Strategic Slip in Seventh Stunts ‘Stros

When the seventh inning began, the Astros’ chances of winning the World Series looked good. With a rolling Zack Greinke, Gerrit Cole available out of the bullpen, and closer Roberto Osuna fresh, the Astros had a clear path to getting the final nine outs and celebrating a title for the second time in three seasons. It didn’t work out that way. The Nationals rallied, the Astros were defeated, and A.J. Hinch’s decision making merits some scrutiny. Bad outcomes can cause us to believe the decisions that led to those outcomes were poor, when that isn’t always the case. Let’s take a look.

We’ll start with Greinke entering the seventh, take a quick detour, and then get back to him. After six innings, Greinke had made a total of 67 pitches; he got through the sixth on just eight pitches, including a strikeout of Trea Turner to end that frame. While Greinke’s velocity is not a big part of his game, his fastball velocity was still fine and he topped 90 mph on one of the pitches to Turner. Heading into the seventh, Astros manager A.J. Hinch had a few different options. He could continue on with Greinke, go to Cole, go to Osuna, or go to someone else, like Will Harris or Jose Urquidy.

There are two causes for concern with respect to Greinke, one sort of real, the other self-imposed by the Astros. The Greinke-related issue is that he was about to face the Nationals the third time through the order. He had retired Turner, but Adam Eaton, Anthony Rendon, and Juan Soto loomed. But at this point in the game, facing the order a third time should have been of minimal concern. The “penalty” pitchers often experience is due to two factors. One is the rising pitch count of the pitcher. With Greinke only at 67 pitches, that really wasn’t an issue. The other factor is that the third time through, pitchers often pitch against a portion of the lineup that is disproportionately comprised of the better hitters at the top of the order. While that was an issue for Greinke with Eaton, Rendon, and Soto, it would have been a problem for the Astros no matter who was on the mound. So then the question is, who is the better pitcher, a rolling Zack Greinke or one of the bullpen arms?

Will Harris as reliever and Zack Greinke as starter put up roughly equivalent numbers, with FIPs around three. Add to that that Harris had pitched the night before and Greinke seems like he was a sound choice. Urquidy pitched well in the fourth game of the series and also performed well in two relief outings earlier in the postseason. There’s an argument to be made that as a reliever, Urquidy might be a little better than Greinke as a starter, but it isn’t an especially compelling one. Roberto Osuna also put up similar numbers to Greinke’s in the regular season. The idea behind pulling starters is to replace them with relievers who are better. Greinke at 67 pitches is one of the 10 best starters in baseball, and as good or better than most of Houston’s relief options. It doesn’t make a ton of sense to pull him while he’s still on his game.

That leaves Gerrit Cole. It’s not clear why Cole was only going to be available for the ninth inning if Houston got the lead. He was warming earlier in the game. He was pitching on two days rest, so it’s possible he was only going to be available for an inning, and it seems reasonable to want to put him in at the start of an inning so he can be better prepared for it, but having him only available in the ninth to close out a World Series win is an odd choice and makes one wonder if the decision wasn’t entirely baseball-related. In any event, if Cole could have only gone one inning and needed to start it, then sticking with Greinke to start the seventh was completely reasonable.

Here’s where Greinke’s pitches went to the first three batters in that frame, from Baseball Savant.

Against Eaton, he pounded the outside corner away and induced a groundout. Against Rendon, Greinke threw the hardest pitch of his night at 91.8 mph for a ball, and then missed with a changeup that Rendon crushed. The walk to Soto put the winning run on base, but Greinke caught a bit of a bad break during the at-bat. After a 1-0 whiff on an outside curve, Soto took a change outside. Then, he took a change that should have made the count 2-2, but instead made it 3-1. With the count tilted in Soto’s favor, Greinke threw the same curve that got the whiff earlier, but Soto took the pitch and went to first.

With Howie Kendrick coming up, we are faced with a set of questions similar to those from the beginning of the inning. Greinke was now at 80 pitches and with a walk and a homer, the results said he was getting worse. His velocity against Rendon and the tough break against Soto — one of the best hitters in the game regardless of age, with his 155 wRC+ against righties behind only Christian Yelich, Mike Trout, Cody Bellinger, and George Springer (min. 350 PA v RH) this season — it’s not clear that Greinke didn’t do the right thing by not giving in. Cole seemingly needed a clean start to the inning to enter the game, and that logic might have also been true for Urquidy, who had only come in during the middle of an inning with the Astros once. (That relief outing came in the second inning of a September game against the Angels, and while barely worth mentioning, he gave up a single to the first batter.)

In his piece on the same subject, Michael Baumann discussed the reasons why relieving Greinke and bringing in Harris was defensible, though he did acknowledge Harris’ potential wear as a point against it. That Osuna came in later that inning is another point against it (Osuna is better than Harris), but it’s still not clear that pulling Greinke was the right move. With the lineup through Rendon and Soto, unless Greinke was tired, any issues related to the third time through the order were mostly moot. Howie Kendrick has been a good hitter, but he’s not on the level of Rendon or Soto. With Kendrick and then Asdrúbal Cabrera coming up, leaving Greinke in might have been the best play if Hinch was going to bring in a pitcher other than Gerrit Cole or the best reliever. And if Greinke was tiring, he wasn’t really showing it based on velocity and just pitching Soto carefully.

At that point, the decision should have been Osuna or Greinke; if Hinch thought Osuna wasn’t the best available reliever because he had a few slip-ups in the postseason, then the choice should have been sticking with Greinke. Playing by the numbers doesn’t always require pulling the starter. Relievers aren’t necessarily better than the guy currently on the mound, and even good relievers aren’t usually better than a fresh starter if he’s one of the 10 best in the game. Greinke might have had some so-so outings in the playoffs before last night, and his three strikeouts might not have suggested dominance, but 19 called strikes out of the 80 pitches he threw is an indicator that he was keeping Washington off balance. Was relieving Greinke defensible? Sure. Was it the right call? I’m less certain. It’s usually better to take a pitcher out too early than too late, but in the most important plate appearance of the season, Houston’s fifth-best pitcher threw the pitch that lost the Astros the lead and eventually the championship.