« The Kitniyot Problem | Main | More fun with Politifactuality »

The Double Switch

Can someone explain to me about the double switch?

In baseball. The double-switch in baseball. When the manager changes pitchers and simultaneously changes a defensive position, so that the new pitcher comes in to a different spot in the batting order, and the new defender comes in to the batting order where the pitcher was. The idea of it, as far as I can tell, is that the manager wants to replace a pitcher is up next (or next inning) with a relief pitcher who he expects to want to keep in to pitch the following inning, and wants to avoid pinch-hitting for him. Given the way the game is currently played, it seems to me this should be rare.

  • It takes a starter out of the game. Sure, sometimes, you want this. In the last inning of a close game, or a game that has become close, you may want to get your aging shortstop out and your slick-fielding light-hitting young shortstop in. Of course, if that’s the switch, you are guaranteeing that the slick-fielding light-hitting young shortstop will be up next inning (in what was the pitcher’s spot), so there’s that. It kinda defeats the purpose. If you are going to be bunting with the guy anyway, why not just let the pitcher bat?
  • If there’s a rally the next half-inning, of course, the pitcher’s spot may come up anyway. If it does, there will probably be men on, and there’ll be another pinch-hitter gone as well as the pitcher you wanted to keep in. You could figure out some odds, based on the OBP of the batters, how many batters would be needed, and on the pitcher you think they will be facing, so you would have some sort of if the chances of the new pitcher’s spot coming up are more than X formulation, but it wouldn’t be very accurate. My feeling, though, is that the double-switch is pulled in anticipation (or at least hope) of some offensive production.
  • Most teams have umpty-’leven relief pitchers now, anyway, and no actual mop-up man. As far as I can tell, most of those guys are capable of pitching against a few batters two days in a row, or three days out of five, or whatever. Managers don’t seem all that concerned about saving the bullpen anymore. That’s not a bad thing (although The Book on how to use relief pitchers is seriously sub-optimal) but it does mean that it shouldn’t be that much of a problem to just bring in your guy, get your out, and then pinch hit for him in the next inning and bring in someone else.
  • Most of all, it burns a bench player for little reason. When the new pitcher’s spot comes around, you are going to pinch-hit for him, and you have one fewer choice on the bench to do it with, because you put the one guy in already. If you only have five guys on the bench, and one of them is your emergency catcher, it seems like that’s a resource you want to conserve.

Let’s take an example, shall we? You yank the pitcher and do a double-switch. In your half of the inning, it’ll be your 6-7-8 hitters, so you pull your cleanup hitter and put your pitcher in the 4 slot, and the back-up guy in your 9-slot. The next inning, you go through an extra batter—success!—so following your relief pitcher getting another three outs, you start with your 1-2-3 guys. But your 1-2-3 guys can hit (hypothetical), so the pitcher’s spot comes up with men on base, and you need to pinch-hit. Which you do. You now have one position player and two pitchers out of the game. At the end of the inning, though, you will need to put another pitcher in, so you now have two position players and two pitchers out of the game. You may have some choice about who is coming out, now, depending on what positions your pinch-hitter can play, but either way, you have gone through two position players and two pitchers by the end of the inning.

If, on the other hand, you had not pulled your cleanup hitter, but instead kept your relief pitcher in the 9-slot, you would have had to pinch-hit in the bottom of the inning, and then put your new relief pitcher in the 9-slot again. Now, in your next time at bat, when the 4-slot comes up and you have men on base, guess what? You have your clean-up hitter hitting cleanup. Amazing! At the end of that inning, you have burned through two pitchers and one position player. Your second relief pitcher has pitched a full inning, but is able to keep pitching if you want him to.

So if you have a mop-up guy that you want to have in for more than one inning, it seems like the double-switch is the wrong way to go about it. And in general, with the way rosters are set up and the way relief pitchers are used, I would think that it would be better to use up an extra relief pitcher than an extra position player. Right?

The reason I am asking here is that Bruce Bochy, manager of my San Francisco Giants, loves the double switch. He seems to be disappointed if he ever has to pull the pitcher without pulling another player as well. Yesterday, Tim Lincecum was getting shelled, and had to be pulled in the third inning, and he switched out the clean-up hitter. Eight batters later, he had to pinch-hit in the 4-slot. In the fifth, he switched back to put the pitcher in the 9-spot. In the seventh inning, that spot came up and he pinch-hit; at the end of that inning, he left the pinch-hitter in to put the pitcher in the 2-slot, which made no sense to me at all, as the 2-slot was coming up the next inning, and when it did, he pinch-hit, of course. He pinch-hit with the last guy on the bench, our starting catcher who had been given the day off because he had been diagnosed with shingles. We were down by eight runs, at that point, and it was the eighth inning. Had we somehow scored a bunch of runs in that inning, we could easily have been in the position of having to have a relief pitcher bat in the ninth with the tying run on base.

That was an absurd situation, in an absurd game. We didn’t lose that one because of the double-switch, and really, at the time of the first double-switch we had a win expectancy of 3%, so it seems silly to complain about managerial moves at that point. But Mr. Bochy uses the double switch a lot, and while I do get the sense that other managers don’t use it quite that much, it seems to me that it happens far more than it should.

Or, at least as likely, I am missing something. Am I missing something?

Tolerabimus quod tolerare debemus,


It seems to me that the following principles apply:

1) Never remove a pitcher who is pitching well just because you want to change a position player--I've never heard of a double switch that wasn't initiated because of a need to change the pitcher!

2) When you have a lead, late in the game, the marginal value of defense rises significantly versus the marginal value of offense, so if you have a defensive liability out there, the double switch is a convenient way to remove that liability without burning a pinch-hitter;

So, when you have the lead, late in the game, and you need to change pitchers, the double switch can be a moderately effective tool. Using it in any other situation, unless there's some other compelling reason why both a pitcher and a position player need to be changed at the same time, is either useless or counterproductive.

The Pirates have given up 12 runs in 5 games. They are 2-3 because they have scored 9 runs. It's early days. I expect the Giants' pitching will improve. The Pirates' offense? Not so much.

The principles you outline seem sensible, although I would add that with a lead in the ninth inning, the rearranging of the lineup should be unnecessary. Really, then, the double-switch is a good idea iff (a) you have a defensive replacement, and (2) it's the eighth inning, you have a lead of three runs or fewer, and you want to bring in your closer now and have him finish the game.

I suppose I could dredge the data to find out how often the latter actually happens. Not as often as it ought, I suspect.


Comments are closed for this entry. Usually if I close comments for an entry it's because that entry gets a disproportionate amount of spam. If you want to contact me about this entry, feel free to send me email.