FanGraphs Baseball


RSS feed for comments on this post.

  1. So, for Adam Dunn this past year he should always have been bunting to the left?

    Comment by jacob — November 18, 2011 @ 11:51 am

  2. Dear Josh Hamilton…

    Comment by Adam D — November 18, 2011 @ 11:53 am

  3. One thing I love about Brian McCann is that he isn’t afraid to drop down a bunt when the opposing manager employs the shift. It’s shocking that more players don’t do the same thing.

    Comment by JT Grace — November 18, 2011 @ 11:54 am

  4. Does anyone have a good idea of how often a hitter can, on purpose, hit a grounder to a spot on the infield? A lot of MLers can’t even get a bunt attempt in play that often. And pitchers often attempt to pitch to the shift as well, although maybe this is overstated by announcers.

    55% seems very high for a hitter like Texiera to break even, but I guess that’s a sign that the article is adding useful analysis – it surprised me.

    Comment by test — November 18, 2011 @ 12:02 pm

  5. It is shocking. How do you not bunt against a shift like this?

    I think part of the problem is that big left sluggers (the guys who face the most extreme shifts) never practice bunting and there’s a good chance they’d embarrass themselves by popping up or striking out.

    Comment by Yirmiyahu — November 18, 2011 @ 12:09 pm

  6. Great piece, thanks for writing it. I Kinda hope the Sox don’t resign Ortiz cause I am so sick of watching him line rockets at the short rightfielder time and again. Especially when their is absolutely no one on the leftfield side of the infield. Cincho don’t set up and Papi don’t bunt. I think he’s afraid of tearing a seam in his pants sprinting torwards first. I can remember one of the Bill James abstracts talking about the value of bunting at a lousy third basemen who can’t field bunts and have a propensity for throwing the ball into the first base dugout. Seems it would be even a surer bet bunting at an open pasture.

    Comment by sheath1976 — November 18, 2011 @ 12:12 pm

  7. Maybe I missed something, but you used their averaged season performance to get the rates of singles, doubles, walks, etc. Yet those rates are lower hitting into a shift. I would think you would need to find out what any given player hits with a shift on rather than what they hit overall to figure out their needed success rate to hit against the shift. If I’m misunderstanding what you did here please let me know, because I think this is a really interesting topic.

    Another thing I was wondering about this is whether pitchers pitch the same way with a shift on. If we imagine a dead pull hitter under normal conditions, pitchers are generally going to tend to work the outside part of the plate against this hitter to prevent him from pulling the ball. Yet with a shift on, working the outside corner increases the likelihood of the batter hitting the ball well to the opposite field. So I guess what I wonder is whether as a result of the shift, pitchers might actually throw into the batter’s strength when they otherwise wouldn’t because they’re afraid of allowing him to hit to the open hole easily on the opposite field side. If I’m a batter with a shift on against me and the pitcher is throwing me inside pitches, there is no way I’m going to try to hit against the shift, but if he’s working me on the outside half I’m more than happy to hit the ball to the opposite field and take the single.

    Comment by Corey — November 18, 2011 @ 12:13 pm

  8. The shift has hurt specific players of this generation while others have overcome it. Ryan Howard fell into the shift trap, seeing his K’s go up, average down, and OBP down. David Ortiz was feeling the heat from the shift at one time, but has adjusted his approach to combat the shift and still produce great numbers last season. Ortiz’s percentage of ground balls and line drives were up last year and his fly balls were down. Howard’s line drives and fly balls were down last season but his ground balls were up. These stats show that Ortiz is either doing something to combat the shift or is ignoring it but hitting the ball harder, thus making it more difficult for fielders, even in a shift to get to the ball. Howard’s numbers show that he has yet to find a way to beat the shift as he consistently hits grounders (that go to the right side into the shift). I agree that these players need to work on hitting the ball the other way. It is pathetic to see players like Howard and Teixiera hit into the shift game in and game out while the other side of the infield remains vacant. Players don’t have to bunt to the left side, they just need to put in the effort to learn how to hit the ball that way. Lefties need to learn how to take breaking balls away and hit them, sometimes with less power, to left field. They don’t need to become spray hitters. Hitting the ball to the left side once in a while should be enough to stop defenses from shifting.

    Comment by Ben — November 18, 2011 @ 12:17 pm

  9. Aren’t the percentages you’re starting out with (Mark Teixeira’s 15.7%, 3.2%, 0.5%, 6.9%, 11.6%, 62.0%) based on all plate appearances, rather than just plate appearances where he faced the shift? If the shift works (I’m sure it does to some degree), then that first number should be lower, and the overall win expectancy should be lower.

    Comment by Yirmiyahu — November 18, 2011 @ 12:22 pm

  10. I agree with most of this. However, I would think a pitcher would still tend to pound the outside part of the plate against a dead pull hitter even with the shift on. The idea being that they do that anyway against these hitters and that they still pull the ball an ungodly amount of the time so why not keep doing it and stay out of their power zone down and in (for most of these players).

    My guess would be that pitchers would probably just avoid throwing change ups (or curve/slider for LH) on the outside part of the plate more often and just stick with outside FB.

    Comment by Colin — November 18, 2011 @ 12:24 pm

  11. The other thing is that, for the other half of the equation, the run expectancy is frequently much more than one base. Do you know how often, when a 3Bman has to run in on a bunted ball and make a rushed throw to 1B, he throws the ball over the 1Bman’s head?

    Comment by Yirmiyahu — November 18, 2011 @ 12:24 pm

  12. I think while it seems high, if the 3b is over in the SS spot, assuming Teixeira can get a bunt somewhat close to the line he’s not going to have a play. My guess would be that a pitcher is going to have to make the play most of the time in these situations and that success will largely be determined by Teixeira’s ability to place the ball close to the line and less on the defense he is facing. I actually wouldn’t at all be surprised to see >55% against a shift if he is any competent with bunting.

    Comment by Colin — November 18, 2011 @ 12:27 pm

  13. And with that, let their be stats on throwing outside CH or CU/SL….

    Comment by Colin — November 18, 2011 @ 12:28 pm

  14. Your point is definitely valid. The method I presented includes when the opposition shifted, but isn’t limited to just those situations. I don’t have the associated data for when teams shifted and when they didn’t, so I would have had to just assume teams always shifted with the bases empty. I figured this was a safer method and presents more of a worst (or best, depending on how you look at it) case scenario. If the defensive shift does suppress offensive performance (which it more than likely does), then the break-even batting average (or OBP if you want to say a hitter would still take walks) would be lower.

    Comment by Josh Goldman — November 18, 2011 @ 12:31 pm

  15. Yup, I address this lower down in the comments.

    Comment by Josh Goldman — November 18, 2011 @ 12:36 pm

  16. At least with Jim Thome, he would always practice bunting during batting practice. Granted, his stance was horrible and it was clear he was half-assing it, but he still laid down his bunts like everyone else before he started hacking.

    I think it goes back to Josh’s suggestion in the article: Sluggers are paid to hit the ball. These guys would rather go for the high-risk, high-reward extra-base hit instead of the low(er)-risk, low-reward bunt single.

    Comment by Bryz — November 18, 2011 @ 12:40 pm

  17. This seems like an article that should be revisited when Field F/X becomes available.

    Comment by Bryz — November 18, 2011 @ 12:41 pm

  18. Also, looking at the full plate of data probably gives us a better representation of true talent level and sample size becomes less of an issue.

    Comment by Josh Goldman — November 18, 2011 @ 12:45 pm

  19. I think “bases empty” would be a better baseline than all plate appearances. You could also expand your time period to multiple years to alleviate sample size concerns.

    Doesn’t MLB’s gameday system include a notation when the shift is on? Or maybe I’m just thinking of the notations for “runner going” and “bunting”.

    Comment by Yirmiyahu — November 18, 2011 @ 1:18 pm

  20. If Dunn would have learned to bunt, he might have hit .200!

    Comment by MikeS — November 18, 2011 @ 1:22 pm

  21. Those stats also show that Howard is declining and Ortiz is aging well. No wonder Ortiz is angry he can’t get a long term deal, compared to what Howard got.

    Comment by DD — November 18, 2011 @ 1:24 pm

  22. What does Howards K rate have to do with hitting into the shift.

    Comment by sheath1976 — November 18, 2011 @ 1:31 pm

  23. It seems like this should create a fairly powerful incentive for all but the greatest left handed power hitters to get better at bunting and then advertise that fact against the shift. If they can improve their bunting skills to anywhere close to the necessary level defenses will be forced to abandon the shift and by virtue of deciding last the hitter will benefit in all cases.

    Comment by bisonaudit — November 18, 2011 @ 1:33 pm

  24. I was thinking the same thing. Develop the bunt and prove you can use it a few times and the shift is gone for awhile–you can go back to having a more open right side.

    Once you have done this, the opposing team manager risks sever embarrassment using a shift against you. The most likely place to employ it would be late in game 1 of a 4 game series when one team is way out in front. In this case even an emergency pinch-hit for you tells them everything they need to know.

    Comment by Barkey Walker — November 18, 2011 @ 2:19 pm

  25. “Ryan Howard struck out looking to second base”

    Comment by Yirmiyahu — November 18, 2011 @ 2:24 pm

  26. It is sad, but I would have been ecstatic for .200

    Comment by jacob — November 18, 2011 @ 2:47 pm

  27. I know the Yankees never seem to shift against Ortiz. Is their any data out their telling us whom shifts against whom or which managers use the shift most often?

    Comment by sheath1976 — November 18, 2011 @ 2:58 pm

  28. Someone should really take a look at the Rays, they seem to be the team that employs the most aggressive shifts and a wider variety than other teams. Employing shifts is one way teams can attack the BABIP of hitters without changing their players.

    Comment by Phantom Stranger — November 18, 2011 @ 3:22 pm

  29. Thanks for the response. I don’t have a problem with assuming that the bases are empty, at some point you have to narrow things down just to make it comprehensible, I do think that the unstated assumption that the shift doesn’t work is a bit problematic, I like what you did here though, I would just pull the baseline success rates from in-shift situations rather than total situations. I have no idea if it’s even possible to get that data.

    Comment by Corey — November 18, 2011 @ 3:37 pm

  30. Josh, your box “Mark Teixeira, 2011 v R” actually contains his stats as a RH hitter vs LHP, when he does not face the shift. He had about twice as many PAs versus RHP as a LH hitter. Those are the times he faced the shift and he was much, much worse.

    I didn’t read all of the comments, so maybe somebody mentioned this already, if so, pls ignore.

    Comment by BronxBaumer — November 18, 2011 @ 4:03 pm

  31. You’re 100% right. I updated the numbers above. Thanks for catching that.

    Comment by Josh Goldman — November 18, 2011 @ 5:25 pm

  32. The Brewers supposedly committed to using more defensive shifts in 2011. I think Ron Roenicke said something to the effect of, “We’ve got the numbers (of where hitters hit the ball), we might as well use them.”

    Comment by Bryz — November 18, 2011 @ 6:00 pm

  33. What about failure attempts? Attempting a bunt is a per pitch decision, not a per PA decision. A failed bunt results in a strike even on a ball that would otherwise have been a ball. What effect does this have on the analysis?

    What if we changed the analysis to 0-0 count run expectancy charts? Then use league average bunt results to determine the expected results(BIP/foul/strike/ball). Calculate the change in RE given a certain BIP%. Now, take a look at batter performance for 0-0 bases empty. Do the same thing. I would presume more 1-0 counts for the latter and more 0-1 counts for the former. That strongly affects RE for the rest of the plate appearance.

    Comment by Nivra — November 18, 2011 @ 7:46 pm

  34. Also, with Teix batting LH he’s facing a RHP who generally fall off to the 1B side and whose glove is also on the 1B side. In other words, it may be difficult for them to field a bunt that’s not right at the mound.

    FWIW, Mike Fast recently presented average velocity off the bat and Ryan Howard led all hitters with ~78.6 mph. The key for Howard is to not pull outside pitches and go the other way like he did during his big years.

    Depending on their buying prowess, teams still might put on the shift with the mindset of “Hey if their big bat wants to bunt, we’ll take it”. The worst case scenario is a below average runner on 1st half the time (or less).

    Anyone look at charts to see how many hits the shift robs guys like Howard? It might not be as drastic as we think, and the easy singles via bunts might not be as high as we estimate. There might just be a whole lot of easy outs.

    Comment by CircleChange11 — November 18, 2011 @ 8:23 pm

  35. You wrote “whole” when you meant “hole.”

    Comment by J — November 19, 2011 @ 2:58 am

  36. I’ll just note that the former advantage that LHH’s used to have over RHH’s (considering BA only) has pretty much evaporated over the last 10 years:

    1970:LHH +4 points
    1980: LHH +17 (wonder how much of this is George Brett by himself?)
    1990: LHH +12
    2000: LHH +5
    2010: LHH +1

    I suspect that the now common practice of shifting has been the major factor involved here, since you can’t really shift like that against a RHH pull hitter.

    Comment by John DiFool — November 19, 2011 @ 2:09 pm

  37. Also the diminution of speed in the game. A lot of bad left-handed slap hitters used to rack up .283 batting averages every year. A lot more bad right-handed pitchers also used to get longer looks.

    Comment by Matt Trueblood — November 19, 2011 @ 11:29 pm

  38. Carlos Pena bunted for eight hits in 2011 in 10 tries. He could have doubled that and still not have been doing it enough.

    Comment by Matt Trueblood — November 19, 2011 @ 11:30 pm

  39. I think the pitching team would typically prefer to give up more singles in these situations rather than more extra base hits. Considering that we’re usually talking about high-power left-handed sluggers, these are batters that the pitching team would sometimes consider intentionally walking. If they know the batter is going to bunt against the shift, then they should apply the shift whenever they’d consider walking the batter. Now the batter has to try to bunt in order to get the base the pitcher almost gave them anyway, and there’s a good chance he’ll get himself out in the attempt. The pitching team would rather convert a power threat into a slap hitter, even

    Obviously this is situational and wouldn’t always apply.

    Comment by Newcomer — November 20, 2011 @ 12:56 am

  40. It’s not like Barry Bonds still wouldn’t have outhit the shift according to your charts if he bunted every time. .788 would not be a high outcome in the situation of the shift. His average would be closer to 1.000, if not 1.000. All you have to do is hit towards third base. Not even finely. Just hard enough to get past the pitcher.

    Though I’m also not sure about it for the San Francisco Giants teams he was on. The offense behind him was pretty wretched. Though being on base does turn the batting order over, and since it would only be when no one was on, he would still get a chance to drive others in situations when his doubles and homers would be worth more.

    Still, in most game situations, bunting was probably a better option.

    Comment by Crumpled Stiltskin — November 20, 2011 @ 8:14 am

  41. Don’t forget how it can affect a pitcher to go from pitching out of the windup to pitching out of the stretch, there are a lottttt of pitchers out there whose BAA are much higher withg runners on than bases empty.

    Comment by Melkman — November 20, 2011 @ 3:41 pm

  42. In general, hitters fare much better with runners on. It’s a combination of things: pitching out of the stretch, selection bias (if there are runners on, it may be because that pitcher is struggling at the moment), defensive positioning (even if there is no ‘shift’, defenders will be holding the runners on), the fact that the batter no longer has 100% of the attention of the pitcher and defenders.

    Comment by Yirmiyahu — November 20, 2011 @ 11:52 pm

  43. We’ll see how well he’s aging now that they’re testing for HGH.

    Comment by Sean — November 21, 2011 @ 12:01 am

  44. Another thing to consider is the 3B playing closer to SS is going to charge when the LHB squares around. He’s not going to just stand there as the ball rolls toward the 3B position.

    The batter also has the bunt it far enough away from the catcher so he doesn’t jump up and throw the runner out.

    So it has to be away from the catcher, away from the pitcher, bunt not hard enough that the 3B has a play on it … and he can’t square around too early.

    So essentially the batter has to bunt it down the line and beat the throw to 1B. That’s not an easy feat for the average batter let alone for one that would have limited practice.

    I’m not sure this is as easy as “bunt and get on half of the time” as some might think.

    Baseball players in general are kind of a “stick it to ya” type, and we’re not the only guys to think of bunting against the shift. I’d imagine quite a few guys have tried it in BP and found it to be much more difficult than it looks.

    I still think that defenses would welcome Thome, Howard, Dunn, Hamilton and Company laying down a bunt. Even with the shift I think the defense likes their chances.

    Comment by CircleChange11 — November 21, 2011 @ 12:13 am

  45. You’d think so, but Ortiz has definitely laid down some successful bunts. In fact, just last year:

    If they put their mind to it, the big guys can generally bunt all right. I think it’s more an issue that the bunt needs to be a bit of a surprise though, so they can only do it so often.

    Comment by B N — November 21, 2011 @ 4:16 am

  46. Further info, courtesy of the Tango:

    As of 2008, Ortiz had laid down 7 bunts against the shift over a couple of years and reached on 4. Not exactly laying it on thick, but enough to keep the defense honest. … Though Hafner apparently was having none of that BS. He was there to HIT (into the shift).

    Comment by B N — November 21, 2011 @ 4:19 am

Leave a comment

Line and paragraph breaks automatic, e-mail address never displayed, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Close this window.

0.376 Powered by WordPress