The Setup Man Nobody Talks About: Grant Taylor's +17.1% Night
Grant Taylor didn't get a save last night, but he did get the win. By traditional standards, that's the recognition he deserves. But the win only tells part of the story. By win probability added, his value last night went far deeper than a W next to his name.
What Happened Last Night
CWS and BAL were tied 2-2 heading into the bottom of the 7th. Grant Taylor entered from the bullpen. He retired the side — three up, three down. The score stayed 2-2.
In the top of the 8th, the White Sox scored twice to take a 4-2 lead. Taylor came back out. He retired the side again. The White Sox held on to win.
No save. He got the win — but that's likely the extent of the recognition.
But if you care about win probability, Taylor's night looks very different.
By The Numbers
What WPA Actually Measures
Win Probability Added (WPA) answers a simple question: how much did this pitcher move his team's chances of winning?
Every game situation — inning, score, outs, runners on base — has a historical win probability attached to it. Each time a pitcher retires a batter, strands a runner, or gives up a hit, that probability shifts. WPA sums those shifts.
The tricky part is that context matters enormously. A scoreless inning with a 9-2 lead barely moves the needle. A scoreless inning in a tie game in the 7th is worth a lot. That's exactly where Taylor started last night.
The Win Probability Timeline
Inning-by-Inning Breakdown
| Situation | Score | CWS Win Prob (before) | CWS Win Prob (after) | WPA |
|---|---|---|---|---|
| Bot 7th — scoreless | 2-2 (tied) | 41.4% | 50.0% | +0.086 |
| Bot 8th — scoreless | 4-2 (CWS leads) | 84.4% | 92.9% | +0.085 |
| Total | — | 41.4% | 92.9% | +0.171 |
Why the 7th Inning and 8th Inning Win Probs Added Were Almost Identical
Both innings produced identical WPA (+0.086 and +0.085) but for very different reasons.
In a tie game entering the bottom of the 7th, the home team wins 58.6% of the time. That's a meaningful edge — batting last in the late innings is a real advantage. Baltimore still had three more half-innings to bat (the 7th, 8th, and 9th) while the White Sox had only two remaining. More at-bats means more chances to score, and in a tie game that structural edge is significant.
The 8th inning was different. With a two-run lead and six outs to go, CWS was already heavily favored at 84.4%. Taylor's job there was to protect, not to swing the game. He did, pushing the probability to 92.9%. Both innings mattered. The 7th changed the shape of the game. The 8th sealed it.
The Setup Man Problem
Here's the structural issue: closers get saves. Setup men only get a hold.
A closer who enters in the 9th with a 3-run lead and retires the side is worth around +0.050 WPA in a typical situation. Grant Taylor last night was worth +0.171 WPA — more than three times as much — without a single counting stat to show for it.
This isn't an argument against saves as a concept. It's an argument for recognizing what actually happened last night. The White Sox were 41.4% to win when Taylor entered the 7th. After his scoreless inning, they were at 50.0%. Then the offense took over — two runs in the top of the 8th pushed the probability to 84.4%, a 34.4-point jump that had nothing to do with Taylor. He came back out, pitched another scoreless inning, and left the game at 92.9%.
Taylor's actual contribution: +8.6% in the 7th, +8.5% in the 8th. That's the +17.1% that belongs to him. The other 34.4 points? Credit the lineup.
That's what makes WPA honest — it only gives you credit for what you did.
That's worth talking about.
Win probability values derived from the MalterAnalytics win probability matrix, built from 2015–2025 MLB regular season game states. All values represent 0-out, bases-empty situations at the start of each half-inning.