The Boston Celtics’ front office has a model. The model says Mitchell Robinson, when healthy, is worth $47.4 million over three years. The model is, technically, not wrong about this. The model does not have a column for “when.”
https://x.com/ShamsCharania/status/2072341817060724877
Robinson and the Celtics agreed to terms on July 1 — the same day they traded Jaylen Brown to the 76ers for Paul George and a pile of future picks. It is a full non-taxpayer mid-level exception, which means the Celtics are now hard-capped to the first apron, which means the model has also foreclosed most of their flexibility for the foreseeable future. They spent their exception on Robinson; Robinson’s availability has to do the rest.
https://x.com/YossiGozlan/status/2072342199329857979
This is where the numbers become what they are. Robinson’s injury history reads like a body in negotiation with itself: 66 games in year one, 61 in year two, 31 in year three, 72 in year four — the one season he crested the threshold that the league considers a full year — then 59, 31, 17, and 60 last season. Eight years, 397 total games. Roughly 50 per season. The one time he played 72 was 2021-22, and Boston is paying $15.8 million annually on the thesis that 2021-22 is the real Mitchell Robinson, that the other seven seasons were aberrations rather than the data.
Think of it like a reliability rating on a used car with one good year on the odometer; the dealer shows you that year’s receipts, points at the engine, tells you the previous owner just had bad luck. Maybe. Or the car has been telling you something for eight years.
What Robinson offers when he plays is genuinely difficult to argue with: 72.3% shooting last season, 8.8 rebounds, 4.2 offensive boards per game — second in the league — and a presence at the rim that changes geometry. The Celtics were a 56-win team that went home in the first round; they needed a center. The Robinson who played 60 games in 2025-26 looked like the thing they needed. The Celtics ran the projection forward, liked what they saw, and signed it.
The problem is that this is, structurally, a bet that the exception proves the rule; that the bounce-back year is the floor rather than the ceiling. It is the kind of bet that feels irresponsible unless you dress it up in cap-sheet language, at which point it starts to look like due diligence. Boston did not miss on Giannis — Giannis went to Miami — and they did not have another option at center who moved the needle in a different direction. They had Robinson’s 60-game season and a spreadsheet that said it penciled out. So they penciled it out.
There is a version of this that works. Robinson stays healthy, plays 65 games, crashes the offensive glass, gives Jayson Tatum a legitimate interior presence for the first time in two years, and Boston re-establishes itself as a conference contender. That version exists. It just requires that Mitchell Robinson’s body agree to a three-year commitment it has never before honored.
The Celtics signed the player he is when everything goes right; they have paid, in full, for the right to wait and find out whether everything goes right. On a holiday weekend, while the rest of the league was spending its summer, Boston looked at a center who has played 72 games in a season exactly once across eight professional years and decided the spreadsheet said yes. The model is very confident. The model will have an opinion on this in February.
