Will Scherzer’s Shoulder Woes Show in Game Four?

On September 18th, Max Scherzer was removed from his start after two innings because of shoulder woes. There might not be a more troubling word in the world of pitching than ‘shoulder.’ Once it goes, with it goes your velocity, which is much worse than the risk of temporary loss of playing time and control that the elbow offers. We had so much success predicting a low-scoring game for Anibal Sanchez yesterday, let’s wash off the template and start again. Did something happen in mid-September that fundamentally changed Scherzer’s game?

Again, to the gun we go first:

That’s not good news. Scherzer lost about two miles per hour off his fastball that day and hasn’t really gotten it back. And lest you think that three games isn’t a great sample, Jeff Zimmerman found that three starts of data after a return from the DL can predict future fastball velocity within one MPH 95% of the time.

It’s not *quite* that simple, though. In Scherzer’s last start, he bumped his fastball back up to 93.7 mph from the 91.4 he showed in his first start after being pulled on September 13. Looks like the ten days off did some good, and this time he’s getting seven days off. Though the hetero-chromatic one was actually averaging over 95 mph in the second half, 93.7 mph isn’t far from his full-season fastball average of 94.2 mph. (And it’s dead-on his career fastball average of 93.5 mph.) So let’s say the gun says he’s slightly reduced when it comes to fastball velocity — and that slight reduction might easily disappear when the postseason adrenaline kicks in.

Watching the gun will reveal something early on, though, and it’s easy. Pitching mix is harder to discern while watching baseball in the moment, but it can be as revealing. Since the injury, Scherzer has used his fastball less. According to Joe Lefkowitz’s PITCHf/x tool, he threw the fastball 57.3% of the time to righties and 64.4% of the time to lefties before September 18th. After, those numbers dropped to 46.3% and 53.6% respectively. Where we ascribed some of Anibal Sanchez’s success to his reduced fastball usage, here it seems a little worrisome. After all, with a swinging strike rate of 12.2% on the season (second in baseball among qualified starters), Scherzer’s problem isn’t garnering more swinging strikes. He’s even retained all his excellent swinging strike rates since his injury.

What is his problem, then? Well, he hasn’t topped five innings since September 18th, so it might be durability. And that’s actually been a thing all year. Scherzer averaged 5.85 innings per start, and that was 15th-worst among the 88 qualified starters this year. With someone that throws so many off-speed pitches anyway — 38.3% on the year, and 20.8% sliders — that finding may not be startling. But a reduced fastball velocity, begetting a reduced reliance on the fastball… what if that, in turn, begets another short start?

If that’s the case, we’ve established the superiority of the A’s bullpen from the start. Judging from recent usage, it’s probably Drew Smyly that’s the first out of the pen after a short outing from Scherzer, and though the rookie was really good this year, and should be better in a short stint, his 3.83 FIP on the year falls short of Scherzer’s 3.27 FIP… but it does match up well with A.J. Griffin’s 3.85 FIP.

The real ‘problem,’ then, doesn’t look so much like the first five or six innings, when even a short outing from Mad Max is still likely to keep the game a close, low-scoring affair — it’s the next three innings, when the Tigers try to match Sean Doolittle, Ryan Cook and Grant Balfour with Octavio Dotel, Phil Coke, Joaquin Benoit and Jose Valverde. As good as Dotel looked Tuesday night, it was Doolitle that got the fans roaring by striking out the side.