Wakefield’s Curious Platoon Split

It looks like most of my fellow writers are more sensible than I, and have moved on from constant attention to the two-week-old Fangraphs’ splits. But I cannot help myself: as a pitchf/x-er I have a fondness for Tim Wakefield and I had always remembered hearing that the knuckleball does not have a platoon split. So it was something I had to check out.

Since 2002 Wake has posted an xFIP of 4.79 against LHBs and an xFIP of 4.73 against RHBs — not much of a split. His FIP actually shows a reverse split: 4.38 against LHBs and 4.73 against RHBs. The interesting aspect is that the components show rather large — but complementary — splits.

Looking just at strikeout and walks Wakefield does much better against righties, with 6.56 K/9, 2.78 BB/9 against RHBs and 5.52 K/9, 3.34 BB/9 against LHBs. That would portend an big split. But once the ball is in play the story changes. Righties have a BABIP of .286, with 38% GB and 10.5% HR/FB. Lefties a .264 BABIP, with 42% GB and 8% HR/FB. So lefties are hitting a lower BABIP; more ground balls, so fewer flies; and a lower percentage of those flies make it over the fence. In every way lefties make poorer contact against Wakefield.

I thought that maybe this had to do with Wakefield throwing his pitches in different percentages to RHBs and LHBs. But his numbers a very similar, against both about 85% knucklers,10% fastballs and 5% curves.

So the difference must be in how LHBs and RHBs deal with his knuckleball, and the per-pitch numbers bear this out. Righties whiff on his knuckler more often (20% versus 16%) and swing at more of them out of the zone (27% versus 25%), but when they hit it they make much better contact (.512 slugging on contact versus .435). I don’t know whether this is a difference in approach — righties go up there looking to knock one out of the park while lefties just look to make contact, but I don’t know why that would be — or whether it has to do with how lefties versus righties pick up as it is delivered or what is causing this difference. Anyway it is one more in a long list of wonderful mysteries of the knuckle ball.

Put it all together and Wakefield probably has a slight reverse split. His FIP, which would credit the difference in HR/FB as real while xFIP would ignore the difference, is better against lefties. And that does not credit him for his very low BABIP against LHBs, which after so many batters faced is probably a real difference and makes his performance agianst LHBs even better than his FIP would suggest.

But anyway you look at it the platoon split is pretty small. All else equal if you have two guys and one spot in the rotation and one spot in the pen you would give the rotation spot to the guy with the smaller platoon split — thus opposing managers could not take advantage of it in lineup creation and you could leverage the bigger split of the guy in the bullpen. So that is one argument for putting Wakefield in the rotation. But all else is not equal and, even accounting for Wake’s splitless-ness, there are probably five better starters than Wakefield on Boston’s depth chart. Anyway the point will probably be moot, as there is little chance all six pitchers are healthy at the same time and Wake will get his share of starts and we shall all rejoice.

Print This Post

Dave Allen's other baseball work can be found at Baseball Analysts.

19 Responses to “Wakefield’s Curious Platoon Split”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. JoeR43 says:

    Posting just to say I approve of any blog post written about Tim Wakefield.

    +13 Vote -1 Vote +1

  2. TsB says:

    Interesting article. I’ve always been fascinated by the knuckler.

    In regards to the difference between righties and lefties, I wonder if the fact that he’s played his career at Fenway could be responsible for some of those numbers? Specifically the difference in approach/mentality of visiting righties and lefties? I don’t know where to find data to try and back that up (I find it tricky trying to get to know more when getting in to sabermetrics), so it’s just a theory. If you have them, it would be interesting to see the home/away lefty/righty/ splits.

    Vote -1 Vote +1

  3. Dirty Water says:

    “righties go up there looking to knock one out of the park while lefties just look to make contact, but I don’t know why that would be”

    I suppose that would be because RF is far more daunting than the Monstah. It may not be a herculean task to pound a 50 mph floater over Fanway’s RF wall, but RH’s looking to pull the ball surely have the easier task.

    Vote -1 Vote +1

  4. I can’t remember who did it (and it probably was a while ago), but I seem to recall there being switch-hitters that would hit right-handed against Wakefield.

    (Yeah, this comment’s not going to be very useful unless someone comes up with the guy who did it. I’m sorry.)

    Vote -1 Vote +1

  5. walkoffblast says:

    I think a lot of switch hitters choose to bat righty against wakefield but I cannot remember the reason.

    Vote -1 Vote +1

  6. Brian says:

    how about the fact that most people are naturally right-handed?

    If most of the pitchers are righties, then the fact that far more than the normal societal % of lefty hitters find their way to the majors probably means that as a whole, the righties that DID make it are better PURE hitters. And if there were such a thing as a pitcher without a platoon split, then the righties should hit him better than the lefties because, again, they have to be better hitters to be in the big leagues.

    I have absolutely nothing to back this up. I just know that if Jacque Jones were a right-handed hitter and his splits were flipped, his career baseball earnings would be 6 figures rather than 8.

    Vote -1 Vote +1

  7. Evan Kirkwood says:

    Almost all switch hitters face Wake from the right side because it’s easier to pick up the knuckler from the right side as opposed to the left. I can’t remember a single switch hitter batting from the left side against Wake in about 5 years of watching Red Sox baseball.

    Vote -1 Vote +1

  8. JoeR43 says:

    Since it’s been addressed already, is there any way to check Wakefield’s L/R splits on the road? That would take a ton of digging but could shine some extra light on him.

    Even though if you’re an AL East fan and haven’t seen enough of Tim Wakefield to evaluate him yet, you pretty much can’t make a visual observation on anything, ever.

    Vote -1 Vote +1

  9. JoeR43 says:

    More Wakefield:

    His 1994 season in Buffalo (AAA):
    5-15, 5.84 ERA, 1.679 WHIP, 4.25 K/9, 5.02 BB/9.

    That just makes his 1995 season in Boston even more astounding, going from a guy who looked one bad outing away from selling Buicks to a Cy Young candidacy.

    Obviously he’s never approached that since, but it takes a lot of crying to not appreciate a guy who constantly starts 30 games w/ a better-than-average ERA. Fortunately even most Red Sox fans are immune to that level of annoyance.

    Vote -1 Vote +1

  10. Craig says:

    Tim Wakefield’s numbers are absolutely fascinating. There is almost no difference between batter’s numbers when the pitch is in the strike zone vs when it isn’t either.

    I think the knuckle ball is so random that it basically defies any kind of logical split.

    Vote -1 Vote +1

    • Craig says:

      I think Wake may in fact be the only guy with that many innings where one split shows batters hitting better on balls than strikes.

      Wakefield in 09
      LHB hit .287 and slugged .449 on strikes and .268/.429 on balls
      RHB hit .261/.399 on strikes and IMPROVE to .265/.429 on balls!

      Vote -1 Vote +1

      • David says:

        I wouldn’t be surprised if some power relievers with great stuff, great movement, but no control had similar splits. I could imagine a pitcher who is close to unhittable in the zone (so has low OBP and low SLG), but throws a ton of balls so gives up a lot of walks (high OBP and low SLG outside the zone).

        Vote -1 Vote +1

      • David says:

        unless, of course, those statistics only account for balls in play in which case my point is invalid.

        Vote -1 Vote +1

      • Craig says:

        Yeah, they are numbers in official at bats (which is why I used BA and SLG rather than OBP.)

        Vote -1 Vote +1

  11. bballer319 says:

    I’d have to think that if he hangs a few that don’t quite knuckle, it becomes softball out there for a few outings (why balls would be easier to hit than strikes maybe). Also, switch hitters bat righty vs Mariano Rivera because of the cutter.

    For switch hitters batting right-handed, Joe Maddon had the right idea:


    “Some switch hitters will bat right handed against Tim Wakefield because they are worried that facing his knuckleball delivery would disrupt their normal left handed timing against other right handed pitchers. ”


    and “Two of the switch-hitters in the Indians’ lineup, Victor Martinez and Asdrubal Cabrera, even tried a unique approach to solved Wakefield. They eschewed the platoon advantage and batted from the right side after the right-hander pitcher. “It was Victor’s idea, because the way a knuckleball breaks, it moves away from a left-handed hitter instead of into the body,” said Cabrera, the rookie second baseman. “You actually see the ball better by moving over to the other side.” The numbers suggest that their changing gears was a good move—right-handed batters had a 780 OPS against Wakefield in the regular season, while lefties had a 716 mark; during Wakefield’s 16-year career, the splits are 765/722, again slightly worse against hitters stepping in from the right side.”


    Vote -1 Vote +1

  12. Glen says:

    Jim Bouton (coincidentally, also converted into a knuckler and purportedly a “knuckle-head”, but that’s besides the point) writes in his book, Ball Four, that he treated all of his hitters as flesh and would just go after them regardless of how good or bad they were. Maybe this is approach by Wake is an explanation of why he throws hitters 85% knuckleballs, 10% fastballs, and 5% curves. Also, it might be an explanation as to why with a similar approach to hitters the splits are similar. Just a thought, even though I’m not selling it as anything definite (more information needed).

    Vote -1 Vote +1

  13. Mike says:

    Hey Dave – here’s a random question for you. Wakefield’s PitchFX tab of this site is showing that his knuckleball has very consistently been a “rising” pitch, much like a fastball with backspin. As I understand it, this metric is relative to a pitch with no spin.

    However… Wakefield’s knuckleball has no spin! I mean, it has a tiny amount of spin sometimes, but for the most part, I would guess it’s backspin is 1/50th of that of a normal backspun fastball.

    Shouldn’t his vertical pitch fx be just about 0, just like his horizontal? Does this demonstrate an inaccuracy in the pitch fx system if it’s showing a knuckleball as having “rise”?

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>