the front office is looking for the player that creates the greatest value for the organization over the next several seasons.
None of those questions are wrong. They’re simply optimizing different time horizons.
Choosing between two futures
Fans often believe the final roster should reward the “best camp.”
The veteran receiver catches everything in camp. The rookie drops a couple of passes. The veteran survives final cuts. Nobody complains.
But sometimes the opposite happens. The veteran is released. The rookie stays. Fans immediately conclude the coaches made the wrong decision because the rookie had the weaker preseason
And maybe the coaches did make the wrong decision. Or maybe the organization concluded that the rookie’s preseason was a less accurate reflection of the player he will become. Perhaps he faced tougher competition. Perhaps he was learning multiple positions. Perhaps his athletic traits are impossible to teach while the mistakes are entirely coachable.
Those are investment decisions. They’re not rewards for August production, they’re about deciding which player creates more value going forward.
The Cowboys’ balancing act
This is the balancing act every NFL team faces during roster cutdowns. Keep the dependable veteran who probably gives you slightly better football this season? Or keep the young prospect whose best football may still be two years away? Neither decision is automatically correct.
If you’re one player away from competing for a Super Bowl, immediate production may outweigh long-term upside. If you’re building toward sustained success, sacrificing a little today may pay enormous dividends tomorrow. The difficult part is knowing the difference.
That’s the real challenge of roster construction. The final cuts aren’t rewards for the best preseason performances. They’re predictions about who will become the best NFL players.
