Why Pat Murphy Was Unequivocally Right to Remove Tobias Myers Sunday

It's never a good look when a manager removes a starting pitcher who appears to be cruising, while their pitch count remains low and the game is in the balance. That's what Pat Murphy did on Sunday, and when taking out Tobias Myers after five innings pitched and 64 pitches blew up with an immediate three-run Washington Nationals rally, it was only natural to ask: what was the skipper thinking?

Well, we can start by listening to him in his own words. Here they are, as reported by Milwaukee Journal-Sentinel beat reporter Curt Hogg:

As some fans were quick to point out, Murphy isn't perfectly accurate in that description of his own relief corps. While it's been stellar on the year as a whole, it's now struggling through a period of attrition and regression, with too much turnover and too few arms who feel like they're at full strength. In the last 30 days, they're 10th in MLB in reliever ERA, but 23rd in reliever FIP and 17th in both strikeout rate and strikeout rate-minus-walk rate. If you tighten the window to the period since the All-Star break, they're a hair better by ERA and FIP, but a bit worse in terms of missing bats and preventing free passes.

So, maybe Murphy was drawing from a well he thought to be deeper than it really is right now. Those remarks also betray a tendency he has to harbor extremely stringent, slightly unfair expectations of his defenders. There was nothing cheap about the first two singles in the inning, and after the luckless dribbler that did pin them to the wall a bit, James Wood hit a ball that Jackson Chourio never had a real chance to reach. Murphy seemed to believe the rookie could have made that play, but the best-case scenario was that he'd cut it off and hold Wood to a single. The result was not unearned, even if it was a tad unlucky.

He glazed over the real reasons for the move, rattling them off in non-specific fashion, but we should focus our attention there. Here are the essential claims to consider, moving beyond the momentary strength of the bullpen as a whole or the role of luck in determining the outcome of the game.

1. The top of the lineup was coming up a third time.
2. The day was hot and humid.
3. Myers is young.
4. The bullpen was full and needed to be used.

We can stipulate to the middle two pretty readily. They shouldn't play a huge role in making decisions like these, but Murphy is far from alone in lifting a starter a bit more proactively when it's in the mid-80s. Starters have averaged about slightly fewer pitches per game under such conditions over the last few seasons than when it's in the 70s or even low 80s.

Besides, that part of the talk is code, right? It's really Murphy saying that this staff can ill-afford more injuries, and occasionally lifting a pitcher without experience working deep into a season and piling up innings is the best way to try to avoid them. Myers built up to 140 innings last season, but he's already up to 105 this year, and pitching in the big leagues is a different animal from pitching only every sixth day in Double-A, as he did for most of last year.

Let's allow that aspect of the justification, while admitting that it's far from sufficient on its own. We can best tackle this by asking: Was Myers likely to succeed the third time through the Washington batting order?

No. He was not.

This season, Myers has pretty substantial platoon splits. Righties have just a .572 OPS against him, but lefties hit for a much more robust .763 figure. He's also felt the times-through-the-order penalty pretty powerfully. Opponents' OPS actually dips from .675 to .624 in going from the first time through to the second, but the third time, it's .769. At a pitch level, each time through, batters chase outside the zone against him a bit less, and whiff a bit less when they swing, and hit the ball hard a bit more often. That's all fairly routine.

Admittedly, the samples I'm about to cite are small, but check them out. Here's how Myers has performed by times through the order and batter handedness.

Even with small-sample caveats in play, one of these things is not like the others. By the third time they see him, so far, left-handed batters get it. They understand Myers, and they know what they want to do with him. Of the top five hitters in Washington's lineup Sunday, four are either left-handed batters or switch-hitters. Murphy and his staff were right to think that the best way to keep the game within reach was to go away from their young starter.

The bullpen was rested, sure, but just as importantly, they knew they had an off day coming Monday. They had Milner, plus Jared Koenig, each of whom had only worked once in the previous four days, meaning two southpaws could help build a bridge to Devin Williams--who remains on a somewhat scheduled regimen and needed to get work in Sunday, anyway. That broke any remaining ties, and the decision became easy.

In the long run, you want Myers to learn to face lefties a third time and find success. That's why, when the opportunity has been there, Murphy has pushed Myers a bit. This was the second straight start Myers departed after 18 batters faced, but over the previous four, he averaged over 26. He's gotten some chances to show and to develop that ability, but in a close game on a hot getaway day, with some relief arms itching for the work anyway, it was no time to force him through such a gauntlet.

The move didn't work. Milner struggled to miss bats, which has been a pattern for him lately. The Crew will need him to be good down the stretch to win the NL Central, though, just as they'll need Myers healthy and at his best. The coaching staff followed a good process by turning to the lefty for that lefty-laden segment of the batting order, especially given their starter's inability to handle such hitters under those circumstances so far. The outcome just wasn't what they wanted. Sometimes, it goes that way.