Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Wednesday, August 26, 2009

Kabak: The bunt that drove us all crazy

Golly. So Singleton and Cone were wrong…again.

In fact, it has been proven that at no point in the game does giving up an out in exchange for a base lead to a better chance at scoring runs — and runs, after all, represent the ultimate goal of a baseball game. Before we arrive at Nick Swisher, Joe Girardi and the bunt that made me want to punch a wall, take a look at Baseball Prospectus’ run matrix. This chart details how many runs a team at bat scores in any given situation. For example, with runners on 1st and 2nd with no one out, a team is expected to score 1.50766. With one out and runners on 2nd and 3rd, a team is expected to score 1.43489 runs.

There are two key points here. First, with that extra out and extra base, a team’s total runs scored decreases by around 0.07 runs. Is that by itself worth eschewing the bunt? Probably not. After all, a team with runners on 2nd and 3rd and 1 out still scores, on average, more than 1 run in that situation. There is, however, a need to consider who is batting.

And now, we return to the Yankees. According to Jack Curry, Joe Girardi, with two on and no one out, asked Nick Swisher to bunt. Now, Nick Swisher is a power hitter. On the season, he has a 123 OPS+ and has hit 123 career home runs. He also gets on base 37 percent of the time — and, by the way, a team with bases loaded and no one out scores, on average, over two runs in that situation.

Repoz Posted: August 26, 2009 at 03:01 PM | 62 comment(s) Login to Bookmark
  Tags: game recaps, rangers, sabermetrics, yankees

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. JoeHova Posted: August 26, 2009 at 03:13 PM (#3305562)
I didn't like seeing the bunt there. How many guys in a row had reached base? 6, right? Francisco had to welcome seeing them give him the first out, even before the bunt failed.
   2. aleskel Posted: August 26, 2009 at 03:17 PM (#3305568)
I have a feeling Girardi was playing the "we're going to have to do the little things to win in the playoffs, so we better practice doing them now!" game.
   3. Mike Emeigh Posted: August 26, 2009 at 03:39 PM (#3305590)
You can't use generic run expectation matrices in this situation - you have to look at the specific configuration of hitters coming up and tailor your strategies accordingly.

The next two hitters in the Yankee lineup were Cabrera and Jeter. Neither is a power hitter, both are more likely to make contact than Swisher and both are more likely to hit a ground ball than is Swisher (with the possibility of a double play looming with a runner on 1B). The Yankees needed a run to tie, two to win. If Swisher gets the bunt down, you have two guys who are likely to put the ball in play somewhere, against an infield that has to play in (they're not going to IBB Cabrera to get to Jeter, most likely - but if so, so much the better). If Swisher hits a (non-HR) fly ball or strikes out - both of which he does with some frequency - then you have runners on first and second (or maybe first and third) with one out and two guys who put the ball in play on the ground with some frequency. Personally, I like the odds better with second and third, one out, than with either of the other two situations.

In this specific context, given the mix of batters coming up next, I think the bunt made some sense, especially when you add in the potential element of surprise. It wasn't a great strategic decision (sacs rarely are), but neither was it a horrendous one.

-- MWE
   4. The Voice of America Posted: August 26, 2009 at 03:42 PM (#3305594)
I have no idea how to run the numbers. Somebody probably already has. So I ask, is the chance of scoring a single run more likely with 2nd and 3rd and one out than with 1st and 2nd and no outs? I feel dumb asking this but I have no idea.
   5. tfbg9 Posted: August 26, 2009 at 03:42 PM (#3305595)
One of the guys I work with was semi-ranting that the main reason it was a bad play was that Francisco was so very clearly melting down.
   6. Deacon Blues Posted: August 26, 2009 at 03:44 PM (#3305597)
do sabremetrics allow that players can "melt down"? I figured there'd be some sort of study that showed a pitcher who has given up a bunch of walks and hits in a row is just as likely to strike someone out as a pitcher who is dealing.
   7. SoSH U at work Posted: August 26, 2009 at 03:46 PM (#3305598)
I have no idea how to run the numbers. Somebody probably already has. So I ask, is the chance of scoring a single run more likely with 2nd and 3rd and one out than with 1st and 2nd and no outs? I feel dumb asking this but I have no idea.


No, it's a great question, and one that should have been answered in this piece. Getting the score tied is what's most important (never understood the whole play for the win on the road, play for the tie at home nonsense, if that's even the proper order). Do what it takes to get the score tied, then work from there. So knowing what result is most likely to get you one run is your starting point, as far as I'm concerned.
   8. Randy Jones Posted: August 26, 2009 at 03:49 PM (#3305599)
It wasn't a great strategic decision (sacs rarely are), but neither was it a horrendous one.


SG put up a post about this on RLYW that suggests even a successful bunt would have lowered the Yankees' win exp. from 61% to 55%. Seems like a bad decision to me.
   9. The Original SJ Posted: August 26, 2009 at 03:49 PM (#3305600)
In this specific context, given the mix of batters coming up next, I think the bunt made some sense, especially when you add in the potential element of surprise. It wasn't a great strategic decision (sacs rarely are), but neither was it a horrendous one.

Sac bunts are never a good decision if the bunter is a terrible bunter. You have to factor the strengths and, more often, weaknesses of the batter into the equation as well.
   10. Ray (CTL) Posted: August 26, 2009 at 03:53 PM (#3305604)
In this specific context, given the mix of batters coming up next, I think the bunt made some sense, especially when you add in the potential element of surprise. It wasn't a great strategic decision (sacs rarely are), but neither was it a horrendous one.


I don't like the bunt there, especially with Swisher, but your analysis is reasonable, Mike. Although is there something to be said for the idea that the pitcher was struggling, and so by bunting you're essentially giving him a free out even if the bunt works?
   11. APNY Posted: August 26, 2009 at 03:54 PM (#3305607)
After the first pitch wasn't even close you have to have Swisher take.
   12. SoSH U at work Posted: August 26, 2009 at 03:56 PM (#3305609)
SG put up a post about this on RLYW that suggests even a successful bunt would have lowered the Yankees' win exp. from 61% to 55%. Seems like a bad decision to me.


What about an even more successful bunt attempt, such as a failed attempt to get lead runner or some other situation where everyone's safe? Or a really, really succesful bunt attempt, where the fielder throws ball away and everyone advances? Were those figured into the calculations? (That's a serious question, since I don't know)

By the way, I'm not defending this specific attempt, just trying to look at all of the possibilities more fully.
   13. Fat Al Posted: August 26, 2009 at 03:57 PM (#3305611)
You can't use generic run expectation matrices in this situation - you have to look at the specific configuration of hitters coming up and tailor your strategies accordingly.

The next two hitters in the Yankee lineup were Cabrera and Jeter. Neither is a power hitter, both are more likely to make contact than Swisher and both are more likely to hit a ground ball than is Swisher (with the possibility of a double play looming with a runner on 1B). The Yankees needed a run to tie, two to win. If Swisher gets the bunt down, you have two guys who are likely to put the ball in play somewhere, against an infield that has to play in (they're not going to IBB Cabrera to get to Jeter, most likely - but if so, so much the better). If Swisher hits a (non-HR) fly ball or strikes out - both of which he does with some frequency - then you have runners on first and second (or maybe first and third) with one out and two guys who put the ball in play on the ground with some frequency. Personally, I like the odds better with second and third, one out, than with either of the other two situations.

In this specific context, given the mix of batters coming up next, I think the bunt made some sense, especially when you add in the potential element of surprise. It wasn't a great strategic decision (sacs rarely are), but neither was it a horrendous one.


This is a rational defense, but sitting there it seemed so wrong at the time. Francisco had thrown probably 25 pitches at that point and not to defend the "melting down" concept, but the wheels were definitely coming off the Texas bus. A double play instead of the blown bunt would have been just as deflating, but it really seemed screwy to me.
   14. aleskel Posted: August 26, 2009 at 03:57 PM (#3305612)
fallacy of the pre-determined outcome: if Swisher gets the bunt down and the runners move up to 2nd and 3rd, the infield likely plays in, Melky's liner likely gets through, Yankees win the game. Them's the breaks.
   15. Tango Posted: August 26, 2009 at 03:58 PM (#3305614)
There are two false statements being made:

1. "...runs, after all, represent the ultimate goal of a baseball game"

Actually, the ultimate goal is WINS. The tradeoff between bases, runs and outs is not static at every point in the game.

2. "This chart details how many runs a team at bat scores in any given situation. "

Actually, this chart shows how many runs will score in 24 average situations. Baseball is not about just 24 situations.

I highly recommend the 50-page chapter in The Book written by MGL as the best single source of information if one needs to quote something. I think most, if not all of it, is available for free from Amazon's Look Inside.

There are plenty of times where a bunt is justified, not the least of which is to bunt occasionally just to keep the defense honest.
   16. Darnell McDonald had a farm Posted: August 26, 2009 at 04:01 PM (#3305615)
I don't like seeing managers ask players to do things that they can't do, if Girardi was dead set on bunting there he should have pinch hit for Swisher with Molina
   17. BDC Posted: August 26, 2009 at 04:02 PM (#3305616)
Funny: from the play-by-play of the inning on Yahoo:

- J. Hairston Jr. ran for J. Posada
- R. Cano singled to shallow left, A. Rodriguez and H. Matsui scored, J. Hairston Jr. to second
- M. Cabrera lined into double play, J. Hairston Jr. out at third


It's as if the bunt attempt was so repulsive they can't even bring themselves to record it :)
   18. SoSH U at work Posted: August 26, 2009 at 04:08 PM (#3305619)
I don't like seeing managers ask players to do things that they can't do, if Girardi was dead set on bunting there he should have pinch hit for Swisher with Molina


I'd say this could be one of those situations where a manager could be thinking more long term. Swisher (and most all other major leaguers) should be able to bunt. And if Girardi thinks there's a chance he'd like Swisher to bunt in a more important game down the road, then showing Swisher (and others watching) that bunting is something he needs to be able to do is defensible. Whether Swisher and co. would get that message, or could become reasonable bunters between now and the end of the season, is a separate question.

Similarly, Girardi may have used the opportunity to see if Swisher is capable of getting a bunt down in a game situation, to inform later decsions.
   19. McCoy Posted: August 26, 2009 at 04:11 PM (#3305625)
you have to look at the specific configuration of hitters coming up and tailor your strategies accordingly.

And you have to look at the ability of the guy attempting to bunt. If the guy sucks at it you are probably better off doing something else.


edit: owe a coke.
   20. Lassus Posted: August 26, 2009 at 04:12 PM (#3305628)
An unassisted double play to kill a rally and lose a game?

Pfffft!


I was watching the game with a crowd of people, and from a non-SABR perspective, I can say the occurrence of the of the bunt attempt there was universally and violently opposed.
   21. The importance of being Ernest Riles Posted: August 26, 2009 at 04:13 PM (#3305629)

What about an even more successful bunt attempt, such as a failed attempt to get lead runner or some other situation where everyone's safe? Or a really, really succesful bunt attempt, where the fielder throws ball away and everyone advances? Were those figured into the calculations? (That's a serious question, since I don't know)

By the way, I'm not defending this specific attempt, just trying to look at all of the possibilities more fully.


Good points. In order to to do a full analysis, you need estimate the probability of an unsuccessful bunt, a successful bunt, an infield single, ROE, fielder's choice, etc. If the choice were "successful bunt" versus "unsuccessful bunt," then I am almost certain the correct choice in that situation (not looking at the batters behind Swish) is to NOT bunt. But, given the upcoming batters, the bunt is certainly defensible.

However, in addition to considering the batters behind Swisher in the lineup (as Mike noted), you also have to estimate the bunting ability of the batter himself. Having watched Swisher in Oakland, I can't vouch positively for his bunting abilities. I'm pretty sure he's a terrible bunter. Given that, I think the decision to bunt goes from defensible to terrible. Intuitively, the success probability of the bunt in that situation has to be way higher than Swish's bunting ability.

edit: cokes as appropriate
   22. Tom Nawrocki Posted: August 26, 2009 at 04:17 PM (#3305633)
You can't use generic run expectation matrices in this situation - you have to look at the specific configuration of hitters coming up and tailor your strategies accordingly.


QFT. The run expectancy matrix is useful when you have an average hitter at the plate, followed by a series of average hitters, facing an average pitcher who is backed by an average defense.

All other times, it's close to useless.
   23. Ron Johnson Posted: August 26, 2009 at 04:19 PM (#3305636)
#12 The number I have are well over a decade old. Still I think the range of outcomes is broadly similar:

Sac 71%
PO 10%
FC 7%
Hit 5%
Sac+E 3%
FCN 2%
DP 2%


FCN is a fielder's choice with no out recorded.

IOW, the base assumption of base for an out is a silly one.

#4, Another old table

This one shows the probability of scoring at least one run given the 24 bases/out situations:

_______ Bases ________Outs
_____________ 0 1 2
_____ empty .30 .18 .08
________ 1st .45 .29 .15
_______ 2nd .64 .42 .22
____ 1st, 2nd .66 .44 .24
________ 3rd .86 .67 .28
____ 1st, 3rd .88 .66 .29
____ 2nd, 3rd .85 .71 .29
1st, 2nd, 3rd .89 .69 .34

(Hopefully this will be readable)

EDIT: So close
   24. SoSH U at work Posted: August 26, 2009 at 04:23 PM (#3305643)
Thanks Ron. Even if slightly out of date, both sets of numbers are still informative.
   25. The Voice of America Posted: August 26, 2009 at 04:42 PM (#3305665)
Thank you Ron. So based on those numbers and ignoring the fact that it was Swisher. I think is a good decision to try to tie the game putting the runners on 2nd and 3rd if you are down to your last chance.

Or maybe not, I guess you would have to multiply the 2nd and 3rd with 1 out probability of scoring a single run with the probability of the bunt having the effect you want on the situation and see if it is still higher than the probability of one run with 1st and 2nd and no outs. Right?
   26. The Voice of America Posted: August 26, 2009 at 04:44 PM (#3305668)
Also, isn't it great that there is better chance of a scoreless inning with bases loaded and one out than with 2nd and 3rd and one out?
   27. Obama Bomaye Posted: August 26, 2009 at 04:44 PM (#3305669)
Sac bunts are never a good decision if the bunter is a terrible bunter.

I don't like seeing managers ask players to do things that they can't do

And you have to look at the ability of the guy attempting to bunt. If the guy sucks at it you are probably better off doing something else.

I'm pretty sure he's a terrible bunter.


Because he failed to get the bunt down does not mean he can't do it. He has 3 sacs this season, 7 in his career at a 54% success rate. He's probably not a good bunter, but he is not incapable.


fallacy of the pre-determined outcome: if Swisher gets the bunt down and the runners move up to 2nd and 3rd, the infield likely plays in, Melky's liner likely gets through, Yankees win the game.

I wonder -- would they play in there? Knowing that a grounder through the infield probably loses the game, maybe they just concede the tie, and play back to ensure they keep the tie? Or just play the corners in or something.


After the first pitch wasn't even close you have to have Swisher take.

And Girardi often does that, taking off the bunt when the count goes in his favor, putting it back on if there's a strike.
   28. The Buddy Biancalana Hit Counter Posted: August 26, 2009 at 04:47 PM (#3305673)
Matt Diaz attempted to bunt for hit with two out in the bottom of the ninth with runners on first and third. His attempt caught me be surprise, appeared to catch Heath Bell by surprise (though that may just be what Heath Bell looks like when asked to move quickly), but the bunt went straight to Adrian Gonzalez who made the play at first unassisted.
   29. Ray (CTL) Posted: August 26, 2009 at 04:49 PM (#3305675)
QFT. The run expectancy matrix is useful when you have an average hitter at the plate, followed by a series of average hitters, facing an average pitcher who is backed by an average defense.

All other times, it's close to useless.


I agree with the point but the matrix does provide a useful foundation for broadly understanding what you're dealing with as far as tradeoffs go. Then you look at the specific hitters/pitchers involved, etc.
   30. McCoy Posted: August 26, 2009 at 04:50 PM (#3305678)
Because he failed to get the bunt down does not mean he can't do it. He has 3 sacs this season, 7 in his career at a 54% success rate. He's probably not a good bunter, but he is not incapable.

That sounds pretty incapable to me.
   31. Mike Emeigh Posted: August 26, 2009 at 04:52 PM (#3305681)
Also, isn't it great that there is better chance of a scoreless inning with bases loaded and one out than with 2nd and 3rd and one out?


Double play.

-- MWE
   32. The Voice of America Posted: August 26, 2009 at 04:53 PM (#3305684)
I know, still a great statistical bit.
   33. Obama Bomaye Posted: August 26, 2009 at 05:03 PM (#3305689)
That sounds pretty incapable to me.

7 does not = 0.

In 2006-2007, Derek Jeter was successful on 10 of 17 sac attempts, 59%. Is he an incapable bunter as well?

Oh, even better, in 2001-2003 he was 42%. Everyone who watches the Yankees knows he is a worse-than-incapable bunter.
   34. Ron Johnson Posted: August 26, 2009 at 05:05 PM (#3305695)
Tom, I think useless is a vast overbid. They're pretty much first cut. Often it's easy to see than no set of assumptions will make a given option a good one. But you can easily use base/out (or win expectation tables) in a testing mode.

From an old usenet post (and you can tell how old it was from the player selected -- Joe Girardi):

Assuming for the moment that there's nobody out, runner on second, Girardi at the plate against a league average right-handed pitcher and that Girardi is a league average bunter: (and ignore for now the possibility of baserunners being thrown out and line into DP).

prob is the probability of an given offensive event
(IE Girardi could be expected to walk in 5.6% of his PAs)

exp is the expected runs scored for the resulting situation:
(IE a walk results is runners on 12t and 2nd, none out and
you'd expect to average 1.63 runs)

... prob exp
BB. 5.6% 1.63
1B. 4.8% 1.92
... 14.4% 1.97
2B. 5.3% 2.18
3B. 0.3% 2.52
HR. 1.3% 2.57
K.. 16.1% 0.73
PO. 17.4% 1.00
ROE 0.8% 1.92
GO. 33.9% 0.73
TOT 1.18

In this case, I've estimated a 75% chance that a single will score the runner (thus the two values for a single), that he'll reach on 1.5% of his non K outs (a low total) and that 1/2 of his groundouts (2/3 of his outs) will advance the runner (A tad generous). In case it's not clear, ROE is reached on error, PO is productive outs (IE an out that will advance the baserunner) and GO is a generic out (IE an out that doesn't advance the runner)

The beauty of this approach is that you can freely play with any of the assumptions and see how it impacts the results. If you think current stats represent an actual ability level, you can use them instead. (And if you want you can plug in different tables) And you can modify for the pitcher.

Applying a similar type of analysis to the bunt given an expected run value of 1.02 -- assuming he gets the bunt down of course.

One thing not considered in this analysis is the chance that the hitter will fail to get the bunt down and will have to hit away with the count against him. I've never seen any stats on this, but from what I can tell from a few old Stats Scoreboard pieces this could happen upwards of 40% of the time to guys who aren't good bunters, and something close to a quarter of the time to guys who are only competent bunters.

And this really tips the scales in the bunt/no bunt equation.

Yeah, lots of assumptions here but you can freely play with them. It always comes back to the same thing though. The bunt is rarely a bad play if base for an out is the likely consolation prize. The player needs to be a good bunter and he needs to have some chance of reaching. That chance could be his speed, defensive positioning, bad fielders, whatever -- doesn't matter.
   35. McCoy Posted: August 26, 2009 at 05:09 PM (#3305697)
In 2006-2007, Derek Jeter was successful on 10 of 17 sac attempts, 59%. Is he an incapable bunter as well?

Yes.


Oh, even better, in 2001-2003 he was 42%. Everyone who watches the Yankees knows he is a worse-than-incapable bunter.


correct.
   36. Gaelan Posted: August 26, 2009 at 05:22 PM (#3305704)
I think this is a pretty good move in this situation especially if it isn't telegraphed. I hate the sac bunt where the player just gives himself up but if you get the surprise element then you dramatically increase the chances of a hit or an error and then it is a win--win--win. I find this happens a surprising amount in stressful situations. I'll bet it happens at least once this playoffs.

That said I also think the fact that the pitcher was falling apart is something to consider. Though even that can help because if he's frustrated he's less likely to be concentrating at all aspects of the game and thus more likely to field the ball badly.
   37. Rally Posted: August 26, 2009 at 05:45 PM (#3305737)
In 2006-2007, Derek Jeter was successful on 10 of 17 sac attempts, 59%. Is he an incapable bunter as well?


What were the results of the unsuccessful sac attempts? Popups or forceouts on the lead runners?

Those sorts of plays seem rare to me. Does the above include bunt attempts where the Jeter gets two strikes and then has to swing normally?
   38. fret Posted: August 26, 2009 at 05:57 PM (#3305748)
I think this is a pretty good move in this situation especially if it isn't telegraphed. I hate the sac bunt where the player just gives himself up but if you get the surprise element then you dramatically increase the chances of a hit or an error and then it is a win--win--win.

Good point. I do wonder whether the batter is less likely to get the bunt down if he squares at the last second.
   39. Famous Original Joe C Posted: August 26, 2009 at 05:58 PM (#3305751)
fallacy of the pre-determined outcome: if Swisher gets the bunt down and the runners move up to 2nd and 3rd, the infield likely plays in, Melky's liner likely gets through, Yankees win the game. Them's the breaks.

Except if the bunt works, runners on 2nd and 3rd, and Melky is likely IBBed to set up the double play or at least the force at home. So he never hits a liner in the first place.
   40. SoSH U at work Posted: August 26, 2009 at 06:00 PM (#3305753)
Except if the bunt works, runners on 2nd and 3rd, and Melky is likely IBBed to set up the double play or at least the force at home. So he never hits a liner in the first place.


I'm not sure they're IBBing Melky to get to Jeter (and beyond), particularly when leading by a run. Maybe if the game was tied and the runner on third was the only one that mattered.
   41. Famous Original Joe C Posted: August 26, 2009 at 06:07 PM (#3305762)
I'm not sure they're IBBing Melky to get to Jeter (and beyond), particularly when leading by a run. Maybe if the game was tied and the runner on third was the only one that mattered.

Maybe you're right - but it also doesn't move up the runner at 2nd, which is the one that really counts. The runner at 1st makes no difference.

As for facing Jeter or Melky, well, yeah, you'd rather face Melky.
   42. The Original SJ Posted: August 26, 2009 at 06:08 PM (#3305764)
Similarly, Girardi may have used the opportunity to see if Swisher is capable of getting a bunt down in a game situation, to inform later decsions.

there is a time and a place for this. February, in Florida.
   43. SoSH U at work Posted: August 26, 2009 at 06:12 PM (#3305773)
Maybe you're right - but it also doesn't move up the runner at 2nd, which is the one that really counts. The runner at 1st makes no difference.


No, but having a runner on first does make it, in some ways, easier to score that runner from third.

there is a time and a place for this. February, in Florida.


February in Florida doesn't provide the exact same conditions. And if you've got a 6.5-game lead in the divisional race in late August, I think you've got a little more wiggle room for decisions with long-term implications.
   44. Ron Johnson Posted: August 26, 2009 at 06:18 PM (#3305780)
Sean, looks to be all bunts -- including bunting for a base hit. But not including the 0-2

And it ignores the 5 bunt base hits.

What I've got for Jeter in 2006-07 is 19 attempted bunts of all types. 10 sacs, 5 base hits.

10-10 in sac attempts. 5-9 on bunts for base hits (some of these could be blown sac attempts scored as an 0-1, but some of the base hits were also sacs attempting to reach. I'm using billjamesonline.net as the data source and it doesn't break things down enough to tell)

All in all, looks like a pretty effective use of the bunt to me. Consistent with the bulk of his career. Last couple of years he's cut back on the bunting even more.

No idea the frequency he's gone 0-2. Doubt it's as high as 20% of the time.
   45. baerga1 Posted: August 26, 2009 at 06:32 PM (#3305800)
#37 alluded to this, but I would imagine (though could be wrong) that bunt statistics do not include bunt attempts early in the count resulting in 2 strikes and the bunt taken off. Of course the batter is then at a huge disadvantage.

This happens all the time and frustrates me considerably. If it isn't included in the statistics, then that is a serious problem for their reliability IMO.
   46. ERROR---Jolly Old St. Nick Posted: August 26, 2009 at 06:55 PM (#3305817)
One of the guys I work with was semi-ranting that the main reason it was a bad play was that Francisco was so very clearly melting down.

That's the understatement of the year.

I don't like seeing managers ask players to do things that they can't do, if Girardi was dead set on bunting there he should have pinch hit for Swisher with Molina


I'd say this could be one of those situations where a manager could be thinking more long term. Swisher (and most all other major leaguers) should be able to bunt. And if Girardi thinks there's a chance he'd like Swisher to bunt in a more important game down the road, then showing Swisher (and others watching) that bunting is something he needs to be able to do is defensible. Whether Swisher and co. would get that message, or could become reasonable bunters between now and the end of the season, is a separate question.

Similarly, Girardi may have used the opportunity to see if Swisher is capable of getting a bunt down in a game situation, to inform later decsions.


Look, if Girardi thinks that bunting in this situation is the best strategy (which I think is nuts, but never mind that), then every non-pitcher on the roster should have been doing bunting drills for at least five minutes a day during spring training, and every day during the regular season, just so that they won't embarrass themselves the way my poor namesake did last night.

I was screaming at the TV the second the Rangers announcers mentioned the possibility of the bunt. You could have heard me in the next block by the time he squared around the first time. And when he spastically popped it up to Young, I was ready to fly up to the Bronx and slip a noose around Girardi's neck. I almost turned off the set right there, and I almost wish that I had.

Either teach everyone to bunt, and drill it into them until they can lay it down the first time nearly every time, or forget it. Wishful thinking can't bunt a 95 MPH fastball, and that's all that Girardi had going for him with Swisher up there.

Jesus, you've got four runs in, nobody out, a pitcher with no command at all (including the first pitch to Swisher), every break going your way, the crowd on its feet, and then this. It's like you're all juiced up with a woman and ready to go, and your English teacher pops in and wants you to discuss T.S.Eliot.

I've probably watched the Yankees lose at least 400 games since I got Extra Innings back in 2002, and without a doubt, this was the most disgusting regular season loss I've ever seen. I'd almost rather lose it on a dropped popup that let the tying and winning runs score---losing a ball in the lights is at least understandable. Losing your brain in the dugout is not.
   47. SoSH U at work Posted: August 26, 2009 at 07:11 PM (#3305843)
Look, if Girardi thinks that bunting in this situation is the best strategy (which I think is nuts, but never mind that), then every non-pitcher on the roster should have been doing bunting drills for at least five minutes a day during spring training, and every day during the regular season, just so that they won't embarrass themselves the way my poor namesake did last night.


I think every non-pitcher on the roster should be doing bunting drills five minutes a day during spring training. I have no idea if the Yankees follow that.

Furthermore, if Obama's numbers are correct, the majority of the time Nick Swisher has been asked to bunt he has not embarassed himself, and has actually executed the play properly, so it seems you're overstating his incompetence.

I'm not saying the bunt was a good play. I don't have enough information for that. It's just that I think there is far, far more to consider when evaluating these in-game managerial decisions than is done by fans, here and elsewhere.
   48. Obama Bomaye Posted: August 26, 2009 at 07:14 PM (#3305849)
What were the results of the unsuccessful sac attempts?

Not sure. I can't get the B-R Event Finder to show me each of his sac attempts right now, and I'm certainly not combing through every game log to find them.


Does the above include bunt attempts where the Jeter gets two strikes and then has to swing normally?

No.


What I've got for Jeter in 2006-07 is 19 attempted bunts of all types. 10 sacs, 5 base hits.

10-10 in sac attempts. 5-9 on bunts for base hits (some of these could be blown sac attempts scored as an 0-1, but some of the base hits were also sacs attempting to reach. I'm using billjamesonline.net as the data source and it doesn't break things down enough to tell)


Odd, this is quite different from B-R's numbers. B-R does show him with 5 bunt basehits but I'm not sure where or if they have his attempts.
   49. Obama Bomaye Posted: August 26, 2009 at 07:17 PM (#3305853)
BTW, is there any chance Swisher bunted on his own? The camera showed Girardi stomping down the dugout after the pop up. Could just be frustration with the outcome but it kinda looked like a typical managerial reaction to an ill-advised play. I assume this was addressed in the post-game interview though.
   50. Zipperholes Posted: August 26, 2009 at 07:18 PM (#3305854)
This one shows the probability of scoring at least one run given the 24 bases/out situations:
2nd, 3rd .85 .71 .29
1st, 2nd, 3rd .89 .69 .34


I'm sure I'm missing something here, but it's interesting that in a situation where your only concern is preventing 1 run (e.g, tie game in the bottom of the 9th), with 1 out, it's better to intentionally load the bases, whereas with no outs and 2 outs, it's better to pitch to the guy.

What am I missing?
   51. SoSH U at work Posted: August 26, 2009 at 07:22 PM (#3305859)
What am I missing?


With one out, the double play prevents any runs from scoring. With zero outs, it's not enough, and with two outs, its one too many outs. In those cases, the chances for the bases loaded walk/HBP have more weight than the force at any base situation.
   52. Ray (CTL) Posted: August 26, 2009 at 07:29 PM (#3305872)
BTW, is there any chance Swisher bunted on his own? The camera showed Girardi stomping down the dugout after the pop up. Could just be frustration with the outcome but it kinda looked like a typical managerial reaction to an ill-advised play. I assume this was addressed in the post-game interview though.


It would be hard to believe that a player made such a crucial strategic decision himself, though.

And something tells me players want to hit away there, especially someone like Swisher.
   53. Voros McCracken of Pinkus Posted: August 26, 2009 at 07:48 PM (#3305895)
As has been demonstrated here, the number of relevant variables in a situation like this is pretty immense. Hell, you could even go so far as arguing that win expectancy isn't even the most relevant number, but rather World Series win expectancy.

After all, the Yankees are an extremely strong team with a six game lead with only 37 to play. They're quite likely to win the division and get home field even with the loss. More innings gives more chances for injury, more wear and tear on the bullpen, etc.

I'm, of course, not arguing that should outweigh trying to win the game, but when you really think about it, there's just so much that can be weighed, pro and con, that you can argue about. That's why it's sometimes useful to simplify things, and go from there. I think simplistically it's a bad move, so doing it would have to set things up for down the line.
   54. Fat Al Posted: August 26, 2009 at 07:55 PM (#3305911)
BTW, is there any chance Swisher bunted on his own? The camera showed Girardi stomping down the dugout after the pop up. Could just be frustration with the outcome but it kinda looked like a typical managerial reaction to an ill-advised play. I assume this was addressed in the post-game interview though.


No, Girardi confirmed that he called for the bunt.
   55. Zipperholes Posted: August 26, 2009 at 07:59 PM (#3305919)
With one out, the double play prevents any runs from scoring. With zero outs, it's not enough, and with two outs, its one too many outs. In those cases, the chances for the bases loaded walk/HBP have more weight than the force at any base situation


Thanks. So in general, with the based loaded, the detriment/possibility of a walk/HBP exceeds the benefit/possibility of a forceout. Except that, with one out, the power of the DP (based both on the possibility of it occurring and the benefit it brings when it does) is so strong that it reverses this. Correct?
   56. Mike Emeigh Posted: August 26, 2009 at 08:05 PM (#3305934)
I'm, of course, not arguing that should outweigh trying to win the game, but when you really think about it, there's just so much that can be weighed, pro and con, that you can argue about. That's why it's sometimes useful to simplify things, and go from there.


As long as when you simplify, you don't exclude or hide anything relevant. In this case, simplifying to the BPros base/out matrix excludes or hides almost everything that is relevant.

-- MWE
   57. SoSH U at work Posted: August 26, 2009 at 08:08 PM (#3305938)
Thanks. So in general, with the based loaded, the detriment/possibility of a walk/HBP exceeds the benefit/possibility of a forceout. Except that, with one out, the power of the DP (based both on the possibility of it occurring and the benefit it brings when it does) is so strong that it reverses this. Correct?


Yes, to an extent.

Here's where I'm not sure if the proper informnation is being captured by the data that's being used. If all of of this data is created under normal conditions (run expectancies based on the boilerplate conditions - bases loaded, runner on first and second), it won't necessarily answer how the game will play out under the specific conditions of the last inning.

In other words, the various outcomes may occur because the defense is already playing with the idea of conceding a run, and the actual expectations may change when the defense is specifically designed to minimize the chances of one run being scored.
   58. Long-Time Fan Posted: August 26, 2009 at 09:18 PM (#3306015)
Every hitter to that point in the inning had reached based successfully on their own. Girardi over-manages and had to get into the act with his input. Bad thought process and bad results.

It reminded me of a game about a month ago, when Phil Hughes struck out the first two batters in the eighth inning, and looked unhittable. Girardi immediately came out to bring in Phil Coke to face a lefty. Coke was lit up and the Yankees lost.

When questioned after the game, Girardi's answer was that "he had made up his mind to bring in Coke to face the lefty before the inning started".

Oh, OK, just ignore the stuff happening on the field, and mail your decisions in ahead of time.
   59. ERROR---Jolly Old St. Nick Posted: August 26, 2009 at 10:35 PM (#3306092)
My thoughts exactly, Long-Time Fan. Girardi's a good manager overall, but Jesus, did he ever botch this one.
   60. Voros McCracken of Pinkus Posted: August 26, 2009 at 11:59 PM (#3306204)
As long as when you simplify, you don't exclude or hide anything relevant.

But it's all relevant. How relevant is the issue. Yes issues about bullpen overwork might not matter all that much, but then the run and win expectancy matrices don't actually change all that much depending on hitters (except when pitchers hitting is involved) either. Nor does the differences in win expectancy and run expectancy change all that much in this situation.

So it all matters, which is why a basic statistical framework (win expectancy in this case) can be so helpful. The numbers aren't all that matters, but I don't think you can reach consistently good decisions without factoring them in.
   61. drdr Posted: August 27, 2009 at 07:52 AM (#3306457)
Girardi is quite prone to overmanaging. Torre was more inclined to just let players play.

In this situation except the standard run-scoring probabilities, he had to take into account that the opposing pitcher has just allowed 6 runners to reach without an out. While Swisher doesn't hit good at home, he walks a lot (44/43 BB/SO in 50-something games at home). The instructions should have been: hit if you get something good to hit, otherwise, walk is perfectly acceptable.
   62. ERROR---Jolly Old St. Nick Posted: August 27, 2009 at 10:49 AM (#3306467)
In this situation except the standard run-scoring probabilities, he had to take into account that the opposing pitcher has just allowed 6 runners to reach without an out. While Swisher doesn't hit good at home, he walks a lot (44/43 BB/SO in 50-something games at home). The instructions should have been: hit if you get something good to hit, otherwise, walk is perfectly acceptable.

Which might have fit into Swisher's overall modus operandi a bit better than "Hey, let's pretend you're Phil Rizzuto!"

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
Tuque
for his generous support.

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogHouston vs. Atlanta Is Rob Manfred’s Nightmare World Series
(20 - 1:14am, Oct 27)
Last: David Nieporent (now, with children)

NewsblogMajor League Baseball work stoppage almost certain on Dec. 2
(11 - 1:10am, Oct 27)
Last: David Nieporent (now, with children)

NewsblogMLB Just Tried a Bunch of Experimental Rules in the Minors. How Well Did They Work?
(84 - 12:47am, Oct 27)
Last: Doug Jones threw harder than me

Newsblog2021 World Series OMNICHATTER!
(78 - 12:30am, Oct 27)
Last: What did Billy Ripken have against ElRoy Face?

NewsblogWith Experts on the Negro Leagues Involved, the Hall of Fame’s Era Committee Plans Are Emerging
(12 - 11:57pm, Oct 26)
Last: John Northey

NewsblogReport: Padres interview Ozzie Guillen to be next manager
(21 - 11:38pm, Oct 26)
Last: John Northey

Sox TherapyMeet the 2022 Red Sox
(11 - 11:24pm, Oct 26)
Last: The Yankee Clapper

NewsblogNBA 2021-2022 Season Thread
(359 - 11:23pm, Oct 26)
Last: Moses Taylor hashes out the rumpus

Hall of Merit2022 Hall of Merit Ballot Discussion
(195 - 10:47pm, Oct 26)
Last: Jaack

NewsblogReport: Yankees' Cole, Gardner had altercation after clubhouse prank
(36 - 10:45pm, Oct 26)
Last: My name is Votto, and I love to get Moppo

NewsblogBill James: The Strongest Franchises
(25 - 10:20pm, Oct 26)
Last: cardsfanboy

NewsblogOT Soccer Thread - Transfer! Kits! Other Stuff!
(481 - 8:44pm, Oct 26)
Last: Fancy Pants Handle struck out swinging

NewsblogBRAVES' RELIEVER TYLER MATZEK OVERCAME THE YIPS TO REACH MLB'S BIGGEST STAGE
(2 - 7:44pm, Oct 26)
Last: Howie Menckel

Newsblog‘Dear Rays, Montreal won’t pay for your new stadium’ billboard up near Trop
(11 - 5:00pm, Oct 26)
Last: Paul d mobile

NewsblogSilver Slugger Award finalists announced
(14 - 4:15pm, Oct 26)
Last: Walt Davis

Page rendered in 0.3963 seconds
48 querie(s) executed