Playoff Prospectus: PE... (10/09)
the archives are now free.
12 comments have been left for this article. BP Comment Quick Links
| |
|
|
October 9, 2015 Playoff ProspectusThe Royals' Unforced Error: ALDS Game 1
Alex Gordon batted eighth for the Royals in Game 1 of the ALDS, which is insane. There’s no way around that. Alcides Escobar batted first, which we all more or less knew would happen, but which remains insane. Perhaps more criminally, though, Ned Yost put Salvador Perez in front of Gordon in the lineup, batting seventh, which accomplished two things: 1. Breaking up the sequence of six Royals hitters with seasonal True Averages between .289 and .301; and 2. Creating roughly a 78-percent chance that Escobar would bat one time more than Gordon during the game. If those sound like things the Astros, not the Royals, should have been trying to accomplish, well, yeah. Ned Yost put his team in a worse position from the get-go on Thursday night. Batting order makes very little difference; we know this. In any given game, though, it can find ways to matter, and the chances it will hurt depend on the egregiousness of any errors one makes in putting the lineup together. Yost’s mistake—putting his natural first and eighth hitters in each other’s places—was about as egregious as they get. The reason Yost did it—the Perez-Gordon misfire, for sure, and maybe the Escobar-Gordon part, too—was that Collin McHugh started for the Astros, and McHugh had a reverse platoon split this year. Does that sound familiar? It seems as though there are more of these guys than ever, these reverse-split right-handed starters. And you know what? There are! Or at least, there are more of them than there have been at any time in the last 10 years. Of 107 right-handers who pitched at least 100 innings this year, 43 had a zero or reverse platoon split (we’re using raw OPS here; forgive the primitive tools). That’s 40.2 percent of the population, which continues a three-year trend of righties having unprecedented relative success against left-handed hitters.
That’s an interesting thing to look at another time. It seems eminently possible that that’s a phantom phenomenon, driven either by a change in the broad-spectrum quality of left-handed hitters league-wide; some strike-zone irregularity that affects different hitters and pitchers differently; or the randomness that governs so much of our existence. But for now, the numbers are there, and teams react to the numbers. Yost had some thin justification, some cover, for building his lineup the way he did, even if it remained a wrong decision. By the time the Royals even had a chance to mount any attack with their imperfect lineup, though, they were down by two runs. The Astros—without any easy outs, let alone a near-automatic one, at the top of their order—got to Yordano Ventura early. The pivotal plate appearance of the first inning was, one might argue, Carlos Correa’s. Ventura allowed a quick single to Jose Altuve, then got stuck in a nine-pitch confrontation with George Springer that ended with a walk. With the crowd roaring and Ventura seemingly trying to find a comfort zone, Correa took the first four pitches of his at-bat. Two of them were strikes, but by showing the willingness to wait out a deep count and work against Ventura’s high-octane stuff even in a two-strike situation, Correa set up the pitcher well. He fouled off a couple pitches, then hit a rocket of a liner to right for a single. The inning turned. Ventura did fairly well from there, but ‘fairly well’ from a starting point of the bases being loaded with nobody out still means allowing two runs to score. The rain delay made this game and series more interesting. It was a huge blessing for Yost, who got from it an excuse to remove Ventura (who was fooling no one, but hadn’t been bad enough to proactively pull after two innings) and bring him back on short rest in Game Four. Chris Young came on and either pitched really, really well, or benefited from a sort of reverse R.A. Dickey Effect, or simply took advantage of starting with Correa and working downhill through a Houston lineup that really isn’t so scary in the lower half. He finally got in trouble when he faced Altuve and Springer, and Yost probably should have pulled him then, before Young had to wade through the order a second time. Instead, he asked him not only to finish that inning, but to come back out and handle the sixth. Young did that, and didn’t give up any more runs, but it took a double-play grounder from Jason Castro with two on and one out to get through it. That was the first double play Young had induced since May 22th. Yost could have gone to Danny Duffy or Kris Medlen somewhere in there, especially knowing he would be bringing Ventura back for Game Four. That particular error was small, though, and didn’t bite him. McHugh ended up losing a full hour between official pitches, though he threw during the delay to stay loose, but he responded quite well to that. He gave up one solo homer before the delay and one after it, both to Morales, and he got out of there after six innings without having to face Mike Moustakas, Perez or Gordon a third time. (Instead, Gordon got Tony Sipp with two outs and no one on in the seventh.) The game rolled into the late innings with Houston holding a two-run lead, and it never felt like the Royals had a good chance to close that gap. From there, the bullpens did mostly what these bullpens both do. Colby Rasmus jumped a first pitch from Ryan Madson to widen the Houston lead, just the way he jumped a first pitch from Masahiro Tanaka to give the Astros an early lead in the Wild Card Game. From now on, if Rasmus leads off an inning, he’s worth a first-pitch curve at the ankles. Madson giving up a home run has to set off a few alarm bells for Royals fans, because he’s been Yost’s highest-leverage relief horse lately.
The real problem, though, is that Kansas City just lost a game to an Astros starter not named Dallas Keuchel, at home. The series is now set up very differently. The Astros might have just swung the pendulum. Having Ventura set for Game Four is nice, but the Royals need to get their lineup back in order, keeping those Super Six chained together. The Astros are proving that they’re ready to punish mistakes in the way that really counts: by putting them in the seats. The Royals can do that, too, but they’re not going to match Houston there, so they need to be ready to punish Astros hurlers who get into trouble. To do that, they need to have those great hitters in sequence, and accept that empty Perez-Escobar-Rios innings will be offset by the crooked numbers they can put up in other frames.
Matthew Trueblood is an author of Baseball Prospectus. Follow @MATrueblood
|
Player Profile: Drew Hutchison
FWIW, I have Mchugh with an estimated true platoon split of 14 points in wOBA, which is slightly smaller than the average RHP.
I use historical stats, adjusted for the RH and LH talent of the batters faced, as well as THEIR platoon splits, and I use arm angle and pitch repertoire in a regression formula to establish the baseline toward which to regress.
So I think my platoon estimates are pretty good. Very few of those pitchers who exhibited reverse splits for one season have true actual reverse splits. An easy way to see that is to simply look at that group in another year.
My guess is that of the 40% of the population that shows reverse splits, 80% of them will show positive splits in any other year.
A RH SP typically faces around 300 to 400 lefty batters in one season. You have to add around 2000 PA of league average platoon splits to estimate a RH pitcher's true platoon split. That means you are regressing a one year split for a full-time RH starter around 83% toward league average!
As you can see, it is almost impossible to estimate a reverse split for a RH pitcher from one year stats. It also means that almost all of the actual one-year splits you see in a pitcher is noise.
In fact, you can safely IGNORE one year splits and just assume that a RH pitcher has a true league-average platoon split. That's A LOT better than using the actual split to assume his true split. When you regress something 80 or 85% of the way toward league average, you might as well just assume league-average.
I get so tired of hearing about a pitcher's one year splits on TV. And then, they quote batting average against. For a pitcher! Take a worthless stat to begin with, BA against for a pitcher, and make it even more worthless (if that is possible) by quoting a one-year split. Might as well quote a split for pitchers based on the phases of the moon and assume that that has meaning in the present or future too.
For a left-handed pitcher, the regression is about half as much. Of course, they typically face many fewer LHB so your sample size in one season is a lot smaller. So for a full-time lefty, you would regress maybe around 50%, so one part actual splits and one part league average. For a full time lefty starter.
Don't even talk about one-year reliever splits...
Interesting. Are your "true splits" publicly-available?
Nope
How large a sample do you need to stabilize the data, do you think? Is it possible to tell If the magnitude of the split has been reduced, league wide, over recent years (and with the more frequent use of shifting the infielders against lefties)?
There is no such thing as "a certain sample size to stabilize the metric."
Yes, it is possible to tell if that is the case. I have not done that, but I suspect that if it has changed, it has changed very little. The two factors for magnitude of splits is arm angle and pitch repertoire. I doubt that arm angles have changed recently and I don't know whether pitch frequencies have changed much.
If anything with more and more reliever "specialists" you would think the average platoon split would have increased and not decreased.
I don't know that shifting would have anything to do with platoon splits, but you never know.
Thank you for the quick reply. I guess what I'm wondering is how large a sample would you need to reach a reasonable confidence level in the results? I'm not a statistician, scientist, etc., and my recollections from my one experimental design class in grad school are hazy at best.
From your description above, we find that platoon splits exhibited in one season tend not to repeat. How many seasons/batters would you need in your sample to be confident in the result for an individual hurler, or is it truly, "You can never have enough" regardless of arm angle and pitch repertoire. I can accept the latter, just verifying I understand the answer.
Regarding the shift, I am making an unfounded hypothesis that, league-wide, RHP's platoon split has narrowed over the past few years. If teams are shifting their infield defense against lefties, and it is effective at reducing BABIP, then this would improve their effectiveness against the opposite-handed batters, but have no impact on effectiveness vs. same-side batters. This new baseline should be reflected in a larger number of RHPs having a reverse-split (particularly if the variation is so large that you can never have enough at bats to be confident in the results).
For LHPs, the shift would improve performance vs. same-side batters, so we would tend to see a wider platoon split than in years before the shift was frequently exercised.
Of course if possible, one might try to analyze the data with 'team' as a separate variable, to see if these hypothesized results are more evident in teams that shift more frequently.
And of course these sample sizes may be so low that any effect is indiscernible, on a league-wide or team-wide basis. A large sample would be our friend, as a classmate of mine once proclaimed, "The n's justify the means."
Have a great weekend!
Certainty of the estimates increases with sample size AND with other non-empirical data like arm angle and pitch repertoire. For example, if a pitcher throws like Brad Ziegler, even without any empirical data, it is almost guaranteed that he will have a large split. Similarly, if a pitcher throws tons of sliders like Romo, it is almost guaranteed that he will have a large split.
Here's where people go wrong though. The certainty of estimates has virtually no bearing on strategy. If what you are estimating is relevant to a strategy decision, then you must use the estimate whether there is much certainty in it or not.
For example, say a left-handed pitcher comes in and you know nothing about him other than he has had a huge reverse split in around 50 IP or the entire season so far. Another lefty comes in and he has an average split in around 1000 IP. A third lefty comes in and you also have 50 IP data on him and he has a huge PLUS split.
In the first instance you pretty much estimate his true splits at around league average for a lefty or PLUS 29 points despite his huge negative split for the entire season (you know nothing about him prior to that). Maybe your estimate is only plus 27 or 28 because the huge negative in 50 IP does count for SOMETHING.
In the second case, you of course estimate his true splits to be the same as he has been for the last 15 years - league average.
In the first case, you are not very confident of your plus 27 estimate. In the second case, you are VERY confident of your plus 29 estimate. In the third case, like the first one, you are not very confident of your +30 estimate.
So, all 3 pitchers have around the same expected splits - around PLUS 29 points. But, pitcher one has had HUGE NEGATIVE splits (he actually pitched better v. RHB) all season and that's all you know. Pitcher two has league average splits his whole, long career. Pitcher three has had HUGE positive splits all season long and that's all you know. He has been deadly on LHB and/or RHB have murdered him.
Do you respond to any of these pitchers and differently? No! You treat them exactly the same. You MUST assume that they all have around the same true platoon split. That when you send up you batter to the plate, a RHB will do much better than an equally talented LHB. Against all 3 pitchers. Yes, you are MUCH more certain that that is the case with pitcher #2, but certainty is irrelevant in your decision of whom to bat against him.
This is counter-intuitive but true.
Now if you had any more information on these players such as arm angle (which you obviously do, but for this exercise we assumed no other information or data), that information would have MUCH more of an effect on your estimate for pitchers 1 and 3 then for pitcher 2. That is where the certainty comes into play. Or if you were deciding which pitcher to acquire and you wanted a LOOGY (or not) then the certainty might come into play in your decision. But in terms of how to respond to this pitcher, certainty is irrelevant.
MGL, your adjustments are great for explaining platoon splits for individual pitchers but I don't see how it could explain an increase in reverses splits for RHP across baseball. That would require a different explanation and increased RHP effectiveness vs. LHB due to the shift seems like a good one.
My models have nothing nothing to do with explaining the decrease in splits for RHP.
Could be pitcher repertoire (maybe more changeups or 2-seamers). Could be average TBF per pitcher. If the number of batters faced per pitcher decreases (i.e., more specialists or starters pitching to fewer batters) then the number of outliers by chance will increase.