Mar
01

Lining a pitch up the middle is not progress

By

Curtis Granderson can help a ball club in a number of ways. He can play a solid center field. He can hit for power. He can get on base at an above-average clip, and afterward he can motor around the bases. But as we learned in 2009, the mere ability to do something doesn’t necessarily bring results. Granderson can get on base at an above-average clip, but in 2009 he didn’t. Since he’s done it before, we don’t say that he can’t, but rather that he didn’t. In the same way, just because he hasn’t hit lefties well in the past doesn’t mean he can’t.


Photo credit: Kathy Willens/AP

It appears that questions about Granderson’s ability to hit lefties will surround him this spring. We’ve already heard stories about him working with Kevin Long on ways to better approach lefty pitchers. Those are encouraging, but far from a sign that Granderson will put his woes behind him. Still, that won’t stop certain newspapers from taking one small detail and making a huge deal out of it. Apparently, according to the NY Post, hitting a line drive in batting practice off Kei Igawa represents progress. Oh, the crazy storylines of the spring.

I agree with SG’s statement that Granderson will probably never hit well against lefties, but that he can certainly outperform his career numbers to this point. One aspect of splits that often goes unmentioned is the small sample they provide. Granderson has come to bat 2,896 times in his career, but only 24 percent of those have come against lefties. That leaves his total at 685, or just under a full season’s worth of plate appearances. We don’t judge a player based on a single year of his career, so why would we judge Granderson based on one year’s worth of data?

Because of this small sample against lefties, we need to regress the figures in order to get a better idea of Granderson’s true skill against lefties. Thankfully, Matt Klaassen of FanGraphs already did the calculation, which adds 1,000 plate appearance of league average splits to Granderson’s existing 685 PA. If Granderson hits to his CHONE projected .359 wOBA, we can expect that to be .374 against righties and .311 against lefties. That .311 number is still below average, and it’s even blow the .323 wOBA he posted against lefties in 2008. It is, however, a bit more optimistic than CAIRO, which pegs him at .299 against lefties, and certainly better than his .266 career wOBA split.

Deliberate and focussed practice could help Granderson even further against lefties. Again, echoing SG, we can’t really expect Granderson to make leaps in his split figures, but it’s certainly possible. We saw such a transformation first hand in Paul O’Neill. While in Cincinnati O’Neill hit lefties poorly, routinely posting OPS numbers in the .500 range. Even in his first year with the Yankees he struggled against lefties. But then, at age 31, he came around, posting OPS numbers in the .700 range or better for the next five seasons. We might set ourselves up for disappointment by expecting Granderson to make the same transformation, but we know that it is certainly possible.

The discussion of Granderson’s skills and results against lefties reminds me of what a good all-around player he is. We’re not questioning his defense, his speed, or his ability to get on base. Instead, we’re focussed on his performance in fewer than a quarter of his plate appearances. I think we’ll all be satisfied with his performance in the other 75 percent, and maybe even a bit more satisfied with that 25 percent sample.

Categories : Offense

127 Comments»

  1. Steve H says:

    Should he consider giving up switch hitting altogether?

    /PP’d

    • bexarama says:

      you beat me to the joke :(

      I have faith in KLong, though I’m not expecting Grandy to be a monster against lefties.

  2. JPR says:

    He is not a switch hitter, he bats left and throws right. I dont really care about the hit up the middle but I did like what Granderson was saying about how he sees that he was moving his head when hitting lefties.He is trying to hit lefties and righties with the same swing now….the same setup, which is the way it has to be done, because whether the ball comes out of a left hand or a right hand it still has to come over the plate…Cano is a savant hitter…he hits stuff all over the zone.They are paid to hit the ball, not the pitcher.It seems like he understands that and does not need to pull every ball down the right field line. Noting he is a notorious pull hitter and hitting to the opposite or center field is a good deal for him. He was attempting to have different swings like Jeter or Damon…there are not many guys like that.

  3. Ben says:

    You bring up a good point. I’m desperately hoping though that Granderson improves against lefties this year. By the way, when will the next chat likely be?

  4. Templeton "Brendog" Peck says:

    ugh, when ir ead that article in the post i was on the crapper and without thinking about it went oh come on!! the guy in the next stall must’ve been like WTF!??!?! haha

  5. Jamal G. says:

    The following line annoyed me greatly in that article:

    The Yankees wouldn’t have made the trade for Granderson if Long couldn’t [solve his struggles versus southpaw pitchers].”

    I am sorry, but I refuse to believe that Brian Cashman would have not traded for *the most valuable center fielder in Major League Baseball over the past three years that is not named Grady Sizemore because of a flaw in his game that he has dealt with whilst producing to such a level.

    *Fangraphs has Granderson as the fourth-most productive CF over the past three years, but I discount Ichiro because he has been a full-time center field during just one season in the time frame (2007); and factoring in B-Pro’s baserunning metrics, Carlos Beltran (15.1 WAR) trails Granderson (15.5) and Sizemore (15.6).

  6. mko says:

    In my opinion you are often correct in pointing out small sample size when other writers jump to quick conclusions. But I just don’t think that 685 PA or 25% of his whole career are a small sample size anymore…

    • Templeton "Brendog" Peck says:

      do you consider one season’s worth of at bats a small sample size? as in, this guy has played 5 seasons but his 5th one was awful while his first 4 were hall worthy? is that 5th seasson of ~600+ atbats a SSS?

      then why isn’t the accumulated atbats against lefties when it equals the same number of atbats?

      • Eh… I mean, the guy’s been an MLB regular for 4+ seasons already. It’s not like anyone is looking at an arbitrary situational stat that rarely occurs and overemphasizing its meaning. I think pronounced LHP/RHP splits, over the course of 4+ years, are relatively reliable and meaningful.

        (That’s not to say the guy can’t outperform his career numbers in this particular discipline (hitting LHP).)

      • Snakes on the mother effin plane says:

        Where this comparison breaks down is that a single season is confined to precisely that: a single season. Lots of things can explain a single down year in an otherwise great career – an injury, a personal problem, contract distractions, trade talk distractions, whatever.

        But a similar # of poor PAs spread relatively evenly over the course of 5 years is most likely explained, in this case, by not hitting lefties well. Full stop.

  7. X says:

    i would not consider 645 pa against lefties a small sample size at all. You are reaching in order to argue your theory.

    • Read again. I said 645 PA is about one season’s worth of data, and we don’t judge players based on one season.

      • Riddering says:

        Maybe you don’t, sir, but we outside of the blogosphere believe in a little thing called what my eyes saw last week.

        /village idiot’d

      • Mike HC says:

        “Because of this small sample against lefties”

        But I still agree that it is a small sample size. Granderson has only played 4 full seasons in the major leagues. It has become obvious that hitting lefties is a weakness, and he is now beginning to really focus on improving. There is no reason to believe that Granderson will not be able to get better at this at this early stage of his career.

      • Read again. I said 645 PA is about one season’s worth of data, and we don’t judge players based on one season.

        See also: Nick Swisher’s 2008 White Sox campaign, Javy Vazquez’s 2004 Yankees season.

        • Steve H says:

          Javier’s a nice pitcher. He had a great year in Atlanta, but the AL East is a different division. If I was the Yankees, I’d look at what he did in 2004 and wonder “Are we getting that Javy, or the Javy from the other 11 seasons of his career?” It’s tough to say. He’s definitely an upgrade over what they had in the back of the rotation last year, but they have to hope he pitches well so they don’t have to keeping starting guys on short rest in the playoffs again

        • Mike HC says:

          I think it has more to do with improving though, than luck. Dismissing or diminishing the predictive value of a single season is usually due to variations in luck. Here, I don’t think Granderson has been unlucky. Just inexperienced.

        • X says:

          Swishers 08 campaign consisted of hitting both lefties and righties. He’s arguing that 645 PA against JUST left handeed pitchers, spread over 4 years is not enough to judge how badly Granderson sucks against them, which to me is contrary

          • The point

            ————-

            your head

            It doesn’t matter that Swisher’s 588 poor PA’s occurred against both righties and lefties in one single year while Granderson’s poor 645 PA’s occurred against only lefties over 4 calendar years.

            The point is, both of them are fairly small sample sizes. Whether you’ve only slept with four women total in one year or four women total over the span of four years is a secondary consideration to the main fact that you’ve only slept with four women. You’re inexperienced. Your sample size is too small to say that this is what all women are like (or what all your plate appearances will be like) going forward.

            • A.D. says:

              Exactly

            • andrew says:

              It actually does matter though. Because Swisher’s poor 2008 was only a portion of his career. Swisher’s 2008 was the outlier. Granderson has NEVER hit lefties well. His poor stats against lefties have never been good. Last year was not an outlier in a much bigger sample. The ENTIRE sample claims that he is not good at hitting lefties. There is a big difference between dismissing one season out of five because it is merely 20% of the sample and dismissing 100% of the sample.

              • No one’s dismissing a sample. It’s clear that Granderson has had issues facing lefties. What I’m arguing is that it came in a small enough sample that we can’t say that it’s indicative of his skill level.

                • andrew says:

                  I know there’s a chance he’s improving, Joe, and will one day be better against lefties. I am not arguing with your logic. I am arguing with the idea that dismissing Granderson’s sample is the same as Swisher’s. Swisher’s sample was small because it only made up 20-25% of the data on him. With Granderson, we are looking at 100% of the data, and still dismissing them.

                  Yes, there is a chance he improves, but dismissing 100% of the sample size because it is as big as Swisher’s 2008 is not correct.

                • I am arguing with the idea that dismissing Granderson’s sample is the same as Swisher’s.

                  Nobody is arguing that Swisher’s sample and Granderson’s sample are identical. What we are saying is they do have similarities. And they do.

                  Swisher’s sample was small because it only made up 20-25% of the data on him. With Granderson, we are looking at 100% of the data, and still dismissing them.

                  False. Granderson’s sample is also only 20-25% of the data on him. That’s a fact.

              • Templeton "Brendog" Peck says:

                dismissing one season – 600 abs

                dismissing grandersons splits vs lefties – 600 abs

              • X says:

                thank you

              • Swisher’s 2008 was the outlier.

                Agreed.

                Granderson has NEVER hit lefties well.

                Swisher’s 2008: an outlier
                Granderson’s career against lefties: also an outlier

                His poor stats against lefties have never been good. Last year was not an outlier in a much bigger sample.

                Granderson’s tripleslash against lefties is an outlier in the much larger sample of his entire career against both lefties and righties.

                The ENTIRE sample claims that he is not good at hitting lefties.

                No, the entire sample says he’s good at hitting. The small sample against lefties says he’s not good at hitting lefties, but it’s still a small sample that disagrees with (i.e. is an outlier for) the larger sample that says he’s a damn good hitter.

                There is a big difference between dismissing one season out of five because it is merely 20% of the sample and dismissing 100% of the sample.

                No, not really, there’s not.

                • Jack says:

                  Granderson’s tripleslash against lefties is an outlier in the much larger sample of his entire career against both lefties and righties.

                  So, his career is defined by a large sample? Is that what you’re saying?

                • andrew says:

                  Granderson’s career against lefties: also an outlier

                  Really? explain to me the outlier? He has never hit lefties well. Where is the outlier?

                  No, the entire sample says he’s good at hitting

                  But that’s not true. Hitting righties and hitting lefties are not in the same sample. They are different, and the stats bear that out.

                  Dismissing a sample that is 680 PA’s out of 680 PA’s is not at all the same as 650 PA’s out of 2500 PA’s.

                • Big Juan says:

                  Well played, Jack. Well played.

                • Dismissing a sample that is 680 PA’s out of 680 PA’s is not at all the same as 650 PA’s out of 2500 PA’s.

                  But it’s only 680 PAs out of 680 PAs because you’re arbitrarily limiting it to 680 PAs. Nobody is forcing you to limit it in this way. Curtis Granderson does have a larger sample size. It’s called his career, which currently has 2,896 PAs worth of data. They count too.

                • Really? explain to me the outlier? He has never hit lefties well. Where is the outlier?

                  The outlier is he hits well, so that data set (his tripleslash against lefties) where he doesn’t hit well lies outside the normal data clustering for him. He’s normally a good hitter, in this situation he’s not a good hitter, so said situation is an outlier.

                  His career tripleslash is .272/.344/.484. His lefty split of .210/.270/.344 is thus an outlier. It doesn’t matter that it didn’t happen all at once, it also doesn’t matter that there may (and I stress MAY) be a valid reason for it being an outlier, it’s still an outlier.

              • Mike HC says:

                Very true, but it was also the first 645 at bats of his career against lefties. In which he had a 178 at bat stretch in 2008 where he slugged .739 against them. Showing he has potential to hit well enough against lefties, for a lefty hitter in the future.

                The 645 at bats is a small sample because of his inexperience. It would be the equivalent of judging a players future performance against righties based on their first 646 at bats, which would come in only their first season and a half.

    • Jamal G. says:

      Actually, when dealing with splits against a southpaw, a sample of ~1,000 plate appearances is needed in order for the results to become reliable for a same-handed hitter (~2,000 for RHB); so, in this case, yes, the amount of plate appearances that Granderson has against southpaws are inadequate.

      • Can you provide some sort of substantiation of those numbers? I’m not arguing with you, I’m just interested in how those numbers were ascertained, I think it’s interesting.

        • Jamal G. says:

          Sorry, the best that I can currently do is cite page 163 of The Book:

          A right-handed hitter needs around 2,000 plate appearances against left-handed pitchers before his measured platoon spltis can be reliable (in other words, using the measured platoon split is more accurate than assuming has an average split). From a practical standpoint, right-handers are best assumed to have average platoon skills, unless one is willing to make the calculations needed to accurately estimate a player’s platoon skill. For lefties, the number is about 1,000, which means that only veteran starters have reliable platoon splits.

          • Thanks. I don’t think I buy the notion that a player like Granderson’s platoon splits, at this point in his career, aren’t a somewhat reliable indicator of his ability in the near future to hit LHP, but obviously I’m totally open to being shown that my opinion is faulty.

            Is the question – the gulf between those saying this is too small a sample size and those saying it’s not – about whether his platoon splits at this point in his career are a good indicator of how he’ll perform for the rest of his career vs. how well he’ll perform in the next few years? I’m afraid I’m not communicating this clearly… It just seems strange to me that we’d disregard the predictive ability of 4+ years of pronounced platoon splits from a big league regular just because we haven’t hit that 1000 PA marker. I would get it if the point (that you outlined above) is that we can’t predict what will happen for the rest of a guy’s career when the guy has less than the 1000 PA marker, but can we really not get some idea of how the guy will perform in that particular situation for the near future?

        • Jamal G. says:

          I do realize that instead of asking you to take my word for it, I am asking you to take another’s; however, I hope that the “word” of Andrew Dolphin, Mitchel Lichtman and Tom Tango hold a bit more weight in your eyes. ;)

      • Andy in Sunny Daytona says:

        Just being a Devils’ advocate, but what if Granderson does show improvement over the next +300 plate appearances, do we use the data from the entire 1000 and judge him on that, or only based on the last +300?

        • Zack says:

          I think you have to use both.

          Acknowledge he struggled against LHP in the past, but he also had a good 300 PA. Then you have to wait and see what he does next year to see if it was just one year or see if it was a change in his approach that brought better results.

  8. A.D. says:

    Apparently, according to the NY Post, hitting a line drive in batting practice off Kei Igawa represents progress.

    If only all the lefties he faced were Igawa he’d probably crush lefties (as would everyone else)

    • Thomas says:

      I actually checked earlier how well Granderson hit Igawa remembering that horrible Igawa-Det game.

      Granderson actually didn’t face Igawa. Leyland thought it was a better plan to start Raburn in center. Granderson did come in the game in the 6th, so I assume Leyland didn’t want Granderson facing a lefty. If I was Granderson, I’d be pissed I didn’t get to face Igawa (or embarrassed my manager didn’t have enough faith to let me face Igawa).

  9. Riddering says:

    Joe, this post does a great job of putting Granderson’s lefty struggles in perspective. Before, during and after the trade the majority of talk about Grandy has been what he will do this year against LHPs. While that’s important and we all hope for an improvement it’s true that the intense focus has made the issue bigger than it has to be when you consider the entire package he brings as a center fielder.

    You know what I love most about the SSS in regards to Granderson and lefties? The fallacy of claiming that Gardner hits them so much better with his career 91 PAs.

  10. 28 is on the way says:

    Why are you always scrutinizing other people’s work? I’ve noticed that RAB likes to break down other people’s writing and go on and on about how this person is wrong. Go get a job at a newspaper, some of these writers may not have the time to go through Fan Graphs for every piece they write. It’s spring training, give the guy a break he’s just trying to write something optimistic for New Yorkers to read on they train ride to work. Sheesh.

    • Steve H says:

      some of these writers may not have the time to go through Fan Graphs for every piece they write.

      They should do what they need to do to be good at their job, right? If that includes reading Fan Graphs, that includes reading Fan Graphs. If they are writing on any subject, shouldn’t they research it?

    • Mike HC says:

      “give the guy a break he’s just trying to write something optimistic for New Yorkers to read on they train ride to work. Sheesh.”

      give Joe a break he’s just trying to write something statistical savvy, and optimistic for RABers to read on they train ride to work. Sheesh.

    • Andy in Sunny Daytona says:

      This is going to be like buying a caravan of a group of Pikey’s.

    • “Go get a job at a newspaper, some of these writers may not have the time to go through Fan Graphs for every piece they write.”

      They don’t have the time to check fangraphs for every piece they write? THAT’S THEIR JOB. When I wonder about baseball-related issues, I check fangraphs. It takes no more than a few minutes. If they want to write unsubstantiated, illogical articles, people will criticize those articles and point out why they’re wrong and how/why we can look at the particular issue in question in a more enlightening, intelligent manner. Such criticism leads to more intelligent discussion, which leads to more intelligent and educated fans. This is, in no way, a bad thing.

      If I ever had occassion to use your excuse for baseball journalists at my job, I would immediately be dismissed from my job. I can envision the reaction I’d get:

      Employee: My bad, I didn’t get this right because I didn’t do the necessary research in the readily available source materials.

      Employer: You didn’t have the time to research the relevant background information regarding this issue on which you worked? You’re fired, it’s your job to get things right and to do the necessary work to make sure you get things right. If you can’t do that, why should we pay you?

      • MOSS
        I don’t gotta take this s#$t.

        BLAKE
        You certainly don’t pal, because the good news is, you’re fired.

        (Moss stops.)

        BLAKE
        The bad news is, you’ve got, ALL YOU’VE got, one week to regain your jobs, starting with tonight. Starting with tonight’s sit…

        Oh, have I got your attention now? Good. Because we’re adding a little something to this month’s sales contest; you all know, first prize is a Cadillac El Dorado–anybody wanna see second prize? Second prize is a set of steak knives.

        Third prize is: You’re fired. Get the picture? You laughing now?

      • A.D. says:

        If they want to write unsubstantiated, illogical articles, people will criticize those articles and point out why they’re wrong and how/why we can look at the particular issue in question in a more enlightening, intelligent manner. Such criticism leads to more intelligent discussion, which leads to more intelligent and educated fans. This is, in no way, a bad thing.

        Exactly, people need feedback, then they can adjust their work to that feedback, just why there is performance reviews at work, grades at school, and posters responding to other posters on RAB.

    • Big Juan says:

      Yankees outfielder Curtis Granderson took live batting practice off a left-hander on Field 2 yesterday and early in the second round lined a pitch up the middle. That’s progress.

      When someone says something this useless, they deserve to be ridiculed.

    • Go get a job at a newspaper, some of these writers may not have the time to go through Fan Graphs for every piece they write. It’s spring training, give the guy a break he’s just trying to write something optimistic for New Yorkers to read on they train ride to work.

      MR. PINK
      Look, I ordered coffee. Now we’ve been here a long f#$%ing time, and she’s only filled my cup three times. When I order coffee, I want it filled six times.

      MR. BLONDE
      What if she’s too f#$%ing busy?

      MR. PINK
      The words “too f#$%ing busy” shouldn’t be in a waitress’s vocabulary.

      NICE GUY EDDIE
      Excuse me, Mr. Pink, but the last thing you need is another cup of
      coffee.

      (They all laugh.)

      MR. PINK
      These ladies aren’t starvin to death. They make minimum wage. When I worked for minimum wage, I wasn’t lucky enough to have a job that society deemed tipworthy.

      MR. WHITE
      So you don’t care that they’re counting on your tip to live?

      (Mr. Pink rubs two of his fingers together.)

      MR. PINK
      Do you know what this is? It’s the world’s smallest violin, playing just for the waitresses.

      /MQBB’d

  11. Jack says:

    I love how the only picture they have on that article is him bunting. Is that his new approach?

  12. andrew says:

    I know i posted a bit of this above, but dismissing Granderson’s lefty statistics because they only account for one season of stats is NOT on par with dismissing a one year sample from a group of one year samples. Many of you have compared it to Swisher’s 2008 campaign, but we are able to dismiss this data (or at least take it with a grain of salt) because Swisher’s 2008 campaign was at the time, only 25% of his career. We didn’t take the 600 AB sample of 2008 as the be all and end all because 75% of the data disagreed with that 2008 sample.

    Granderson, on the other hand, has 620 AB’s vs lefties in his career, and that is our sample. The difference though, is that those 620 at bats are the ENTIRE sample. This is not at all the same as the other SSS cases. Generally when we complain of small sample sizes, it is in the case of Swisher, when people choose to look at only a small portion of a greater sample to draw their conclusions. But with Granderson, we are looking at the whole sample.

    This is a rant, but I just wanted to point out the issues with comparing this “small sample,” with the small sample issues that people brought up with Javy and Swish.

    • Many of you have compared it to Swisher’s 2008 campaign, but we are able to dismiss this data (or at least take it with a grain of salt) because Swisher’s 2008 campaign was at the time, only 25% of his career.

      Only 619 of Curtis Granderson’s career 2,896 plate appearances have come against left handed pitchers. That’s 21.3% of his total career.

      http://rlv.zcache.com/im_just_.....lf_400.jpg (safe)

      Granderson, on the other hand, has 620 AB’s vs lefties in his career, and that is our sample. The difference though, is that those 620 at bats are the ENTIRE sample. This is not at all the same as the other SSS cases.

      Yeah… no. The order in which the small sample comes does not change the fact that it’s a small sample. To wit: ARod’s playoff struggles came in a small sample scattered throughout his long career. It was still a small sample that was not enough information to make declarative statements about his true talent level.

      • But A-Rod’s SSS (playoff performance) wasn’t related to a discernable skill. Hitting in the playoffs is not much different, if it’s different at all, than hitting in the regular season. Hitting against LHP and RHP, on the other hand, are two discernably different skills. We know that RHB hit LHP better than they hit RHP, as a general rule, and vice versa for LHB.

        A-Rod’s playoff numbers were a SSS because they were a very small portion of his overall body of work. Granderson’s platoon splits aren’t a SSS in the same way because they’re the complete story of his body of work at this point in his career. Yes, he’s young and he hasn’t played all that much yet, so he might improve on those numbers… But they’re not a SSS in the same way as A-Rod’s postseason numbers were.

        • Hitting against LHP and RHP, on the other hand, are two discernably different skills.

          I don’t know if I can agree with that. I’d say that hitting against LHP and RHP can produce different results, but that doesn’t mean per se that they are discernibly different skills. Good hitters who hit lefties and righties equally well aren’t good hitters because they possess both of the two independed skill sets, they’re simply good hitters and don’t differentiate between the handedness of the pitcher.

          • Ok… But would you agree that, in general, LHB hit RHP better than LHP and RHB hit LHP better than RHP?

            Call it something other than a “skill” if you prefer, but the point stands.

            If you took a random and totally generic baseball player and gave him 1000 PAs in the regular season against RHP, 1000 PAs in the regular season against LHP, and 2000 PAs in the postseason (split evenly between LHP and RHP)… Where would you most reasonably expect to see a discrepancy in the numbers: Between the PAs against LHP and RHP or between the total regular season PAs and the total postseason PAs? Between the PAs against LHP and RHP, right?

            • If you took a random and totally generic baseball player and gave him 1000 PAs in the regular season against RHP, 1000 PAs in the regular season against LHP, and 2000 PAs in the postseason (split evenly between LHP and RHP)… Where would you most reasonably expect to see a discrepancy in the numbers: Between the PAs against LHP and RHP or between the total regular season PAs and the total postseason PAs? Between the PAs against LHP and RHP, right?

              I agree with your conclusion drawn from this tangent that’s not really at all related to what we’re talking about.

              • It’s very much related to what we’re talking about, whether you want to snarkily dismiss it or not. Here’s why:

                You made the A-Rod comparison because you thought it would cut against putting weight into Granderson’s platoon splits, you can’t then turn around and dismiss the comparison as irrelevant just because I reacted to what you said and debunked it.

                I showed that A-Rod’s postseason sample size was irrelevant because hitting in the postseason isn’t much different than hitting in the regular season, thus making his postseason numbers an extremely small percentage of his overall body of work and giving us no reason to think that split would continue to hold or provide any predictive power for subsequent postseasons.

                I also showed that hitting against LHP and RHP in most cases produce different results for hitters, thus, unlike the regular season vs. postseason comparison, we can’t dismiss a sample of PAs against LHP just because it’s smaller than a sample against RHP or because it’s a relatively small sample of the overall number of PAs.

                I’m sorry, but if you agree with the distinction I drew between samples of postseason PAs and platoon splits, then that distinction is relevant and important in this conversation, whether you want it to be or not.

    • I think this is the reason I was taking issue with the SSS thing in my comments above. My very quick reaction is that I agree with this comment.

      • I think the only point most of us who cite the small sample size are saying is that while there’s data that says he’s not a good hitter against lefties, that data still isn’t large enough to say FOR CERTAIN that he’s always going to be bad against lefties.

        We’re not saying that just because Nick Swisher rebounded from his small sample size to hit well again that Curtis Granderson will hit lefties well in the future, just pointing out exactly how little information we really have. You can look at 685 PA against lefties in 4 years and say “This is enough data to extrapolate and make accurate predictions on”; we acknowledge your concerns and the fact that you may be proven right eventually but disagree on the certainty based on the size of the sample.

        • “I think the only point most of us who cite the small sample size are saying is that while there’s data that says he’s not a good hitter against lefties, that data still isn’t large enough to say FOR CERTAIN that he’s always going to be bad against lefties.”

          Ok, so that’s more like what I said above.

          I don’t think anyone would argue that we don’t have a relatively small sample size here… I mean, Granderson’s young, so we just don’t have as many PAs to look at as we’d like. But, at the same time, we have to be careful not to dismiss the sample size we do have, because it does tell us something. I don’t think anyone here (other than a few fringe loonies, I’m sure) thinks that Granderson’s poor platoon splits at this point in his career mean he’ll never hit lefties well or even adequately, that he can’t improve at all. I know I’ve mentioned a few times that just because I put some stock into his platoon splits doesn’t mean I don’t think that, in the long-term, he can’t improve on this particular skill. You’re kind of making this too much of a black-or-white issue. I’m totally fine with saying that we can’t take 685 PAs and say the guy will never hit LHP adequately but, at the same time, I think it’s totally fair to look at that sample and conclude that hitting LHP is currently a weak-spot in his game and probably will be for at least the near future.

          • Mike HC says:

            Mondesi is on point here. My thoughts exactly.

          • I don’t think anyone would argue that we don’t have a relatively small sample size here…

            Clearly, some people would argue that. X and andrew have been arguing it all thread long. Our responses have been largely aimed at them.

            • I’m calling BS on this one, I’m sorry.

              Andrew: “I know there’s a chance he’s improving, Joe, and will one day be better against lefties. I am not arguing with your logic. I am arguing with the idea that dismissing Granderson’s sample is the same as Swisher’s. Swisher’s sample was small because it only made up 20-25% of the data on him. With Granderson, we are looking at 100% of the data, and still dismissing them.
              Yes, there is a chance he improves, but dismissing 100% of the sample size because it is as big as Swisher’s 2008 is not correct.”

              Nobody was saying this isn’t, in the grand scheme of things, a small sample size. Like, compared to a player who has played for 5 more seasons than Granderson has, I’m pretty sure Andrew would conclude that Granderson’s numbers against LHP are a small sample.

              What Andrew was saying is pretty similar to what I’ve been saying, I think. Yes, Granderson might improve… The numbers we have aren’t enough to tell us he’ll always perform like this throughout his entire career. But, at the same time, we can’t dismiss his platoon splits, and they tell us plenty of important information.

              I think the dissenters’ opinions in this thread have been distorted and unfairly maligned, in most cases.

              • andrew says:

                My bad, i didn’t see that you had already referenced my post. Thanks for having my back.

                • Okay, I misspoke when I lumped you and X into the same boat. You’re clearly saying something different than he is, and I apologize.

                  I still quibble with your claim, though, that you want to call Granderson’s lefty split the “entire” sample. You’re not explicitly saying anywhere in this thread (that I can see) that his lefty split is either a large sample or a small one, but you repeatedly refer to it as the entire sample when contrasting it with Swisher’s 2008 (which is also an “entire” sample in a way I think is misleading.

                  JMHO. That’s what caused me to disagree with your characterization. You’ve repeatedly said in this thread that Swisher’s 2008 sample is different than Granderson’s vLHP sample because Swisher’s sample is “25%” of his total while Granderson’s sample is “100%” of his.

                  I think you’re being a bit duplicitous in that regard. The only point of bringing up the Swisher sample is to put an easy to understand comparison of exactly how large or small Granderson’s vLHP split sample size is; i.e. not that big. It’s only 25% of his total career, just like Swisher’s 2008 was only 25% of his total career.

                • I think the distinction has been argued/explained a few times… And very-well explained by Snakes on the mother effin plane. Swisher’s 2008 was 25% of his career and it came all in one season. Granderson’s vLHP split is 25% of his total career but 100% of his vLHP career, and came over the course of his entire career, not all in one season.

                  Those distinctions are very relevant.

            • andrew says:

              But that’s not what I’ve been arguing. What i’ve been arguing is that dismissing Granderson’s sample is not in the same class as dismissing Swisher’s sample.

              You have continued to point out that Granderson’s total sample is his full career, while I believe (as well as Mondesi), that hitting righties and lefties is not the same. Also, as large or as small as you believe the sample size to be, I think more importantly is the fact that I believe it is the entire sample.

              Sorry if I’m being repetitive, but it seems i’m struggling to get my point across. Mondesi has seemed to convey what I mean better than I have.

            • Snakes on the mother effin plane says:

              It’s not a question (for me at least) of whether we do or do not have a small sample size. It’s a question of what we can or cannot (or should or should not) infer from those samples.

              ANd as I posted above there’s a massive breakdown when comparing 600 +/- PAs confined to one specific season to 600 +/- PAs that occur when spread relatively evenly over the course of 4 – 5 years.

              There are MANY reasons a player might have a single off year – an injury, personal issue, contract issue, trade talk issue, new team/city, etc. etc. etc. So dismissing (or downplaying) that one off year because “it’s not like the others” has more validity than in the case of a similarly sized (both absolute and relative) sample spread over many years, which cannot be better explained away than via the simple reason that “he doesn’t hit lefties very well.”

              You can quibble (and I do believe it’s a quibble) that b/c we’re not at 1000 PAs we can’t say anything definitive yet. So fine. We’re at +/- 1 std dev instead of 2. I personally believe that the N is more than high enough. But we’ll see. And I’ll certainly be happy if he turns it around even if that means he hits league avg against lefties…

                • ROBTEN says:

                  The “problem” is, I think, one of context. In other words, a SSS is a SSS, but it appears different because of when it occured.

                  For instance, Swisher’s or Javy’s SSS was in the context of additional information from which one might more easily conclude that the SSS was an outlier from the norm. That is, both Swisher and Javy had established that there were able to achieve a certain level of play prior the SSS in question, such that the SSS appears more clearly as a deviation from the norm.

                  In the case of Granderson, the issue is that the SSS is at the beginning of his career, so that there is no other predicative information upon which to determine, more or less, whether or not it represents a deviation.

                  Further, there is the fact that even with Granderson’s sample size against lefties there is an additional SSS when in 2008 he went .259/.310/.428. Not good, but not as bad has his career .210/.270/.344. So, thus far in his career, even within the small sample, there is variation.

                  Of course, this precludes the question of whether or not 685 PA is a SSS, but it seems that if we agree that it is, then the difference is that with Swisher and Javy, the context of the SSS makes it more evident that it was a SSS and additional information could be used (more or less reliably) to predict a correction whereas with Granderson the context of the SSS means that we don’t have enough information yet to draw a conclusion about whether it is an outlier or not.

    • Mike HC says:

      I agree that the comparison to Swisher and Javy are off. But it is still a small sample size for the same reason that a players first 600 at bats or so against righties in their career is a small sample. Nobody would look at Carlos Beltran’s (just as an example) first 600 at bats against righties and assume he will not improve. Look at the majority of players first 600 at bats against righties compared to the rest of their career and most will have greatly improved. There is clearly a difference between the at bats coming in only a season and a half or so, vs four + seasons, but it is a better comparison than taking one year of a players career, like Javy or Swisher.

  13. I’m curious… Everyone dismissing Granderson’s platoon splits due to the SSS… Do you not worry that Granderson has trouble hitting LHP? Are you taking the approach that ‘we have no reason to think this guy doesn’t hit LHP well, and we’ll see how he does in 2010?’

    • I think the problem lies between the ellipses. No one is dismissing the sample. The argument is that his skill is likely superior to the performance to this point. As I said and then repeated in the article, I don’t think he’ll ever be an elite hitter against lefties. I do think that he’ll outperform his career numbers over the next few years.

      • That. I have concerns; my concerns are allayed by the relative inexperience and youth of Granderson. His inexperience will become experience; his youth will become sagacity; those two things mean he’s far from a finished product.

        I have concern but there is valid reason for hope because his poor results against lefties thus far are such a small sample that they can’t be considered a final statement on his ability.

      • “No one is dismissing the sample.”

        That’s definitely a matter of interpretation, I think a few comments in this thread have at the very least implied (if not said pretty explicitly) that this is too small a sample to be relied upon. I’m not going to re-print all those comments here and get into a ‘he-said/she-said’ about it, but I disagree with saying that nobody dismissed Granderson’s platoon splits as a small sample size.

    • A.D. says:

      No one should dismiss the sample, they should realize it’s not very big, and therefore doesn’t have the same strength that it’s likely given. That said per SG, he’s probably never going to be great at hitting lefties, but there’s a good chance he can be better than he has been.

  14. Rose says:

    Honest opinion…

    Who exceeds expectations further?

    Curtis Granderson or Nick Johnson

    Note: Not who’s better in 2010…but who does vastly better than what we expect of them with their respective talents.

    • A.D. says:

      Hmmm, tough one, I think Granderson mainly because one expects him to be poor against lefties, and I think its more likely he improves against lefties than NJ suddenly hits for a ton of power or gets on at a 500 clip.

    • Sweet Dick Willie says:

      FYI – mentioning the subject of the post in a comment totally unrelated to the post does not make it on topic.

      • Rose says:

        I appreciate your concerns, Sweet Dick Willie. While you weren’t able to contribute anything relevant yourself throughout the entire post, I’m glad you at least got to play the role of ‘police man’ anyway. Thanks again.

  15. Rick in Boston says:

    Given that we’re dealing with some SSS issues here, I wonder if Granderson’s true lefty-hitting skill has been masked by his home field.

    My bigger issue with Granderson and lefties in his career: his walk rate is almost 4% less (10.5% vs. righties compared to 6.6% vs. lefties).

Leave a Reply

You may use <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong> in your comment.

If this is your first time commenting on River Ave. Blues, please review the RAB Commenter Guidelines. Login for commenting features. Register for RAB.