Wednesday, September 2, 2009

Should Hitters be Platooned More Often?

Ryan Howard is a good hitter. A very, very good hitter. He has a .921 2009 OPS and a .961 career OPS (142 OPS+). Howard's splits look like this:

vs. RHP, career: .307/.409/.561 (1.071)
vs. LHP, career: .224/.308/.446 (.754)

Against righties, Howard is every bit the monster he's reputed to be. Against lefties, he's a well below-average first baseman. But wait, there's more:

vs. LHP, 2008-09: .212/.290/.406 (.696)
vs. LHP, 2009: .198/.284/.348 (.632)

Every club's AAA squad has a right-handed hitter who could probably play a passable first base and put up a .700 OPS against left-handed pitching.

Of course, you can't platoon Howard. You should, but you can't. His overall numbers and his resultant reputation are just too good. He's not going to lead the NL in homers and RBIs every year by platooning. Also, in his (almost deserved) MVP year of 2006, he did put up a .923 OPS against lefties (which is pretty much the whole difference between MVP-quality Howard and the last couple years' pretty-decent-first-baseman Howard).

But consider another case. Tonight, the Twins were facing White Sox southpaw John Danks. They started Jason Kubel at DH and in his customary #5 slot in the order, and they started Delmon Young in left field and in the #8 slot. Forget for a moment that it's crazy to play either of these guys in the field, and just consider this (vs.LHP/vs.RHP):

Kubel's splits, career: .667/.844, 2009: .622/1.010
Young's splits, career: .805/.697, 2009: .861/.578.

Kubel, looking at his total line, has always been a good hitter, and has been one of the 15 or so best hitters in the league in 2009, with a 140 OPS+ and .387 wOBA. Young, on the other hand, has been as disastrous as ever, with a 78 OPS+ and .288 wOBA. Yet: Kubel is just as helpless against lefties now as he's always been, or even more so--the only difference is that he's crushing righties rather than just holding his own against them. No matter how lovely his overall numbers are (and add a .300 average, 22 homers and 77 RBI to that OPS), Kubel has no business ever serving as the designated hitter against a left-handed pitcher. Ever. Delmon Young is Kubel's perfect platoon partner, and DH vs. LHP may be the only role for which Young is actually suited.

There's another great reason to platoon, too. Say you're playing a team with a southpaw starter but a shortage of lefties in the 'pen, or a right-handed closer that you know they're going to use in the ninth. How awesome is it to have the luxury of using Ryan Howard or Jason Kubel (or your righty thumper if the situation is reversed) at exactly the right time, rather than just hoping his turn in the order comes up when you need it to?

So here's my idea, for some future really, really ballsy manager and/or GM:

We need to stop thinking of "hitting" as a skill. Rather, there's hitting vs. LHP and there's hitting vs. RHP, and they're totally separate skills, and your ability to do one doesn't make it a whole lot more or less likely that you can do the other.

So Ryan Howard has been awesome, and has put up some awesome stats, but he hasn't somehow earned the right to keep sucking against LHP by virtue of being awesome against RHP, any more than Tim Lincecum has earned the right to start in center field by virtue of being an awesome pitcher.

This kind of thinking would lead to a lot more platoons in more extreme situations (and there are a lot of them), and teams would properly value right-handed hitters whose numbers look bad because they hit against RHP 70% of the time, but who are highly valuable as the less-used half of a lefty/righty platoon. But even more commonly, it would change the way managers set batting orders and rest players. Almost every player (as we all know, but which I don't think people pay enough attention to) has a significant split in favor of opposite-handed pitchers. For instance, Mauer is a Hall of Fame .951 career vs. LHP and a merely pretty-good-for-a-catcher .762 vs. LHP, and while he deserves to start most games against both, he's probably not a #3 hitter against lefties, and he should never get a day of rest when a righty is on the mound unless the team has faced six righties in a row. Even Roberto Alomar, a switch-hitter and future Hall of Famer, probably batted high in the order far too often against LHP, against whom he had an OBP 50 points lower than he had against RHP (.337/.386).

It's not a big deal on a case-by-case basis (except in extreme cases like Howard and Kubel), but a manager who really looked at these things, roster spot by roster spot, and utilized significant platoon advantages whenever possible -- in setting the lineup and order and actually using platoons where appropriate, not merely pinch-hitting at the end of the game -- might pick up an extra win or two over the course of the season. And DHing Kubel against lefties just has to stop.


  1. It's an interesting idea, but would it mess the hitter's timing/mechanics/mojo? Most don't like platoons because it makes playing time so unpredictable and hard to get into a rhythm at the plate. Would more platoons just exacerbate the problem? Would it make the players worse against the pitchers they are supposed to hit against?

    Presumably, hitters do well against the other-handed pitcher 1) because that's how it works for several reasons and 2) they're in a sort of rhythm at the plate. Now, I don't know if it would really affect hitters at all. Maybe it doesn't really matter if they get consistent playing time. Something to consider.

  2. It's a good question, but I'm inclined to say that while there would be a lot of grumbling in that direction, it would make almost no difference to most or all hitters. How often do we see the media amazed at how a guy has come back from a long injury or suspension and just been on fire (like Mauer in May, for instance)? And when they slump badly, it usually turns out they were still injured. I just don't think things like "rhythm" factor into real-life baseball nearly as much as they factor into people talking about baseball. In The Book, for instance, they studied hitting streaks and found that being on a long hitting streak didn't increase the odds of getting a hit in your next game at all (suggesting that hitters don't really get "hot streaks"--at least not ones that have any predictive value--and suggesting to me that "rhythm" isn't a real thing in baseball).

    I'd really like to study this (considered doing it for tomorrow)--find some guys who had been both platoon players and full-time players in their primes and compare, for instance--but unfortunately, splits are just way too inconsistent from one year to the next due to sample size. Your boys' Matt Diaz, for instance, is a perfect example of the righty masher who can't hit righties that I was talking about, except he had one year ('08?) where he CRUSHED righties and didn't hit lefties. So the only way to really find out if that's a concern or not, I'm afraid, would be to give it a whirl.

  3. The biggest problem I see with platoons - because the thought has crossed my mind before - is the roster space. Even setting aside the costs (in dollars) of paying two people to perform one role, the limited roster spots makes things hard. With teams struggling to suit up the right number of bench players anyway (let's see: 13 pitchers, 8 starters, 1-2 backup catchers - that leaves you basically with 2-3 guys off the bench), it'd be hard to justify having more than one guy as a strictly platoon player.

    An intelligently designed roster might be able to maximize these precious few spots so that you're not wasting a player in any particular game, but it'd be hard. I like the idea of smarter platoons, but I'm not sure they can fit on a 25-man roster (especially when pitching staffs are now at 12-13 players)

  4. Very valid point, lar, but of course nobody actually needs 13 pitchers (or even 12). This hypothetical management team would theoretically also be smart enough to go with 11 pitchers and 14 position players.

    But of course roster space is still valuable, and for that reason, guys who can play more than one position (especially righties who can hit lefties and play more than one position) would be exceptionally valuable. So that if you've got a big lefty 1B who can't hit lefties, maybe you get a righty 1B/3B/LF, so that he can be a platoon partner but also the de facto backup at all the corner positions. Or if you're platooning in the outfield, get a backup who can cover all three OF positions. Etc. Catcher, of course, is prime for platoons (if you can find a good lefty-hitting one), since you usually need to carry two guys who can pretty much only catch anyway.

    It definitely does limit how much you can do it, though...but I think with the right guys you could have 3 or even 4 of them in the lineup.

  5. You wrote:

    "How awesome is it to have the luxury of using Ryan Howard or Jason Kubel (or your righty thumper if the situation is reversed) at exactly the right time, rather than just hoping his turn in the order comes up when you need it to?"

    I think pinch hitting is hard. I don't have any data handy, but if Howard gets his first AB in the 9th, he might not hit like he normally does. But maybe if players did this more often, they would do better. Also, you have to estimate what a guy's splits will be before you decide to platoon. A year or two of extreme splits might not hold in the future so you might not need to platoon as a guy reverts to normal or regresses to the mean

  6. Thanks for the comment, Cyril.

    No, you're right, there's a very real pinch-hitting penalty. The Book had a chapter on it. I was just kind of glazing over that part, since it wasn't really the point of the post.

    And of course you're also right that you need to be sure that you're not overreacting to a small sample, and they do fluctuate a lot from year to year. But looking at Kubel's career and how Howard has declined over the last three years, I'm comfortable in saying that neither can hit lefties, and it's at least equally important not to overreact if he ends up hitting .300 after 60 or 70 ABs against them one year.

  7. Lar made the point I was going to make; roster size. And of course, Mark made the point I was going to make about full-time and part-time guys and the level of comfort.

    So, all that's left to say is great job.

  8. Roster space is easy. CArry 11 pitchers because you don't really need more, and have a competent RH bat that can play multiple positions.


    You might not need to platoon the lefties every day, depending on matchups, so one can play, and one can sit.

  9. I recently came accross your blog and have been reading along. I thought I would leave my first comment. I dont know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.