Rays vs. Tigers: Same Run Differential, 20-Win Gap Explained

Tampa Bay (85-58) and Detroit (65-78) have nearly identical run differentials, but a huge split in one-run and blowout games explains the gap in the standings.

Published: September 7, 2026 · Figures as of September 7, 2026

Asked: The Rays have a +47 run differential and are 85-58. The Tigers have a +50 run differential and are 65-78. Can you explain?

Same run differential, opposite records — it's almost entirely one-run games

Detroit's +50 run differential is actually slightly better than Tampa Bay's +47, and when you convert both to Pythagorean expectation (runs scored/allowed, exponent 1.83 — the standard baseball value), the two teams come out almost identical: Tampa Bay projects to a .534 win rate (about 76.4 wins over 143 games), Detroit to .538 (about 76.9 wins). Their underlying play has been essentially the same team. The 20-win gap in the standings — Tampa Bay 85-58, Detroit 65-78 — comes from how each team's runs were distributed across games, not from how many runs they scored or allowed overall.

The split shows up cleanly in one-run games and blowouts:

Tampa Bay has won 60% of its one-run games while Detroit has won only a third of its own, which alone accounts for roughly the gap between "should be ~77 wins" and the actual 85-65 split. It runs the other way in lopsided games: Detroit has blown teams out more often (24 wins by 5+ runs) while also eating more blowout losses on Tampa Bay's side (23), which is exactly the kind of thing that inflates run differential without moving the record much.

DataBaller's own team-sustainability read confirms this precisely rather than approximately: Tampa Bay's record sits 8.6 wins ahead of what its scoring margins support (value 0.225, "expect fallback"), and Detroit's sits 11.9 wins behind (value 0.0, "expect recovery") — both as of today, full-season confidence. That's the model-computed version of the same gap the raw Pythagorean numbers above show.

One caveat worth stating plainly: the conventional sabermetric instinct is to wave this off as "just one-run luck," but DataBaller's own testing found MLB one-run records carry more persistence than that story assumes — they aren't pure noise. So this isn't a guarantee both records snap back to .536; it's a lean, and it's the strongest one the data supports. It also can't say why Detroit keeps losing close ones and Tampa Bay keeps winning them (bullpen usage in the late innings is the traditional explanation, but that's not something this measures).

Confidence: High that the gap is a close-game/blowout-distribution story rather than a difference in overall run prevention or scoring — the Pythagorean numbers and the model's own decomposition agree exactly. Medium on how much of the gap closes going forward, since one-run performance regresses only partially, not to zero.