How valuable is a good reliever?

[ASFN Admin]

You must be registered for see images attach

Detroit Tigers pitcher Keider Montero (54) waves at the crowd after pitching change during the seventh inning against Texas Rangers at Comerica Park in Detroit on Saturday, May 2, 2026. | Junfu Han / USA TODAY NETWORK via Imagn Images

Despite a 48-54 record, the Detroit Tigers have a bit of an embarrassment of riches when it comes to good starting pitching. At various points in time, any and all of Tarik Skubal, Casey Mize, Jack Flaherty, and Troy Melton have been on the injured list, yet the team has the 6th highest fWAR and 3rd lowest ERA of any rotation in baseball. A large part of that has been Keider Montero’s stellar 2026 season.

After a strong start against the Cubs on July 22nd, Keider has 22 games (16 starts), 106 innings, a 3.14 ERA and 2.2 fWAR. He hasn’t struck out many (18.6%), but is throwing harder, showing better command, and generating weaker contact than he did in either 2024 or 2025. His walk rate is a career low and 8th best in the league, while his HR/FB rate is 4th. There’s been a little good fortune involved, but not much, and he’s legitimately pitched his way into any future rotation plans.

Despite all that, Montero can hardly hold onto a current rotation spot. He was called up on April 4th to start, pitched excellently, went to the bullpen on June 16th, started on the 21st, went back to the bullpen for July, then started again in that game against the Cubs. It’s been quite a rollercoaster for arguably Detroit’s most valuable pitcher this year.

All of which lends itself to two questions. The Tiger specific one – what’s the best way to use Keider Montero? – and the league wide question – how valuable is a good reliever?

To try and answer those, let’s begin by understanding the two main methods that analysts have available to describe pitcher value. The first is Wins Above Replacement (I’ll use fWAR because I’m more familiar with it, but the broad strokes of this discussion apply to bWAR as well). At its core, fWAR attempts to capture player value as a mix of volume (innings) and context-blind performance (FIP). Multiply the two together, use some adjustment coefficients, bam, fWAR. It’s simple, it’s not perfect, but it generally does a very good job of showcasing who adds the most talent to a team over the course of a full season, and because it uses FIP as its performance metric, it’s fairly predictive moving forward.

The other, lesser used option, is Win Probability Added (WPA). WPA checks Win Expectancy for a game before and after every play, notes the change, and credits that change as a positive or negative to the pitcher and hitter. Crucially, WPA is not context-blind like fWAR. A shutdown save is more important to winning than a shutdown 4th inning in a 9-0 loss. Like fWAR, WPA is accumulative, since every at-bat changes a players WPA over the course of the season.

Both have their merits, but because fWAR is more predictive of future results, it’s more frequently used for analysis. Predicting the future is a huge part of assembling a baseball roster, especially pre-season. However, fWAR can often fail to capture how valuable a reliever is, especially a good one, because their value is accumulated so densely.

Case in point: the single-highest fWAR season of the last half decade for a reliever is 2022 Edwin Diaz at 3.0 with a FIP of 0.90 and a strikeout rate of 50.2%. That is not a typo. Corey Kluber, Adam Wainwright, Miles Mikolas, and Chris Bassitt, in the innings-eating stages of their careers, all posted at least 2.8 fWAR that season. After that season, Diaz signed a $102M extension, while the four starters signed for $146M total, and half were out of baseball by 2024.

My point in this is not to hate on veteran starting pitchers, but to suggest the league does not value all 3-fWAR pitchers the same, and perhaps shed some light on how WPA can be more informative of what teams value besides fWAR. We know not all wins are equally valuable. For example, the LA Dodgers are on pace for 102 wins; it doesn’t matter if they get 98 or 103, they’ll still win their division handily. The inverse is true with the Colorado Rockies. The gap from 65 wins to 70 just won’t impact their trajectory. For the muddy middle, however, every win matters, and teams in this band – from, say, 83 to 90 projected wins – need every edge they can get as they claw for an uncertain playoff spot. Here’s where WPA matters most in the regular season. (Note: a similar theory applies to elite teams but with Championship Probability Added instead. That doesn’t really take hold until the postseason, which is why a team like the Dodgers still spent $300M in free agency last winter).

This is why, while the rotation was fully healthy, having Montero in the bullpen made a surprising amount of sense. The starting five – Skubal, Mize, Melton, Flaherty, and Framber Valdez – was already a top unit in baseball, while the bullpen needed far more help. Through the start of July, when Montero first moved to the pen, the Tigers’ pen was dead last in WPA. No bullpen did a worse job converting winning positions into wins than the Tigers’ without Montero. Since then, they’re 9th! It was more important for Detroit to capitalize on winning games that started well than for Montero to throw 6 good innings in a loss blown by Drew Anderson, for a purely representative example that definitely didn’t happen.

The cool part is it worked! Moving Montero, who’s clearly one of the five best pitchers on the roster, into the bullpen and using him in games that Detroit already established an opportunity to win, worked very well. I calculated Montero’s average WPA per game as a starter and as a reliever. I did the same for every other pitcher with as many innings as Montero did as a starter (90) or a reliever (10) and compared how much value he added. My theory was that by pitching well in relief during games Detroit should win, he’d add more to their chances than by pitching well as a starter in games they might win. I normalized by games played as a starter or reliever because WPA is a counting stat and I wanted to control for the fact he’s had fewer opportunities to contribute as a reliever than as a starter.

In all cases, Montero has been a good pitcher, and so he ranks highly in both categories. As a starter, he ranked 35th out of 179 with a .0675 WPA per game, aka about 7%. Basically, at the time he left the game, Detroit was 7% more likely to win the ballgames Montero started than when the game started. That doesn’t sound terribly impressive, but it’s ahead of guys like Skubal, Dylan Cease, and Paul Skenes. (Aside: sitting in 3rd place is none other than Troy Melton at .177 WPA per game! Take a bow, kid). However, as a reliever, Montero places third among 355 relievers at .095 WPA per game. The team was 10% more likely to win a game after Montero finished than when he came in! More importantly, that 10% average as a reliever is almost 1.5x higher than as a starter, and he can pitch more frequently, too.

Full credit goes to Montero for taking to his short-term relief role like a fish to water. His velocity increased, his strikeout rate jumped, and he routinely pitched multiple innings on shorter rest than he’d be used to as a starter. It’s a difficult role to adjust to, especially midseason, and him accepting whatever innings he’s given is a major reason Detroit has fought their way back to the edge of playoff contention. This plan doesn’t work unless he makes it work.

It also required AJ Hinch recognizing the Tigers’ needed condensed, short-term value more in June and July. Rather than going the traditional route and using a ‘failed starter’ as a long reliever in blowouts, Hinch fit Montero into situations he could save games and, potentially, the entire season. This is why dominant relievers get traded for top-5 prospects, elite closers get hundreds of millions to throw 60 innings a year, and why pitching chaos carried the Tigers to the playoffs in 2024. This is also why former starting pitching prospects who probably could have stayed in a starting role are converted to relievers. They simply wouldn’t have been as impactful as a mid-rotation or backend starter type.

If the Tigers decide to buy, the bullpen is the obvious place to start, especially now that Flaherty is injured and Montero is a starter again. They need enough to squeak into the playoffs, where they can use starters like Montero, Flaherty, or Jackson Jobe in flexible roles to paper over an otherwise pretty terrible bullpen. Those relievers could make the difference between ‘winning’ half games at a time or capitalizing fully on a great rotation. fWAR might not fully capture the story there. But for the Tigers, and many other teams, banking every win possible matters much more right now.

Continue reading...