Playoff Odds (ESPN/Fangraphs): 68.0% (30.9% Division, 37.2% Wild Card)
Playoff Odds (Baseball Prospectus): 64.9% (16.7% Division, 48.2% Wild Card)
If you’re wondering why I’ve decided to close down this blog at the end of the year, yesterday’s game was as good an explanation as any. I feel like I’m trapped in the Royals’ version of Groundhog Day, condemned to write the same columns over and over, first pointing out that Decision X is a bad idea, and then inevitably writing a post-mortem explanation of why Decision X was a bad idea after reality has proven that it was, in fact, a bad idea.
Letting Gil Meche throw 117 pitches with a dead arm. Trading for Mike Jacobs. Signing Jason Kendall. Signing Yuniesky Betancourt. Sticking with Luke Hochevar in the rotation. Etc. Etc.
I’m not saying I always get it right – James Shields says hi – and I would argue that the Royals under Dayton Moore have improved in this regard considerably over the years. But a single tactical decision may have hurt the Royals more than any of the ones I mentioned above, even those that affected the Royals for years, because it came in a pennant race, and one game may well cost the Royals a playoff spot.
And I’ve been writing for at least a month now that AARON CROW ISN’T ANY GOOD, and that JASON FRASOR IS A BETTER PITCHER AND SHOULD BE PITCHING IN KEY SITUATIONS.
It’s true, Ned Yost trusted Crow with a one-run lead in the ninth inning on September 2nd and it worked out, and again used him to pitch the seventh inning in a 2-0 game at Yankee Stadium and it worked out. Which simply proves that – particularly in today’s run environment – even mediocre pitchers will throw a scoreless inning the majority of the time. A pitcher with a 4.50 ERA (ignoring unearned runs for a moment) allows a run every other inning, so unless he literally never allows a crooked number, he’s going to throw a scoreless inning more than 50% of the time. It would take a pitcher with a 6+ ERA before the odds that he pitches a scoreless inning would drop under 50%.
That’s if a pitcher is starting an inning fresh. When a pitcher comes in with men on second and third and one out, those odds drop considerably, and you need your best arms then. As we saw yesterday.
I’m not going to break down the decision in too much detail, both because I have no time and because Yost’s decision to let Daniel Nava bat against Crow with the bases loaded, two outs, and the Royals clinging to a one-run lead in the sixth inning has been the talk of baseball over the last 24 hours. Andy McCullough does a brilliant job of being about as critical in his game recap as a beat writer can be. Joe Posnanski weighed in. Jonah Keri weighed in. The Effectively Wild podcast from Baseball Prospectus weighed in. I don’t need to pile on.
But I do want to point out that there are essentially two mistakes that Yost made. The first one is perhaps the worse one from the standpoint of how much it hurts the team, but it’s also the more forgivable one, because it’s the mistake that the majority of managers in baseball would make.
That’s the decision to not go to one of his elite pitchers in that situation because “Aaron Crow’s inning is the sixth inning”. We can laugh about this if we want; yesterday was just the 9th time in 64 outings for Crow this year that he pitched in the sixth inning at all, and the third time since the All-Star Break. We can get mad if we want; analysts have been arguing for 30 years – basically since the 1970s model of using your best reliever as a “stopper” gave way to the 1980s model of using your best reliever as a “closer” – that saving your best reliever for a late-inning situation is the height of foolishness when the game is on the line a little earlier.
But the reality is that Yost’s decision to wait one more out for Herrera is not at all out of line with Generally Accepted Managerial Principles (GAMP). Yost is a bit of a lightning rod for criticism, in that when he makes a bad tactical decision, the baseball media pounces on him in a way they don’t when another manager makes the exact same move. Mike Petriello discusses this phenomenon here, and months ago Jonah Keri did the same thing. But Yost is not alone. Almost all managers prefer to give their relievers roles, and to not diverge from those roles even when it would help the team.
And in the long run, I’m no longer 100% convinced that’s a bad idea, simply because we continue to see relievers scale heights of dominance that were unthinkable even 10 or 15 years ago. And the Royals, as I’ve documented several times, are as dominant as anyone in this regard. We must at least consider the possibility that the reason why Herrera and Wade Davis and Greg Holland have been historically effective this year is precisely because they know their roles, and they know they will only have to pitch one inning at a time. I’m not saying that’s a fact; I’m saying that when enough teams manage their bullpens the same way, and when bullpens continue to get better and better, you have to at least wonder if there’s a correlation.
That doesn’t excuse Yost for what happened yesterday, though, because you can’t worry about the long run in mid-September. Shields likes to say that September is the postseason – well then, dammit, manage like this is the postseason.
Before today’s game Yost appeared chastened in this regard, saying that – from now on – he will consider using Herrera and Davis for more than three outs. It’s nice to see him closing the barn doors after the horses have disappeared over the horizon, but he shouldn’t have needed the negative reinforcement that he got yesterday to figure this out. Good managers aren’t reactive; they’re proactive.
But anyway, if that were the only issue with having Crow pitch there, I wouldn’t be as upset as I am. Because the second mistake that Yost made was that, even if you’re not going to Herrera, Davis, or Holland there, going to Crow was an indefensible decision.
Jason Frasor, 2014: 2.84 ERA, 3.32 FIP
Aaron Crow, 2014: 4.13 ERA, 5.50 FIP
Jason Frasor, 2013: 2.57 ERA, 3.37 FIP
Aaron Crow, 2013: 3.38 ERA, 4.34 FIP
Please explain to me in what sane world would you pick the second pitcher over the first? Why, because Aaron Crow was a useful reliever two years ago? Two years is a lifetime for a reliever. Crow’s strikeout rate has dropped by almost half since 2012, from 25.0% to 13.7%. His fastball has lost three mph, and continues to trend downwards. It’s not possible that Yost might not have noticed that Crow isn’t the pitcher he used to be.
Or maybe it is. Why otherwise would Yost say that he went to Crow in part because he wanted a strikeout in that situation? Crow’s strikeout rate ranks 300th among the 312 pitchers in the majors this year with 50+ innings. I want to say to Yost what old schoolers want to say to us analysts, which is “get your head out of a scouting report and watch a baseball game some time.” I don’t know how someone could have watched the Royals all season long, without even looking at the stat sheet, and not realize that Aaron Crow isn’t missing any bats. Or that he is now tied for the AL lead in home runs (10) given up by a reliever.
Crow, to his credit, did get a strikeout when he needed one; after walking Yoenis Cespedes to load the bases, he struck out Allen Craig for the second out. This does get to the one thing Crow does okay, which is get right-handed hitters out. He has always had a large platoon split; his career numbers are .229/.310/.328 for RHB, .262/.337/.457 for LHB, not at all surprising for someone who relies on his slider as much as Crow does.
The only problem: Daniel Nava is a switch-hitter. And he crushes right-handed pitching. Really, he should just give up switch-hitting and bat exclusively from the left side. His career line against LHP is .208/.285/.299; vs. RHP it’s .291/.384/.428. That’s an astounding difference.
Nava batted left-handed against Crow. In the key situation of a key game, Ned Yost let a switch-hitting batter who is helpless against left-handed pitching face a right-hander who is helpless against left-handed hitting.
Frasor, in addition to his other advantages over Crow, has a comparatively small platoon split: .230/.303/.363 vs. RHP, .244/.338/.370 vs. LHP in his career. Yost could have gone to Frasor for just one batter, since he was planning to turn the game over to Herrera, Davis, and Holland anyway.
He could have gone to Brandon Finnegan, a left-hander who had retired 11 of the 12 batters he had faced in the majors. Finnegan had pitched an inning the night before, but he was only needed to get one out.
It’s not hard to come the conclusion that Yost was spooked by what happened two months ago in Boston, what Yost has admitted was the one decision he regretted this season, bringing in Scott Downs to pitch – with two outs in the sixth inning – to Jackie Bradley Jr., only to be caught off-guard when the Red Sox pinch-hit with Jonny Gomes, who hit a two-run homer in a game the Royals lost, 5-4.
Gomes is no longer in Boston, but Mike Napoli is, and was on the bench, and presumably would have pinch-hit. But Yost may have learned the lesson too well. The problem wasn’t bringing in a left-hander; it was bringing in Scott Downs. The problem wasn’t taking out a right-handed pitcher; it was taking out James Shields. And the problem wasn’t going for the platoon advantage; it was going for the platoon advantage with Jackie Bradley Jr.
Going with Downs over Shields because you’re worried about Jackie Bradley Jr. is absurd. Going with Finnegan over Crow because you’re worried about Daniel Nava isn’t. You’re bringing in a better pitcher, taking out a worse pitcher, and facing a player who is enormously dangerous against right-handed pitching.
Maybe Finnegan isn’t ready for the big time yet. Maybe he wasn’t ready to pitch in back-to-back games yet. But Jason Frasor not only was ready to pitch, he did pitch, when the game was already out of reach in the eighth inning. He pitched a scoreless inning. And when Nava batted, Frasor got him to ground out.
Meanwhile, there are just 14 games left in the season. The AL Central is slipping away, and if the Mariners get hot, the wild card might slip away too. It’s great if Ned Yost learned from his mistake yesterday. But if he learns from any more mistakes this season, it might be 2015 before he gets a chance to apply what he’s learned.