Curtis Granderson and left-handed pitchers


And boom goes the Grandymite.

One of the biggest stories of the Yankees’ first half has been Curtis Granderson‘s emergence as not just an import piece of the offense, but as a legitimate MVP candidate. He’s currently at 4.7 fWAR and 3.6 bWAR, the sixth and 12th best in baseball, respectively. An important part of his success has been the complete 180 he’s done against lefties, tagging them for a .394 wOBA this season after producing just a .264 wOBA against southpaws from 2007-2010. Grandy’s nine homers are the most by a left-handed batter off left-handed pitchers in baseball, three ahead of Jay Bruce and at least four more than everyone else.

As weird as this sounds, it’s been a while since Curtis took a lefty deep. This shot off Brett Anderson on May 31st was his last homerun off a southpaw, a span of 35 team games. Through May 31st, Granderson was hitting .323/.373/.823 off lefties, but since then just .162/.256/.216. His strikeout rate against southpaws went from 25.8% to 43.2%. Now before you freak out, remember we’re talking about an extremely small sample here. Curtis has just 112 plate appearances against lefties this year, and just 37 have come since that homer off Anderson. That’s nothing. I’m not concerned that Grandy has reverted back to his pre-August 2010 form against lefties, but I do want to see if same-side hurlers have been pitching him differently of late.

The table on the right shows the pitch selection left-handers had been using against Granderson before that homer off Anderson and what they’ve been throwing him since. He’s still seeing the same number off fastballs, though the distribution of offspeed pitches is a little different. Curtis is seeing way more sliders and curveballs than before, but also way fewer changeups. Because of the small sample, this could mean anything. It could mean that lefties have stopped throwing him changeups, or it could just mean they haven’t faced many left-handed changeup pitchers. The important thing is that the ratio of fastballs-to-offspeed pitches is the basically the same. If they’d stopped throwing him hard stuff all together, well that would be a problem.

With some help from Texas Leaguers, let’s look at where pitchers had been attacking Granderson from the start of the season through that May 31st game we keep referencing …

That’s from the catcher’s view, so there’s a huge gaping hole down and in. Just about everything is down and away, which is not uncommon in left-on-left matchups. Granderson took just four pitches total down and in (and in the strike zone) during the first two months of the season, and there’s a pretty good chance they weren’t even supposed to be thrown there in the first place. Now let’s look at the strike zone plot since June 1st…

There’s still a bit of a hole down and in, but it’s not nearly as big. Granderson does hang out over the plate a little bit, so it could be that lefties are trying to get in on him to keep him from extended his hands. This is the called strike zone, so it could also be that Curtis is simply taking more of those down and in pitches from lefties. The swing plots do back that up a bit, though there just isn’t enough data to say anything definitive right now.

Granderson has struggled against left-handers of late, the first time he’s done so since the fix The Fix™ last August. The Yankees are going to see a bunch of lefties in the coming weeks, with series against the Blue Jays (Ricky Romero, Jo-Jo Reyes), Rays (David Price), Athletics (Gio Gonzalez, Josh Outman), and Mariners (Jason Vargas, possibly Erik Bedard) coming up. That’ll give us a chance to see Curtis take some more hacks against same-side pitchers, which will hopefully give us a better idea of whether this latest slump is just a fluke, or if the early season success was the outlier.

Categories : Offense
  • Cut your hair

    To call it a small sample then analyze pitch selection is idiocy – the short bus kind.

    • http://twitter.com/#!/billreichmann breich315

      There’s nothing wrong with looking at data and acknowledging the limitations of said data. No conclusions or predictions were made. Mike simply noted some observations.

    • Moshe Mandel

      Not when the analysis is observational rather than conclusory. Small sample size does not mean irrelevant.

    • http://twitter.com/AnaMariana42 Ana

      How is it idiocy to say “Here’s a small sample piece of information, let’s analyze it further to see if it seems like a fluke or something we should be worried about moving ahead”?

  • Cut your hair

    A small sample by definition is unreliable – for observations, conclusions, flukes. Etc. We are talking about 37 appearances. How many of each pitch type is that?

    The author himself calls it nothing. Then he wastes everyone’s time trying to figure something out.

    That’s idiocy.

    • https://twitter.com/#!/TheRealJeromeS Jerome S.


    • jsbrendog

      mike’s article: not wasting anyone’s time

      your comments: wasting everyone’s time

      • Rick in Boston

        Maybe he wants his RAB refund.

      • http://www.riveraveblues.com Mike Axisa

        It’s fine, he’s free to criticize all he wants.

    • CP

      Why do people complain about a blog ‘wasting everyone’s time’? Isn’t that exactly the point of a blog? I mean, this isn’t some life or death struggle that’s being analyzed.

      • Oscar Gamble’s ‘Fro

        When a blog is mandatory reading, and you’re forced to pay subscription rates, one has absolutely every right to bitch and moan about the blog wasting one’s time. C’mon man, get with the program.

        Small sample size or not, I do hope Grandy gets things rollings against lefties again soon. Would really suck if the progress he seemingly made was a flash in the pan.

    • http://yankeeanalysts.com Moshe Mandel

      Nope. This is wrong. A small sample of at-bats still happened, so to look at what happened and how pitchers approached him during those 37 at-bats is perfectly reasonable. If you have a small sample and limit your conclusions to just that sample without extrapolating further from it, you have done nothing that is “idiotic.”

  • Cut your hair

    Typical lawyer. You say above you can’t draw conclusions from small samples. Now you say you can if you limit them.

    The author admits the sample is nothing. That’s the most accurate statement in the whole post. Everything else is like deigning intensions from the weather.

    There are any number of reasons the numbers look the way they do having little to do with Granderson. That’s what it means to try to explain noise.

    And now I have wasted enough of my time. Have fun explaining Jester’s success leading off the game.

    • http://yankeeanalysts.com Moshe Mandel

      You can’t draw extrinsic conclusions from small samples, but you can certainly try and explain what happened in the sample itself. That is quite obviously what I meant by “observational.”

      And way to go with the “typical lawyer” comment. Good stuff. By the way, the little blue button in the lower right-hand corner that says “reply”? Yeah, that’s for replying.

    • http://yankeeanalysts.com Moshe Mandel

      Just to add to this, by your logic, breaking down a single at-bat never makes sense. If Granderson took a 1-2 slider in the dirt, and then struck out on a high fastball, and he said afterward in an interview that the slider set him up for the fastball, you’d say, “Pure noise!!!” Right? Because that’s basically what Mike did here, on a slightly larger scale.

  • Cut your hair

    One only needs to read the comments here to see how quickly idiocy turns into conclusions. Hell there are conclusions in the post based on what 20 or 30 off-speed pitches? And the distributions change because of Granderson and not the pitchers?

    Are you going to say from one at-bat a hitter has general difficulties?

    Dumb, dumb, dumb…

    • http://yankeeanalysts.com Moshe Mandel

      “Are you going to say from one at-bat a hitter has general difficulties?”

      No, and Mike didn’t do that in the post. There isn’t a firm conclusion in there anywhere, just noting the data and presenting the various ways it might be explained, with constant caveats about the sample.

    • I Voted 4 Kodos

      Some lines from the article:

      “Curtis has just 112 plate appearances against lefties this year, and just 37 have come since that homer off Anderson. That’s nothing. I’m not concerned that Grandy has reverted back to his pre-August 2010 form against lefties, but I do want to see if same-side hurlers have been pitching him differently of late.”

      “The swing plots do back that up a bit, though there just isn’t enough data to say anything definitive right now.”

      “That’ll give us a chance to see Curtis take some more hacks against same-side pitchers, which will hopefully give us a better idea of whether this latest slump is just a fluke, or if the early season success was the outlier.”

      I see Mike going out of his way to say that one cannot draw any conclusions from the data. He examines the numbers/graphs to see if pitchers have changed anything lately then says that, while it is not informative in and of itself, it is something to keep an eye on in the future. You keep saying that it’s stupid to draw conclusions from small samples, but I can’t see where the article really draws any significant conclusions. It just gives up something to look for in upcoming games.

  • Cut your hair

    The mobile interface makes replies non-obvious, Perry Mason.

    • Jim S

      I thought you were done commenting. Pity.

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Ridiculous Upside the Elder

      Why haven’t you linked us all to your blog yet?

      Because it sounds awesome.

  • Cut your hair

    Read the last graph. It uses words like struggled and slump. That’s a conclusion tied to Granderson. The words fluke and outlier then show a basic misunderstanding of the concepts.

    • http://yankeeanalysts.com Moshe Mandel

      Again, those are intrinsic conclusions. He has absolutely struggled and slumped against lefties of late.

    • Jim S

      I can play this game too. You used words like “struggled”, “conclusion”, “misunderstanding” and “concepts”.

      Clearly you’re misunderstanding the conclusion and struggling with the concepts.

      Honestly? “Uses words like?” Are you in a middle school english class?

    • Mike HC

      Think of the article more like tracking Granderson’s performance against lefties. Granderson’s struggles against lefties has been a topic people like to talk about and analyze since he got traded here.

    • http://www.riveraveblues.com Mike Axisa

      .162/.256/.216 in his last 37 PA against lefties is a slump. I made no firm conclusions about why he’s slumping, which you seem to believe I have.

    • I Voted 4 Kodos

      Saying that he’s struggling/slumping against lefties at the moment is stating a fact. Small sample or not, he hasn’t been successful against lefties of late. Mike doesn’t examine all of the data and then conclude that Granderson is slumping/struggling. Those are just words that relay what has happened recently.

      Drawing a conclusion would be saying he has been struggling because of the change in the way pitcher have been pitching him.

      Take issue with the words struggle and slump all you want, but the author clearly is not drawing any sort of conclusions about Granderson.

  • Cut your hair

    Quiet children. The adults are talking.

    • Jim S

      And now I have wasted enough of my time.

      Have you?

    • Rick in Boston

      I’m enjoying how you respond so well to Moshe’s arguments.

  • Cut your hair

    Absolutely, huh? Now you’re falling into the same causal trap.

    Has he also struggled and slumped hitting inthe 4th inning of games this year?

    • https://twitter.com/#!/TheRealJeromeS Jerome S.

      Do you see that “reply” button on the bottom right of every comment?

    • I Voted 4 Kodos

      He’s hitting .185/.313/.333 in the 4th inning this year, so yes, he has struggled. I feel comfortable saying that.

      I cannot conclude why he has struggled, but I can say with certainty that he has.

      • http://www.riveraveblues.com Mike Axisa

        Don’t even bother, he’s just arguing for the sake of arguing while simultaneously missing the point.

        • I Voted 4 Kodos

          You’re absolutely right, I’ll give it up now. I’m bored myself so I went with it longer than I should have.

    • http://yankeeanalysts.com Moshe Mandel

      Yes, he has. You are mistaking observation for conclusion. To assert that he has not struggled against lefties recently would be blatantly false. There is no causal trap here because no causal relationship is being suggested.

  • Cut your hair

    And now he’s outright calling 37 non-continuous appearances a slump. Basic concepts of small samples truly have little meaning here. I just hope Granderson turns it around in the 4th inning.

    • I Voted 4 Kodos

      What number of at bats constitutes a slump? 42 at bats is certainly not too small of a sample for a slump. Isn’t a slump simply a poor performance over a small sample?

  • Cut your hair

    And now the idiocy has reached its logical conclusion. Causality is being assumed from his performance in the 4th inning.

    Lets see the Hahvahd lawyer wiggle out of this one.

    • Jim S


      Seems there are new commenters every day who argue for the sake of arguing and don’t read anyone’s arguments.

    • http://yankeeanalysts.com Moshe Mandel

      I have no idea what you are talking about, quite frankly. He’s struggled in 4th innings. I’m not saying it is anything more than a statistical anomaly or that it would continue going forward. I am not assuming any sort of causality whatsoever. Explain what you mean by “causality is being assumed.” What sort of causality?

  • Cut your hair

    Hah! You just said he has absolutely struggled – drawing a causal conclusion from random data. When presented with an even more absurd slice you stick to you guns. That’s funny and sad.

    • http://yankeeanalysts.com Moshe Mandel

      Struggling isn’t a causal conclusion.

  • Cut your hair

    Children –

    What are your hypotheses for why Granderson has struggled and slumped in the 4th inning this year?

    • http://yankeeanalysts.com Moshe Mandel

      No one has given any hypotheses, because everyone knows it is a statistical fluke. Now you are just trolling.

      • Jim S

        I think he crossed the troll line way back near the beginning of the thread.

  • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Ridiculous Upside the Elder

    Stop hiding behind your phony numbers, Kabak. You know those sample audiences are flawed, they don’t take into account households with more that two television sets, and other things of that nature.

    I hate you, Ben Kabak, I hate you.

  • Cut your hair

    So one is a fluke to be dismissed and the other means he’s absolutely slumping? Poor boy now you’re in no man’s land. But it is fun watching you work this out in real time. Go back and read my first comment then pretend the post was about pitch selection in the 4th inning.

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Ridiculous Upside the Elder


    • http://yankeeanalysts.com Moshe Mandel

      No, no one said he is slumping. We said he has slumped over those at-bats. And if you can’t understand the distinction between the 4th inning thing and the lefty thing which has plagued him for most of his career, you may not be quite as intelligent as you think. Now that you’ve shown a lack of understanding for basic statistical concepts (as you did regarding the Jay Jaffe article a while back, if Im right about who the guy lurking behind that screen name is), this really has become a waste of time. Enjoy talking this over with all the strawmen you created.

  • Cut your hair

    Do you know what a time series is? Neither data set is continuous. They are a small bunch of data points averaged together because we can not because we should. And you should know better. In fact his career splits say he should be excelling in the 4th inning. So that’s even more worrisome.

    Its all idiocy is what it us. Shame on you for digging in.

  • Will (the other one)

    Sorry, don’t mean to feed the troll, but who’s the “Harvard lawyer” he keeps referring to? Is this a reference to Ben? Moshe? Engineer Axisa? I’m confused.

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Ridiculous Upside the Elder

      That’s my favorite part of all this; he apparently thinks Axisa is the law student (and not Kabak). Or maybe he thinks all three of the RABbis are lawyers.

      Cut your hair’s complete dedication to being thoroughly wrong about everything is quite impressive, I must admit.

      • http://www.riveraveblues.com Joe Pawlikowski

        No, he’s pretty clearly referring to Moshe, who is a Harvard law grad.

        • http://www.secondavenuesagas.com Benjamin Kabak

          Yeah. Let’s get it straight. I went to NYU Law, and Moshe went to Harvard. Don’t associate me with those dirty Bostonians. Jeez.

        • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Ridiculous Upside the Elder

          Ah, gotcha. I missed that distinction in all of his crazy ramblings what with his inability to use a reply button, my bad.

  • Rainbow Connection

    Wow. People get trolled so easy. Just ignore him.