“the probability of a double play is seemingly nowhere near 50%…there is a near 50% chance that both make outs in consecutive at-bats.”
What is the probability of Anderson grounding out to a drawn-in infielder, striking out, or popping up? Isn’t that and an out from Ausmus the run-saving alternative to the IBB & DP strategy? Both batters making outs doesn’t help prevent the run from scoring if the first out is a sacrifice fly.
So, what would have been the correct call? If Manuel was worried about the run and he didn’t walk Anderson, he would have had to have had the infield in. It would be interesting to know how much a player’s BABIP goes up when the infield is brought in.
If this is late in the game, it’s a good move (or at least about the only move you have vs. a leadoff triple).
But, it’s the 4th inning and it’s GAnderson.
I don’t think wlaking Anderson to try and get a double play is necessarily a stupid move. BUt, in the 4th inning it’s likely an unecessary risk.
This is nowhere near as dumb as Charlie Manuel allowing Joe Blanton to bat in the 7th with 2-out and the bases loaded with a 1-run lead. Blanton K’d, and then went out and gave up the lead to StL in the next 1/2 inning.
FWIW, the article’s tone reads of “ridicule” right form the start. If you’re going to take that approach, then the situation should be more “obviously dumb” than it is. Don’t be that guy.
Comment by CircleChange11 — July 26, 2010 @ 10:13 am
R.J., do you realize that OBP and wOBA do not determine whether he will hit an infield groundball out or sac fly and score the run anyways. As it is, even if Anderson strikes out, Ausmus still gets the single and they score the run anyways. Are you saying that there is a better probability for a strikeout and an out, then there is for a double play? What are the chances for a double play with bases loaded against the pitcher? Those are the numbers we should be seeing, not wOBA which bears no relevance in a situation where the goal is to plate the run, and not necessarily to reach base.
So when the Mets inevitably miss the playoffs this year is Minaya finally going to get the well-deserved AXE??? Everyone inside (and outside) baseball knew going into this season that the Mets had gaping holes in their rotation, OF, 2B and 1B. If not for completely out of nowhere good performances from R.A. freaking Dickey, Hisanori Takahashi and Jonathon Niese this rotation would be historically bad…
But I guess you can give Minaya credit for not trading Ike Davis and Mejia+prospects for Roy Oswalt and Lance Berkman…
Comment by DonCoburleone — July 26, 2010 @ 12:38 pm
“a gorgeous day in Los Angeles, California, with the sun hanging out overhead like a halo”
No offense, but please leave the literary flourishes to Carson.
You about to pitch to Garrett Anderson and Brad Ausmus, just how worried should he have even been? Anderson has been good for an out near 80% of the times he steps to the plate, Ausmus an out at a 75% clip. Let your pitcher pitch and the problem just might take care of itself.
“Oh, and the Dodgers won by one run. A 13th inning homer by James Loney off Oliver Perez with Francisco Rodriguez still available.”
I think it is pretty standard for the road team to not use its closer in an extra-inning game until they take a lead.
It’s not so much that a manager would fear Anderson, but rather looking at the odds of getting 3 consecutive outs, without a leadoff triple scoring.
One of the ways to increase your odds is to put yourself in a position where you can get 2 outs on a batted ball that would normally just result in 1 out, with the runner scoring from 3rd (for example a grounder up the middle to SS or 2B). A grounder right at the 3B or 1B likely causes the runner to hold while the single out is recorded.
But a grounder resulting in a DP, ends the inning, with no runs scored.
Also, a fly ball by Ausmus scores the run (in either situation).
So, the negative is minimally increased with the IBB, while the potential for the DP, and getting out of the inning without the run scored is increased.
I don’t Manuel would normally be so concerned with allowing the run to score, but given the weakness of the batters upcoming, following the K, presented a higher-than-normal chance at getting out of the inning without a run being scored (even with an out), which is a rare chance worth taking advantage of (perhaps).
Comment by CircleChange11 — July 26, 2010 @ 1:21 pm
Why rely on one person to produce two outs when Garrett Anderson does a pretty good job of getting out on his own? I think intentionally walking one of the worst hitters in MLB over the past two years, a guy who multiple Ps have outproduced at the plate over the last two years, is “obviously dumb” on it’s own.
11% of double-play situations turn into double plays. That means there’s an 89% chance that at least one of those two players will still be on base following Ausmus’s PA. Do you really think that there was less than an 11% chance of getting both Anderson and Ausmus out consecutively?
Just because a lot of teams do it doesn’t mean it’s smart. Girardi was rightly skewered earlier this season for having Chad Gaudin lose a game in the 14th up in Toronto with Mo sitting out in the bullpen.
Because Garrett Anderson’s “out” could still score that run.
Isn’t this extremely obvious?
I’m not saying it was the right call, just that it’s not THAT dumb. Sure, we shouldn’t celebrate the fact that Manual did something that was only maybe a little dumb, but I don’t see the need to write a whole article calling him out for it either.
True. But Girardi’s reasoning was that Mo wasn’t available for 2 innings in Toronto, so he was “saving him” for a lead. Still wrong, but at least he was thinking about it instead of just going by the book.
Because a few weeks later he did use Mo in that situation TWICE within the same week.
a fly ball by Ausmus, if you had gotten Anderson out while stranding the runner doesn’t score the run. Plain and simple, I’d take my chances at Anderson making an out, and then no matter what Ausmus does, if he’s out the inning is over and the run doesn’t score. So Ausmus really isn’t the issue as much as Anderson is and his absolute failure to do anything worthwhile at the plate in a long, long time.
Steve, his out could score the run sure. But then you could just as easily if not more so fail to induce Ausmus into a double play. But why rely on the oft chance that Ausmus gets out the exact way you want, when Anderson has proven so absolutely awful and that makes ANY Ausmus out effective instead of just a single type of out.
I don’t think anyone here feels strongly about it either way. Pitch to Anderson, don’t pitch to Anderson. One may be slightly better than the other, but neither option is mind-bogglingly stupid. That’s really all anyone is saying. Just that it doesn’t really seem that interesting of a decision to write an entire article on Fangraphs about. Managers do stuff like this all the time.
I simply fail to see how this is an obviously awful move worthy of having an article written about it. In these situations, most managers would consider the pitcher to be an easy out. So he’s really just looking to get the 2nd out of the inning. Option 1 is an awful left-handed hitter. Option 2 is an awful right-handed hitter. With a righty on the mound, and considering the option of a double play, I can’t see how the win expectancy (or the run expectancy) would be much different either way.
True enough. The situation revolves around not wanting Anderson to plate the run, while making an out by putting the ball in play.
Anderson is not normally a “high K guy” … but this year he’s K’ing at 22%, so perhaps it’s worth a shot to go at him.
I know each at bat is it’s own probability, but the chances of getting 3 consecutive outs that do not score a run from 3B is rare. I don’t have my copy of ‘The Book’ with me, or I’d look up the stats for game states.
Generally, the situation calls for “concede the run, but nothing more than that”. This situation is a bit different since once Blake K’d (which seemingly changed everything), the quality of upcoming hitters is weak, weak, and weak.
Ausmus over his career, is a ground ball machine at the plate. So perhaps that is taken into account. Anderson on 1st and Ausmus at the plate is a good recipe for an inning-ending DP.
Now, had I been Torre … once they walked Anderson, I safety squeeze with Ausmus.
Comment by CircleChange11 — July 26, 2010 @ 2:34 pm
Do you really think that there was less than an 11% chance of getting both Anderson and Ausmus out consecutively?
Probably worded my comments poorly.
What they are wanting to avoid is Anderson making an out in such a way that scores the guy from 3rd.
Meanwhile, Anderson is slow, and Ausmus is both slow and a groundball hitter.
Comment by CircleChange11 — July 26, 2010 @ 2:36 pm
~60% of Anderson’s batted balls go into the air (LD + FB).
~60% of Asumus’s batted balls go into the ground.
I’m assuming MLB teams have this type of stuff jotted down on notecards or something to use during the game.
I have to wonder, “When was the last time a runner on 3rd and no outs in the 4th inning attracted so much attention?”
Comment by CircleChange11 — July 26, 2010 @ 3:38 pm
Is looking at overall wOBA’s the right way to look at this? While Anderson does not have pronounce platoon splits, Ausmus’ career wOBA splits vs righties is 60-70 points lower (too lazy to look up average R-R platoon splits but even if Ausmus is assumed average, that still lowers the wOBA while Anderson’s would be higher against a righty). You can’t just wave your hand and say wOBA is the same while completely ignoring handiness as another potentially important variable.
Another option is to look at Pelfrey’s career platoon splits..which show a 56% groundball rate vs righties and a 44% ground ball rate vs lefties (regress that as much as needed). His infield fly rate is 4.9% vs lefties and 12.8% vs righties (this obviously is much smaller sample size so hard to say how big this delta actually is)
You think maybe the platoon splits should also be a factor you might want to consider in the probability analysis (both from the pitcher’s and hitter’s perspective)?
This reads a bit like a desire to bash Manuel and the Mets as there are much more obvious examples of unwarranted moves if you want to point out how managers ignore the odds. When you couple that with what seems to be a rather incomplete analysis of the actual situation, this seems like an unwarranted blog post.
I wonder how much discussion took place in the dugout about this situation? Or if once Blake K’d, someone said “What do ya wanna do here, Skip?”
Then some brief info changed hands, likely about the pitcher and groundball tendencies, and a quick decision was made.
A decision had to be made in the time it takes for Blake to get out of the box and Ausmus to get in.
High GB rates for both RHP and RHB put the D in as favorable a position as one could expect for not letting the runner on 3rd score.
Had Ausmus grounded into a 6-4-3 instead of a single up the middle, would anyone have noticed the “genius” of it? Or if Ausmus had popped up or struck out, putting the opposition in the spot of letting the P hit or taking him out of the game in the 4th?
I wonder what Torre thought? Surely he could not of been thrilled with a serious DP threat at the plate, and 1st and 3rd with 1 out.
Interesting that Blake gets off with no blame when just putting the ball in play, scores the run.
Comment by CircleChange11 — July 26, 2010 @ 6:06 pm
A lot of the reason that it’s rare to get three consecutive outs with a runner on 3B and not score the runner is that it’s rare to face three consecutive batters who completely suck. Which is why the Blake out was such a big deal — that left the Mets in exactly that enviable position. Clearly their chances of getting two consecutive outs from Anderson and Ausmus were considerably better than if they were facing Pujols and Holliday.
If I’m doing my math correctly — and lord knows whether or not I am — Anderson and Ausmus have a 59.625% chance of making consecutive outs (using their 2010 stats and assuming neutral pitching and defense). Going for the DP in this situation you’d need to be able to expect at least a 60% chance of success, and that’s entirely unreasonable.
But that “productive out” opportunity doesn’t go away if you walk Anderson. Ausmus can fly out to right and score the run, or bleed through the infield and score the run (which is what happened) just as easily as Anderson could. Actually, probably more easily, since, sad as it sounds, Ausmus has been the better hitter this year.
This is not a bad call IMO. The point is not that he could get two outs in a row, but that the same out that could get one out might instead get two outs, since the first batter’s out might very well plate the run.
Since pelfrey is RH and Anderson left-handed, and Ausmus being right-handed with the pitcher up behind him, the choice he made might even be the better choice.
Not to mention Anderson is an extreme fly-ball hitter. To go after Anderson and not give up a run you need to know what is the chance of every result except GB or Fly-ball, this would not be very high in Andersen’s case especially.
You could bring the IF in, but then you increase the chance that the GB is a hit, and Still a fly ball would plate the run. If you don’t bring the infield in, then a GB is very likely to score the runner, so why not change that to being the ground ball getting a possible double play instead of scoring a run?
By walking Anderson you have a SO, and pop out AND certain % of GB’s being a double play (or not being able to score the runner from 3B anyway). With Ausmus batting and him and Anderson running and Pelfrey (High GB pitcher) pitching the combination seems a much more preferable idea.
And even if Ausmus gets a hit, he is very likely only to get a single at worst and you have the pitcher up next, so you would give the single and have a man on first and second with the pitcher up, meaning that even giving up a single is very likely to only give up one run in the end result anyway.
To me, the over all risk due to the specific circumstances involved actually make this a choice that might even be positive, or at the very least not all that bad.
When I read the title, I was expecting to read how he walked GA to face Loney, Kemp or Manny Mota (circa 1975). When I learned that he was losing by 1 run and he walked a bad left-handed hitter to face a bad, slow right-handed hitter (with a RHP on the hill) with the P in the hole, I realized this was much ado about nothing. If Tony LaRussa had done this instead of Jerry Manuel, no one is writing an article about it.