Beating a Dead Horse - An Analysis of Presley's Bunt
by DG Lewis on Jul 21, 2011 10:56 AM EDT
Email
Print
Was it good small-ball strategy? Was it horribly giving away a valuable out? Let's figure it out!
The logic behind having Alex Presley bunt is that it moves the tying run into scoring position. And, per baseball conventional wisdom, "at home you play for the tie."
But what is the likelihood of the Pirates tying the game with Chase d'Arnaud at the plate, runners on second and third, and one out?
d'Arnaud is hitting .225. However, a single doesn't automatically tie the game, because Matt Diaz isn't exactly a speedster - in his career, he's scored from second on a single only 56% of the time. Furthermore, 7 of d'Arnaud's singles were infield singles, on which Diaz will not score. So the chance of d'Arnaud tying the game is the chance of him hitting a single to the outfield times 56%, plus the chance of d'Arnaud hitting a double or triple. Seeing as how d'Arnaud has never homered in the majors, and homered only 18 times in 1567 minor league PAs, I'm disregarding that possibility.
d'Arnaud has 15 singles, 4 doubles, and 1 triple in 94 PAs. 7 were infield singles and one a bunt single, so 50% of d'Arnaud's non-bunt singles were infield singles on which Diaz will not score, and 50% were outfield singles on which he will score 56% of the time.
15/94 * .5 * .56 = 4.47% chance of d'Arnaud singling and Diaz scoring.
5/94 = 5.32% chance of d'Arnaud doubling or tripling (on which we assume Diaz scores).
So after a successful sacrifice bunt, there's about a 9.8% chance of d'Arnaud tying the game. Is a 10% chance worth giving up an out by taking the bat out of your hottest hitter's hands? You be the judge.
If Presley doesn't bunt, there are a range of possible outcomes:
- He could homer or triple, or could hit a double on which Diaz scores from first (which historically he's done 44% of the time), tying the game or putting the Pirates into the lead. With one home run, three triples, and four doubles in 88 ML PAs, the probability is (1/88) + (3/88) + (4/88) * .44 = 6.55% chance of Presley tying the game or putting the Pirates into the lead.
- He could double and Diaz not score, putting runners on second and third with no one out and the Pirates down 2-1: (4/88) * (1-.44) = 2.55%.
- He could single or walk, loading the bases with no one out. (I'm assuming that McKenry doesn't score from second on a single.). With 19 singles and 8 walks, the probability is 27/88 = 30.7%.
- He could ground into a double play. He's never done so in the majors, but doing some calculations to estimate DP opportunities in his minor league career, I estimated that he's got maybe a 5.9% chance of doing so.
- He could make a "productive out", putting runners at second and third with one out; I calculated that probability at about 32.6%.
- He could make an unproductive out, leaving runners at first and second with one out. I calcuated that probability at about 21.7%.
So basically, by not having Presley bunt, you have about a 40% chance of an outcome better than a successful sacrifice bunt. You have about a 33% chance of an outcome the same as a successful sac bunt, and about a 22% chance of an outcome the same as an unsuccessful sac bunt. And you have about a 6% chance of a worse outcome than an unsuccessful sac bunt.
To me, the upside outweighs the downside. Again, you be the judge.
This is a FanPost and does not necessarily reflect the views of the managing editor (Charlie) or SB Nation. FanPosts are written by Bucs Dugout readers.
52 comments
|
10 recs |
Do you like this story?
Comments
Are you calling me a Wonk(a)?
Just because I typically have 6 spreadsheets open on my desktop at any given time…
Nice work.
To me, a side issue is letting d’Arnaud hit, whether the bunt is successful or not. If it succeeds … maybe he hits. But if it doesn’t (and it didn’t), what’s he doing up there? Not like they didn’t have Harrison on the bench to pinch hit and Ciriaco to go in and play short. In fact, Ciriaco could have gone in to run for McKenry BEFORE the bunt and then stayed in once Harrison hit for d’Arnaud.
But then I guess that leaves you with no options off the bench later in the game if you don’t score, and you might look silly sending a reliever to the plate down two runs in the ninth …
We can look at the WPA changes for a different perspecitve
Of course, these expectancies assume an average team vs another average team, with an average run scoring environment. With the low environment this year, we may want to, for the sake of discussing this specific play, shift the numbers a bit. I wouldn’t move them more then a % point or 3, but a little wiggle room can be justified.
The Win Expectancy for the home team batting in the 7th, down 2 with runners on 1st and 2ns with 0 outs is .373
If the bunt gets down successfully, we have runners on 2nd and 3rd with 1 out. That WE is .388, so it looks like the bunt may have been a good move.
Lets look a little deeper. I am going to use the numbers in the OP because I’m too lazy, and I don’t want to get into arguing his numbers.
A .195 chance of a single and .030 chance of a walk. (I’ll assumee that we do not score from 2nd, to bias this a bit toward the bunt. BB% from Fangraphs) So, we have a .225 chance of bases loaded, no outs, which gives a WE of .548
He has a .043 chance of a double, I will again assume slow/cautious runners giving us now a 2nd and 3rd, no outs, down 1 run for a WE of .685
For a triple a .010 chance, leaving the game tied, runner on 3rd, no outs. WE of .707.
Now for the negatives. There is a .032 chance of a Double play. We will assume here that the lead runner goes to 3rd. Now, runner on 3rd, 2 outs, down 2. WE is .181
The .032 chance of the double play is removed from his overall 1-x OBP (fail rate), which gives us (.750-.032) .718. 20% of that is .143 that we end up with runners on 2nd and 3rd with 1 out. WE of .388
The rest of the ABs result in an out, with the runners where they were. .607 giving us a WE of .304
So lets do the math. We’ll take the probability of the event, times the WE of the state after. This will give us a better picture of teh expected WE, risks included, of letting him bunt. (Prob*WE)
(.225*.548)(.043*.685)(.010*.707)(.032*.181)(143*.388)+(.607*.304)
Which gives us an expected WE, by letting Chase bat, of .406. As I mentioned earlier, if we assume a perfect bunt (unlike what happened) the WE is .388. By bunting we give up about 2% wein expectancy. Not huge, but, like Jonah Keri’s book, ithe the other 2% that counts.
People
seem to be leaving out the fact that even if D’Arnaud does nothing, Walker is still up to bat with guys on 2nd and 3rd with no outs.
I guess a reasonable strategy though, considering how hot Presley is and how hot D’Arnaud is not, is to let Presley swing away, and assuming he makes a non-productive out, you have D’Arnaud bunt and then you are in the same situation.
Personally, regardless of what the numbers say (although that’s a very dangerous thing to utter because it means I’m relying on intutition), down 2 late in the game, it makes sense to put yourself in a position where two guys have a chance to get a hit to tie the game. FWIW, I’m assuming (hoping?) Hurdle would have pinch run for Diaz had the bunt been successful. Honestly, I can’t remember the pinch running scenario was plausable given the game situation. I’m too lazy to look it up.
Bleh, I know this will get shot down, but I honestly don’t care. I don’t know why.
Da'Sean Butler - A Mountaineer Legend
by McCutchenIsTheTruth on Jul 21, 2011 11:55 AM EDT reply actions
if D’Arnaud does nothing, Walker is still up to bat with guys on 2nd and 3rd with no outs.
Don’t you mean two outs?
down 2 late in the game, it makes sense to put yourself in a position where two guys have a chance to get a hit to tie the game.
By bunting, you’re giving two guys a chance to get a hit to tie the game – although in Chase’s case, it’s only a 10% chance. (You can run the same numbers on Walker – he’s got about a 15.5% chance of tying the game (or putting the Pirates in the lead) if he’s batting with runners on second and third and two out.)
But by not bunting, you’re giving three guys a chance to tie the game. Which is better than two.
I guess a reasonable strategy though, considering how hot Presley is and how hot D’Arnaud is not, is to let Presley swing away, and assuming he makes a non-productive out, you have D’Arnaud bunt and then you are in the same situation.
Haven’t run the numbers, but I actually wouldn’t be averse to this strategy. d’Arnaud is more likely to make an out than Presley and more likely to GIDP. And if d’Arnaud gets the bunt down, the Reds aren’t going to pitch around Walker to face Cutch – he’s going to get pitches to hit.
I’m assuming (hoping?) Hurdle would have pinch run for Diaz had the bunt been successful.
Ciriaco could have pinch run. Presumably he’d be more likely to score from second on a single than Diaz, but that would only increase the probability of d’Arnaud tying the game by a percentage point or two.
Yes
two outs. My mistake.
My other points stand I think.
Again, my intuition is fighting what the numbers say right now, and for now my intuition is winning. My real issue is probably the fact that I always bunt in this situation when it comes across in a video game (mlb 2k11 – anytime I have 0 outs and guys on 1st and 2nd). The big difference is a bunt if almost a guarentee to be successful in moving the runners. Hitting the Y button is fairly easy. So on and so forth.
Da'Sean Butler - A Mountaineer Legend
by McCutchenIsTheTruth on Jul 21, 2011 12:57 PM EDT up reply actions
You haven't calculated the probability of a successful bunt.
I think the overall success of sac bunts is about 75%, but that is mostly sac bunts trying to get the runner to second. It’s certainly less for a situation when you have a force out at 3B.
I would guess a sac bunt with runners on 2nd and 3rd is close to twice the difficulty, because you have less area to bunt it into to avoid doing exactly what Presley did.
Seems like the odds are very much against the strategy of bunting in that situation.
Best I can find
is this HT article from 2005, which puts the success rate of sacrifice bunts with runners on 1st and 2nd and nobody out at about 70%.
So bunting gives a 30% chance of 1st/2nd 1 out and 70% chance of 2nd/3rd 1 out.
Not bunting gives a 6% chance of a runner on 3rd and 2 out, a 22% chance of 1st/2nd with 1 out, a 33% chance of 2nd/3rd with 1 out, and a 40% chance of none out with either the bases loaded or one or more runs in. (Numbers don’t add to 100% due to rounding.)
In each case, you have about a 70% chance of “success” and a 30% chance of “failure” – but by swinging away, about 40% of the time your “success” means no out and either bases loaded or runs in, which is never the case if you bunt. And all you’re risking is a 6% chance of a “failure” that leaves you with two out and a runner on third. To me, that’s a no-brainer – swing the bat, Alex.
reality is that it has to be lower with these Pirates...
they are very poor at executing the bunt.
#AllTheBuntsAreBad!
Presley himself is a good bunter. It just shows that it’s not automatic to lay down a perfect bunt in that situation with a slow runner.
by MarkInDallas on Jul 22, 2011 1:05 AM EDT up reply actions
A perfect bunt in that situation would have forced the 3B to field it, not the pitcher.
The slow runner did not help the situation, but if Presley had forced the 3B to field the ball it wouldn’t have mattered who was running.
by insane_sanity on Jul 22, 2011 10:24 AM EDT up reply actions
To me, the upside outweighs the downside. Again, you be the judge.
What position are you taking here? Do you mean the upside of bunting outweighs the downside? Because that’s not what I’m getting from the numbers.
Poorly worded, sorry
My opinion is the upside of having Presley swing away outweighs the downside.
Enough already!
Criminy! You people are enough to “stat” someone to death.
For Pete’s sake. COMMON SENSE tells you in two seconds what it took an entire page worth of stats – or at least a minute even for a speed reader – to tell you.
A bunt by Alex Presley was entriely the WRONG choice.
If more people used their common sense (actually, if more people HAD common sense), baseball – indeed ALL sports – might actually be fun to watch again.
Guess
I’m an idiot.
Da'Sean Butler - A Mountaineer Legend
by McCutchenIsTheTruth on Jul 21, 2011 3:40 PM EDT up reply actions
You've got the touch!!!!
You’ve got the power!
#AllTheBuntsAreBad!
by Slick1 on Jul 21, 2011 4:35 PM EDT up reply actions 2 recs
Feel free to not read any posts that are too difficult or boring for you...
I happen to think this was a well thought out post and provided a solid argument for why bunting was stupid. This argument has gone on all season which is why the post required the detail it had. Oh and, what do you mean you people?
#AllTheBuntsAreBad!
Bunting has once again become a problem with the Pirates. Not only are the pitchers namely Maholm doing a bad job, but some of the position players are as well
The designated hitter position which is now employed at so many levels of baseball, is probably the reason players no longer develop the skill..
err here is something i would like to just throw out there. with a lefty on the hill, should presley have bunted down the first baseline? the point of dropping it down to 3b is to draw that fielder in but does anyone recall if the 1b was in for the bunt. having a lefty stop momentum, and throw the longer distance, i would assume the chances were better for the runner to be safe
Thats what she said! - Michael Gary Scott
no
if the 3B is laying back, odds are the 1B was charging and the 2B was covering 1B.
In that situation, the P is covering the 3B side and it is the bunter’s responsibility to MAKE the 3B have to field the ball by bunting it harder down the 3B line. With Presley’s speed, a well placed bunt down the 3B line (that forces the 3B to field it) could even result in a bunt base hit.
If he doesn’t bunt it far enough, the pitcher fields it and the 3B is staying home to force the runner.
by insane_sanity on Jul 22, 2011 9:54 AM EDT up reply actions
What about a sac fly?
This would have been another positive output if the bunt was successful. Wasn’t it only in the 7th. If so, one run would have been nice.
by ballparkfranks on Jul 22, 2011 12:44 AM EDT reply actions
thats the point, if the bunt is executed, we should at least get 1 run from the inning.
Thats what she said! - Michael Gary Scott
So we would have lost 3-2 instead of 3-1
Considering we didn’t even get another baserunner the rest of the game, 1 run in the inning does not impress me.
I just want to say that it's fantastic
that we can talk about whether a single bunt might have negatively affected the Pirates’ season.
I think we talk about it
as a learning lesson. Or rather, that we hope Hurdle learned some sort of lesson that he doesn’t repeat.
by Central*Scrutinizer on Jul 22, 2011 1:22 AM EDT up reply actions
Actually, I think I saw
a butterfly near home plate during the at bat – did the flap of the butterfly’s wings affect the outcome of this play?
I reccy’d because this is a good, well-worked post but I fall into the “lets not get into paralysis by analysis” camp. By conducting a statistical review of all fanposts, fanshots, and comments on this site, I can safely say there is a 53.127483% that if Hurdle had let Presley swing away, Presley k’s, and we don’t score in this situation, that a substantially similar post blasting Hurdle for not bunting would have been up within 3.2846387 hours…..
What does that mean?!
Other factors for probability adjustments?
Weren’t the corners in on that play? And middle infielders cheating? Wouldn’t that raise Presley’s hit probability also (all I know is the traditional 100 points higher theory on drawn in infields)? Presley has great bat control. It’s amazing how quick he can jump on pitches before defenses have a chance to react. If he shows the bunt and swings away, the defense is out of position. My gut tells me his chance of getting a hit by swinging away was actually somewhere between .45 and .5.
by Central*Scrutinizer on Jul 22, 2011 1:21 AM EDT reply actions
Right Call
Having Presley bunt was absolutely the right call! With how pathetic this offense has been lately, you have to play small ball to score runs. Clint Hurdle has shown that he is willing to play small ball by having McCutchen and Overbay bunt. When those bunts were successful, the call was praised. Had Presley laid the bunt down in a better spot, and possibly event beat the throw to first too, this call would have been praised. People need to stop over analyzing with stats and realize that this team is not going to score much…so we have to manufacture runs wherever we can.
oh, yawn
i dont have the energy to refute the same thing again…
When those bunts were successful, the call was praised.
Let’s see some? Hurdle’s propensity to bunt is a running joke in the gamethreads…
When those bunts were successful, the call was praised.
Not be me they weren’t.
Presley laid the bunt down in a better spot, and possibly event beat the throw to first too, this call would have been praised
Not be it wouldn’t. I would have said that the Pirates got lucky given Hurdle did not put the team in the best position to win the game.
This isn’t a hindsight thing it’s a fundamental disagreement over the value of an old school baseball play and whether said play still holds value today given what we know now. I don’t know about the rest of you guys but when I go to the casino and play Black Jack I go with the odds.
#AllTheBuntsAreBad!
Oh and you can go with your gut at the black jack table all you want...
but be prepared for everyone else at the table to chew you out.
#AllTheBuntsAreBad!
People need to stop over analyzing with stats and realize that this team is not going to score much…
so the Pirates should stop giving the other team an edge by playing with less outs than them (fnished that for you). If you have a team that struggles to hit it’s only logical to give them more chances to do so, not less. Anyway, I don’t know why we’re still arguing when it’s pretty much been proven statistically that it was a stupid move.
#AllTheBuntsAreBad!
The problem with stats. . .
Bunting over a big sample size has shown to be less effective. As is discussed here, even taking into account the past results of the players involved the percentages lean toward not bunting.
The problem I take is that Baseball is not a random number generator. These odds aren’t constant, they fluctuate with every individual situation. The players involved are not exactly the same each at bat. Managers make these decisions in the moment, using the information they have available, they should be informed of stats, but more importantly they have to judge the players in their dugout at the moment, the other team, and that old gut instinct that statistics tell us to ignore.
Maybe it was the wrong move, maybe it didn’t make a difference, who knows what would have happened if Pressley swung, we can point to odds, but even though Chase he hasn’t hit a HR yet doesn’t mean he never will.
My point is that I believe that unless the numbers show a big difference in percentages, like pinch hitting Chase for Walker with the bases loaded (okay that is ridiculously extreme, but you get the idea), the non-quantifiable parts of the game more than make up the gap most bunt or not bunt stats reveal. There are times that the bunt will work out better than swinging, it isn’t as often as the reverse, and you aren’t always going to be right. That doesn’t mean those times where going against the stats is the right call don’t exist. They do exist, and often recognizing when to go against logic is the difference between winning and not winning.
In the end I think there’s a stat we should keep in mind any time we criticize Hurdle, and that would be the gap between how many games our stats say the Pirates should be winning and how many they are winning.
It’s not a perfect stat, but if his decisions were hurting us you’d think it will eventually show. Who knows, it might.
by Phantaskippy on Jul 22, 2011 1:32 PM EDT reply actions 1 recs
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Bucs Dugout
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
Baseball Nation Recent Stories
In Association With
Big League Stew - MLB - Yahoo! Sports
- College player makes impossible over the fence catch to save tournament game (Video)
- Security!: Camden Yards guard collides with Royals’ Mike Moustakas
- The Juice: Nelson Cruz matches career high eight RBIs, Pujols stays hot in Seattle
- Benches clear at Fenway as Red Sox-Rays rivalry heats up
- Keeping good company: Albert Pujols becomes fourth-youngest player to hit 450 home runs
