The NBA’s Most Mispriced Contracts

Here is a thing that happens every July. A team signs a player, the internet decides within forty minutes whether the contract is Good or Bad, and then nobody ever checks.

We checked.

The model prices every player’s season the same way: how many wins he added above a freely available replacement, times what an open market pays for a win. Subtract his salary and you get surplus, the value a team received that it didn’t pay for. In 2025-26, across the 482 players on the board with a salary, teams paid $5359M and got back $11771M worth of basketball. That’s 2.2 dollars of production for every dollar of payroll.¹

Most of that surplus isn’t the market being wrong. It’s the rulebook. Players on rookie-scale deals, four seasons of service or fewer, took home 16% of the pay and produced 40% of the surplus. The 37 max contracts, which exist specifically to stop the best players from being paid what they’re worth, added another $1025M. Nobody made a mistake there. That’s the collective bargaining agreement working exactly as designed, which is to say working for the people who own the teams.

So if you want the market’s actual mistakes, you have to look at the contracts that were actually negotiated: veterans below the max, whose price was set by a front office and an agent staring at each other across a table. There were 232 of them. Here’s where it went sideways, in both directions.

The bargains

The top of the list is Jalen Brunson, at +$78M of surplus: 11.2 wins above replacement on a $35M salary, for production the market prices at $113M. Right behind him is Derrick White at +$75M.

But the best deal dollar for dollar belongs to Payton Pritchard: 7.6 wins for $7.2M. The going rate for a win last season was about $10M. His cost about $0.9M apiece. If you run a team and you have ever felt good about a contract, know that somewhere a capologist felt better.

SalaryMarket value of his production

The disasters, which mostly aren’t

Now the other end. Sort every contract outside the rookie scale by surplus, worst first, and the bottom ten reads less like a list of bad decisions than a trainer’s room: seven of the ten played fewer than half of their team’s games.²

That is a genuinely different problem. A contract that pays a man $46M to spend most of the season in a sport coat isn’t a scouting error. It’s an insurance claim. The model charges the full salary against whatever he produced, because that’s what the team actually got, but nobody should read Trae Young’s line as a verdict on how well he plays basketball.

Yes, Jayson Tatum is on this list. He played 16 games, coming back from a torn Achilles. This tells you a great deal about the list and nothing whatsoever about Jayson Tatum.

The ones who did play are the real mistakes: Nic Claxton, Jerami Grant and Kyle Kuzma. They were on the floor all year, and the floor was worse for it: every one of them posted a negative rating per 100 possessions. That’s what a mispriced contract actually looks like. It isn’t dramatic. It’s a solid-sounding name being paid like a starter and playing like a replacement.

So what did the market get wrong?

Less than you’d think, and not where you’d think. The giant numbers on both ends of the ledger mostly belong to the rulebook and the injury report. Strip those out and what’s left is smaller and more interesting: a handful of veterans who were underpriced because nobody bid hard enough for them, and a handful who were overpriced because somebody did. That’s the market this column is about. We’ll be back.

Figures: the 2025-26 season on the retrospective players board, as of this build. Wins are on this model’s own scale and dollars at the open-market price of a win; the method explains both.

1. This is not because thirty front offices are geniuses. The league caps what any one player can earn and fixes what rookies earn, so the best and the youngest players are paid less than an open market would pay them. Surplus is, in large part, a measurement of those rules. The methodology has the full derivation, including the parts that make economists squint. ↩
2. Forty-one of an 82-game schedule. Games played is the whole season’s, across teams for anyone traded. ↩