(Photo from CBSNewYork)
In what was, for the most part, a pretty perfect Opening Day for the Yankees, the decision to have Brett Gardner bunt twice seemed to be the lone negative for Yankees fans. I have suggested on many an occasion that the “almost all bunts are bad” position that many fans seem to espouse is wrong-headed, and applauded when Joe Girardi suggested that Brett needs to become a better bunter:
So what benefits would working on his bunting have for Gardner? Firstly, working on his bunting to get his success rate over 40% would make bunting for a hit a legitimate weapon in Gardner’s arsenal. If his rate hovers around his OBP, it may not make sense to attempt to bunt for a hit too often, as he would be sacrificing any chance at extra bases without increasing his ability to reach first base. But if he could push his hit rate closer to that of Adam Jones, Angel Pagan, and Ichiro Suzuki, bunting would become a way for Gardner to increase his likelihood of reaching base. As this work by Lucas Apostoleris shows, hitters like Ichiro, Pagan, Jacoby Ellsbury, and Gregor Blanco have actually increased their value through bunting, and there is no reason that a player with Gardner’s speed shouldn’t be able to approach a similar rate of success.
Even if Gardner is not able to increase his success rate, it may make sense to bunt more often in order to keep fielders off balance. Attempting to bunt for a hit more often will force the defense to alter their positioning accordingly. Being that Gardner is a bit of a slap hitter, pulling the defense in at the corners even just a bit makes it more likely that he squeezes some hard hit liners and groundballs through the infield. Furthermore, with a player of Gardner’s speed, you’d like to keep the defense as confused as possible, as forcing defenders to be in-motion before the play begins and having a burner like Gardner going up the line can cause mistakes by fielders. Considering all of this, while Gardner may give up an extra 5 or so outs over the course of the season due to the greater frequency of bunting, it seems likely that he will earn at least that many extra bases due to the impact this strategy will have on opposing defenses.
Now, the key to this strategy is to keep the defense guessing, which means that bunting in every possible situation robs the strategy of the advantage you are trying to cultivate. As noted saberist MGL states in his fantastic game theory analysis of bunting:
On offense though, managers are way too predictable. Their mistake generally is not that they bunt too much or too little (although many managers do). It is that they don’t understand the concept that they should be mixing up their bunts and non-bunts on a random basis. This is a difficult concept for anyone to understand, much less a baseball manager. So, instead, they generally have it in their head that a bunt or non-bunt is 100% warranted in any given situation, and in most cases, the opposing manager on defense knows the same thing.
Now, in situations where a bunt is typically not called for, you might want to keep the defense off-balance by bunting 10 percent of the time, whereas in pure sacrifice situations where a bunt may actually make sense, you might try and keep the defense honest by NOT bunting in 10 percent of situations. As MGL says, doing the same thing in 100% of situations leads to a defense that is well prepared for your strategy and minimizes the chances of a mistake that can help you earn a baserunner rather than an out. Instead, you need to be unpredictable in your bunting, with the amount of times that you do the “unexpected” being dependent on the context.
Obviously, being that we only have one game worth of data to draw from, it is hard to judge whether Gardner is going to be bunting every time he comes up with a runner on first and nobody out. That said, I think we can look at the two bunts today and see two different situations that probably should have called for different strategies.
Instance #1
Inning: 3rd
Score: 1-0 Detroit
Pitcher: Justin Verlander
Baserunners: Russell Martin on 1st
Outs: None
The Play: Gardner attempted a drag bunt to the right side. Miguel Cabrera came in and fielded it, throwing Brett out by a step.
Analysis: Let’s be clear at the outset: Gardner’s OBP is high enough that he should rarely be giving himself up just to gain one base. So unless it is one of those rare situations in which the bunt makes sense in terms of win expectancy, in order for a bunt to be a positive, there needs to be a strategic advantage to the move.
This is the kind of situation where I think it makes sense to encourage Gardner to bunt for a hit on occasion. It was not a clear traditional bunting spot, so the defense, while pulled in a bit as they usually are for Brett, was not pulled in entirely. Furthermore, Miguel Cabrera is not the most nimble of fielders and the field was slick. Lastly, an ace-type pitcher was on the mound, such that moving the runner over and playing for one run if the bunt failed was not a terrible outcome. This is exactly the sort of spot where bunting some of the time will keep the defense thinking and could result in bunt hits and mistakes by fielders over the course of the season. Of course, if they bunt in this sort of spot too frequently, it will sap it of its unpredictability, but taken as a discrete event this was a perfectly acceptable situation in which to try a bunt for a hit.
Instance #2
Inning: 7th
Score: 4-3 Yankees
Pitcher: Phil Coke
Baserunners: Russell Martin on 1st
Outs: None
The Play: Gardner bunted to the right side, where Coke fielded the ball and nipped Gardner with the throw to Rhymes covering first.
Analysis: This was a more predictable bunting situation, as the Yankees had a one run lead late in the game with Rafael Soriano and Mariano Rivera waiting in the pen. Additionally, the lefty on the hill made it more likely that Gardner was going to bunt, being that the matchup was unfavorable. The defense was pulled in, expecting a bunt, and I would guess that Gardner swinging away likely would have come as a great surprise to the Detroit fielders. All of these factors suggest that bunting in this situation was an act of giving up an out for a base rather than a move to gain a long-term strategic advantage, a very poor move when the base in question is 2nd base (gaining 3rd base makes it a more defensible move, particularly when considering the strength of the Yankees bullpen, as an important insurance run could score on an out). Again, we need to add a caveat that it does make sense to bunt occasionally in these spots to keep the defense guessing. But considering the fact that the Yankees had already used that gambit once during this game, using it again in an “obvious” bunting situation was a questionable decision.
—————-
I know that on its face, this seems counterintuitive, in that I suggested they were right to bunt in the “non-bunting” situation and wrong to bunt in the traditional bunting situation. The key is to remember that teams have been bunting for a long time in situations where giving up the out is a terrible mistake, such that our perception of what constitutes a proper bunting situation is skewed. As such, the “non-bunting” situation is one where the defense might be more susceptible to a bunt hit, and it therefore makes sense to bunt in a greater percentage of such situations, at least until the defense adjusts its strategy. Conversely, the traditional bunting situation was one where the win expectancy went down, such that a sacrifice was not justified in of itself, AND the odds of a mistake by the defense were fairly low due to the predictability of the bunt. Those are the situations where bunting should be kept at a minimum, only done occasionally to keep the opponent guessing.
As I suggested above, we only saw two of these situations today, such that it is possible that these bunts are not representative of the decision-making that the team will employ in bunting situations. Hopefully, Joe Girardi and Brett Gardner will use bunting as a tool to disorient defenses and gain baserunners rather than as a way to trade outs for advancing one runner a single base.
If you enjoyed this article, please consider sharing it!
Interesting analysis. I agree that it’s good strategy to mix it up. Here are some thoughts
– During his press conference, Girardi said the bunt (or the bunts) was Gardner’s choice. I was unclear if Girardi was talking about the first bunt only or both of them.
– I thought I heard someone say Gardner wasn’t feeling 100%. Maybe they said he had a cold. Presumably Gardner should be more likely to sacrifice when he’s less likely to get a hit, because he’s unwell or because the pitcher is too tough, etc.
– It’s hard to criticize Gardner’s 2nd bunt, since it did lead to scoring a run.
– Aside from the stragy question, I thought both bunts were well executed.
[Reply]
Moshe Mandel Reply:
March 31st, 2011 at 8:02 pm
These are all fair points, although I will say that the scoring of the run doesnt mean it was good strategy. If he isn’t feeling well, or something to that effect, and he was more comfortable bunting, that is fine.
[Reply]
I pretty clearly heard Girardi say that Gardner bunted on his own, for hits, both times.
If that’s the case, I agree that the first bunt was a better idea than the second. I also think Gardner needs to learn to bunt better. Neither bunt was dragged really well. Maybe it was just the pitches, but you have to hit that hole in the defense when you bunt for a hit.
[Reply]
Moshe Mandel Reply:
March 31st, 2011 at 8:04 pm
This changes the calculus a bit, but not much. The manager has to impart the strategy to the players. So, for example, he could give Gardner the green light while telling him that in certain situations, bunt X% of the time. That way, Girardi is guiding the strategy while Gardner is actually making the in the moment decisions.
[Reply]
I disagree with you Moshe. I may base my 9 AM post around that disagreement.
[Reply]
I think Gardner needs to learn that the key to a successful drag bunt is getting it past the pitcher. He seems intent on getting it 15′ in front of the plate.
[Reply]
T.O. Chris Reply:
March 31st, 2011 at 10:33 pm
I’d like to see Gardner bunt the bal to third more but with intent on killing the ball and letting it die 5′ away, that way the 3rd baseman has to come in so far for the ball with Gardy’s speed it’s a hit every time.
[Reply]
[...] Speaking of Gardner, he dropped down two bunts in this game, neither of them resulting in a base hit, though the first seemed like more of an attempt at a hit, while the second looked more like a pure sacrifice (both were successful sacrifices, whatever the intention may have been). Over at the Yankee Analysts, Moshe Mandel took a closer but sufficiently succinct and measured look at Gardner’s two bunts and the idea of Gardner bunting in general. I agree with every word of his post and suggest you give it a read. [...]
I think there are several flaws with MGL’s game theory theory. The point of a sacrifice is not to trick the defense; it’s to advance runners. If well executed, a bunt will be successful even if announced (which is kind of what squaring around really is). It doesn’t make sense to turn the strategy into a trick play. Either you are willing to give up the out or not. If the motivation is to get an extra baserunner, then why bunt in the first place?
I also find it hard to believe that teams are really going to be fooled by an occasional red herring bunt or two, especially from a team like the Yankees that bunts so infrequently (which begs the question, if the Yankees rarely bunt, why do teams ever try to defense it). Teams usually play the batter and the situation more than the manager’s tendencies.
Having said that, I agree that Gardner’s two bunts were reasonable plays for all the reasons you cited. The variables are what determine whether a bunt is a good play. I don’t think you can make wide sweeping assessments about the strategy.
[Reply]
Moshe Mandel Reply:
April 1st, 2011 at 12:33 pm
Regarding your first point, the idea is that giving up the out to advance the runners is rarely a good move. As such, its value is dependent on it actually gaining runs for you over the course of a season. As I stated in my initial post, a guy like Angel Pagan can bunt for a hit at a greater rate than his OBP, which actually makes it a high percentage play. If Gardner could approach that level, he could raise his effectiveness while impacting the defense every time he comes to the plate. It is simply another thing the defenders have to worry about.
As for your second point, this sort of thing will absolutely show up in scouting reports. In this day and age, where teams are cataloging everything, I have no doubt that a 3rd inning bunt will show up in the data.
[Reply]
William J. Reply:
April 1st, 2011 at 3:43 pm
I don’t believe many, if any, players can bunt for a hit more than 30% of the time if done regularly, so I disagree there. My main point is bunting should only be done when giving up the out isn’t detrimental (like in a walk off situation where the expectancy of one run is improved at the expense of a big inning).
I also don’t think one 3rd inning bunt is going to register on many radars, nor should it. Besides, what is gained by sabotaging a handful of situations to benefit in a handful of others?
The game theory is fun, but I don’t think it has any useful application in a real game.
[Reply]
Moshe Mandel Reply:
April 1st, 2011 at 7:52 pm
Well, the idea is that if you get good enough at it, you aren’t sabatoging anything. And as the data in my first post shows, plenty of hitters do it well enough to do it at a rate better than their OBP and bunting gives them positive value.
[Reply]
William J. Reply:
April 1st, 2011 at 4:14 pm
With the exception of Bourn and Gomez, I don’t think the other names in that list are bunting “regularly”, but I agree that Gardner has the speed to be a very successful bunter.Also, do those kinds of players really need to trick the infield? I have a feeling they draw the infield close anyway (because the opposition is basing its strategy on a specific context).
[Reply]
“[The] value [of giving up the out to advance the runners] is dependent on it actually gaining runs for you over the course of a season.”
Not necessarily. If it’s late in a game, and you’re tied or leading by 1 or 2 runs, getting 2 additional runs is not twice as valuable as getting 1 additional run. In these situations, you want to maximize your probability of getting at least 1 additional run, rather than maximize the expected number of additional runs.
[Reply]
Moshe Mandel Reply:
April 1st, 2011 at 6:23 pm
Agreed. As I noted (or at least tried to allude to) in the post, there are situations where playing for one run is the right move, and a bunt makes sense in of itself.
[Reply]