Joe Maddon’s Bunting Identity Crisis

Two facts, with which you as a FanGraphs reader are likely familiar:

  • The Tampa Bay Rays are among the most sabermetrically-inclined organizations in major league baseball.
  • Sabermetrically-inclined folk generally are against the decision to sacrifice bunt.

One more fact, with which you are less likely to be familiar:

  • The Tampa Bay Rays have attempted 58 non-pitcher sacrifice bunts this season, by far the highest mark in the major leagues. No other team has even 50.


So that’s weird. Since he began leading the Rays in 2006, Joe Maddon has been known as one of the more progressive MLB managers. He does funky things with his lineup. He’s batted the pitcher eighth. He made heavy use of the defensive shift before it was cool. If there was a manager to be considered the posterchild of what a future MLB manager might look like, it has been Joe Maddon.

He’s even spoke out publicly against sac bunting in the past:

“For that group of people out there that want guys to bunt all the time, you don’t know the outcome when you choose to do that,” Maddon said, of choosing not to bunt with two runners on base and no outs in the ninth inning, and again following a leadoff double in the 10th. “I think the bunt is an overrated play.”

Joe Maddon thinks the bunt is an overrated play, yet here we are in the year 2014 and the Rays have bunted more than anyone. Well, attempted to bunt more than anyone. And therein lies the issue with Maddon’s bunting habits this season.

League-wide, the use of sacrifice bunts are trending downwards.

sac2

This makes sense. As more data suggests sacrifice bunting is generally bad, managers will slowly but surely catch on and begin picking their spots wisely. That leads to this:

sac

It seems that MLB managers are picking smarter times to use the sacrifice bunt. League-wide WPA on sac bunts is positive for first time in 10 years and probably longer, though I did not have the patience to continue my research, nor the wherewithal to devise a more efficient method than the one I was employing.

So that’s how the league is adapting with regards to sacrifice bunts. Doing it less often, but doing it more efficiently. Let’s get back to Joe Maddon’s Rays.

As previously stated, the Rays have attempted to bunt more than anyone, with 58 non-pitcher attempts. But they’re not in first in completed sacrifices. Just 35 of those 58 attempts have turned into “successful” sacrifice bunts. I put successful in quotes because we know that, more often than not, a sacrifice bunt even when executed does more harm than good. Despite all of their bunting efforts this season, the Rays sacrifice bunts have amassed a negative WPA. That is to say, all the free outs they’ve given away haven’t really improved their win expectancy at all. It’s hurt them, if anything. To compound the issue, let’s go back to that 35-of-58 number.

35-of-58 yields a 60% success rate. That’s bad. The league average success rate for a sacrifice bunt is 71%. Only five teams have lower success rates on bunts than the Rays this year. All of those teams have attempted at least 11 fewer sac bunts than Tampa. Brandon Guyer is the biggest offender. His 12 sacrifice bunt attempts are second on the team, yet he has only successfully laid down five for a 42% success rate. Desmond Jennings is 9-for-13. Jose Molina is 4-for-6. Yunel Escobar, Kevin Kiermaier and Sean Rodriguez are all 3-for-5. There isn’t a single Rays player who has attempted more than two sacrifice bunts this season that has actually got them down at a league-average rate.

We’re going to get slightly off topic here for a second, but something else caught my eye while doing this research. Two paragraphs above, I presented the fact that the Rays, despite having attempted more sac bunts than anyone, have not executed more sac bunts than anyone. Instead, that title goes to Terry Francona’s Indians, with a league-leading 38 successful sacrifice bunts. The Indians, like the Rays, are known as one of the most progressive organizations in baseball and Francona has a reputation as a progressive manager from his time with the Theo Epstein-led Red Sox who didn’t bunt at all. From 2004-2011, when Francona was at the helm in Boston, the Red Sox bunted just 176 times, 33 fewer than the next-lowest team.

Part of that stems from the fact that the Red Sox had one of the best lineups in baseball during that stretch with a 110 wRC+ and therefore didn’t need to rely on small ball strategy to push runners across the board. Then again, both the Indians (104 wRC+) and Rays (102 wRC+) have top-1o offenses in baseball this season which begs the question of why Joe Maddon and Terry Francona have suddenly fallen in love with the sacrifice bunt?

The Indians have at least bunted well, which is more than the Rays can say, with an 82% success rate that is topped only by the Rangers’ 86%. But all those “successful” sacrifice bunts haven’t yielded a positive WPA for the Indians, either.

To be honest, I really can’t think of a good explanation as to why Maddon and Francona have fallen in love with the sacrifice bunt this year. Both have proven to be anti-bunt in the past and have strong lineups, yet rely on the bunt more than any other manager in baseball seemingly to a fault.

Just for fun, since we’re talking about the Rays and the Indians, what do the bunting habits of the Moneyball A’s look like? Fewest in the league, with just 12. Part of that is due in part to their league-worst 44% success rate, but they’ve also attempted just 24, the sixth-fewest in the MLB.

Three of the teams in the MLB most generally perceived as “progressive” have had unique seasons with regards to the sac bunt this year. The Indians have laid down more sac bunts than any team in baseball, which would appear to be bad on the surface, but at least they’ve been among the most efficient at getting them down. The Athletics have chosen not to sac bunt at all, but part of that is because they’ve just been terrible at it. Then there’s the Rays, who have combined the worst of both worlds, being both terrible at bunting while also attempting them more than anyone. Maybe Joe Maddon should listen to his own advice and concede that the bunt is overrated.



Print This Post



August used to cover the Indians for MLB and ohio.com, but now he's here and thinks writing these in the third person is weird. So you can reach me on Twitter @AugustFG_ or e-mail at august.fagerstrom@fangraphs.com.


Sort by:   newest | oldest | most voted
Nick
Guest
Nick
1 year 10 months ago

“Maybe Joe Maddon should listen to his own advice and concede that the bunt is overrated.”

Maybe we should concede that Joe Maddon has earned enough respect to do something we think is odd and have a damn good reason for it..

LaLoosh
Guest
1 year 10 months ago

wait, so if Kirk Gibson were leading the league in sac bunts, you wouldn’t be calling him a thick-headed Neanderthal? and Joe Maddon conversely can never do anything that falls into the “questionable” category?

that sounds fair…..

Cody
Guest
Cody
1 year 10 months ago

Track record accounts for something… fair or not. It is why the idea of a reputation exists.

Nick
Guest
Nick
1 year 10 months ago

Questioning him and concluding that he should just stop are two different things. You don’t think its possible that he knows what he is doing? Maybe he should just stop managing completely, since we can just give the Rays a win probability for every decision and run the team that way..

He has built up enough credibility that I am willing to see what he is up to before judging it a failure and telling him to stop.

Yirmiyahu
Member
1 year 10 months ago

Okay, so let’s make it a conversation. The statistical evidence indicates he should stop. What’s the argument to the contrary?

Buns Slugsworth
Guest
Buns Slugsworth
1 year 10 months ago

Strikeouts and ground balls are up and batting average is down. This doesn’t defend bunting inefficiently, but if the trends continue it will be nice to have an organization ready to bunt.

LaLoosh
Guest
1 year 10 months ago

this is the “Sabr-” boilerplate answer: when the question is about the A’s, Theo, Alderson, or the Rays the answer is always “well, they must know something we don’t know.” And when the question is about Dayton Moore, Brian Sabean, Kevin Towers or JackZ, the answer is always “well, they are old skool dumb fuck nitwits lucky to be part of the old boy network…”

Buns Slugsworth
Guest
Buns Slugsworth
1 year 10 months ago

“organization” =/= “The Rays”
August posted the WPA graph above with the conclusion that teams are bunting at more opportune times, but the recent batting trends show that bunting is more often opportune this year than it has been for many years. I am clearly not concluding that Maddon has made the correct decisions or that bunting needs to increase. I am continuing dialogue as the OP requested with a couple of ideas as to what Maddon or the Rays may be thinking.

Buns Slugsworth
Guest
Buns Slugsworth
1 year 10 months ago

It is “Sabr-” to evaluate.

Tim
Guest
Tim
1 year 10 months ago

There’s no worthwhile statistical evidence in 58 bunts. Even league-wide year-to-year WPA probably has too mach variance to be particularly useful. In order to learn anything from the bunts it seems to me that we should look at them and their situations individually.

I would rephrase Nick’s original comment to suggest that Joe Maddon has earned a much deeper analysis of his unconventional or unexpected choices than is provided in this article.

Matt
Guest
1 year 10 months ago

You’re an idiot, LaLoosh.

AK7007
Member
AK7007
1 year 10 months ago

You must be new here! Welcome to the wide world of “sabr sucks” shouted from LaLoosh’s, and many other fine individual’s keyboards.

To be fair though, Nick wasn’t exactly helping the conversation here – a more helpful comment would have been to brainstorm ways that Maddon might not be crazy instead of proclaiming faith and devotion.

Johnston
Member
Member
Johnston
1 year 10 months ago

Matt, welcome to FanGraphs!

Nick
Guest
Nick
1 year 10 months ago

To be fair – I am an Orioles fan and could really not care less about Joe Maddon, so it’s not faith and devotion. My point was that this article could have been shortened to say “Joe Maddon suddenly likes bunts. Bunts suck. Joe Maddon should stop liking bunts.” and we would have gotten the same amount out of it. Yes I could have suggested some ways that he might not be crazy, but my point stands that Joe Maddon is a better baseball manager then I, and seems to understand what he is doing, so I will trust that he has not just randomly decided to bunt more this year. I don’t think he woke up on opening day and said to himself “you know what I am going to bunt more this year.” If he did then he isn’t as thoughtful as I give him credit for. What those reasons are, I don’t know.

Three Little Birds
Member
Three Little Birds
1 year 9 months ago

IMHO, Joe Maddon isn’t a great manager, he’s not even a good manager. Forget about batting Jeff Keppinger cleanup, why would he be starting him? No, not in 2014, just using past examples. He’s pinch hit with Jason Bartlett in the past. His success has come in telling pitchers like David Price, James Shields, Matt Moore, Alex Cobb, etc. go take the mound. A monkey could do that. In fact, look at what he did this year. His team has a +RD, yet he’s managed the team to a <.500 record. Great job!!!!

Pirates Hurdles
Guest
Pirates Hurdles
1 year 9 months ago

Isn’t this the same defense that was used for TB and Oak spending big money on RP this past winter?

Its probably best said that even the best run teams do some funny looking things at times, interpreting or misinterpreting nonpublic data.

LaLoosh
Guest
1 year 10 months ago

have any of those recorded as sac bunts actually been bunts put on into a shifted defense?

Steven
Guest
Steven
1 year 10 months ago

Yeah, I was going to ask how bunts are differentiated between trying to get a hit and trying to make an out.

Bats Left, Throws Right
Member
Member
Bats Left, Throws Right
1 year 10 months ago

Official scorers give sacrifice bunts when the batter bunts and gets out (or reaches on error/FC) but a runner advances regardless of perceived intent.

LaLoosh
Guest
1 year 10 months ago

not sure about that. I think it’s always been a scorer’s discretion to call sac bunt or bunt for hit.

Bats Left, Throws Right
Member
Member
Bats Left, Throws Right
1 year 9 months ago

I have been an official scorer. In practice, any time the batter bunts and gets out when a runner advances a sac hit will be awarded.

Nathaniel Dawson
Guest
Nathaniel Dawson
1 year 10 months ago

Teams shift far less often when there are runners on than with the bases empty. So it’s unlikely that very many of those attempts took place against a shift.

Spit Ball
Guest
Spit Ball
1 year 10 months ago

Plenty of teams still shift when their is a runner on first. When their is a runner at 2nd, the shape of the shift is often changed. Yes their is a discrepancy and although I do not know, I’m guessing that many (although less) sacrifice bunts have been attempted into the shift. Perhaps it has something to do with Joe Maddon’s increased bunt attempts now that I think about it.

hbar
Guest
hbar
1 year 10 months ago

what defines a successful sac bunt in this study? Moving the runner(s) up, or simply getting the bunt down? Are you counting missed or foul bunts (with less than 2 strikes) failed attempts?

Justin
Guest
Justin
1 year 10 months ago

Batting a pitcher eighth should never be viewed as “the posterchild of what a future MLB manager might look like”…it was stupid when La Russa did it and its stupid when Maddon does it

Yirmiyahu
Member
1 year 10 months ago

Well, if Jose Molina is catching at an NL park, there’s an argument to be made he’s the 9th best hitter on the team…

Bats Left, Throws Right
Member
Member
Bats Left, Throws Right
1 year 10 months ago

In The Book it is shown that a team could get a small boost by having the pitcher bat eighth. So small that it’s probably not worth confusing people or pissing them off, but when you’re Maddon you can get away with it.

Sparkles Peterson
Guest
Sparkles Peterson
1 year 10 months ago

Possibly stupid compared to a perfectly optimized lineup. When compared to the same lineup with the pitcher batting 9th, models consistently show that it’s superior.

MGL
Guest
MGL
1 year 10 months ago

That is not correct. I did not do the research for the pitcher batting 8th strategy in The Book, but my lineup simulation models have suggested that it is NOT optimal. It probably depends on who the #8/9 batter is, as well as the top of the order.

Bats Left, Throws Right
Member
Member
Bats Left, Throws Right
1 year 9 months ago

What’s the source of the discrepancy?

Spit Ball
Guest
Spit Ball
1 year 9 months ago

Citation? Documentation? Do tell.

Bobby Bonilla
Guest
Bobby Bonilla
1 year 9 months ago

Gonna go ahead and say that your models suck.

Mike
Guest
Mike
1 year 10 months ago

Does your analysis consider the squeeze bunts the Rays typically lay down any differently than simply a sacrifice moving a guy up to second or over to third? The Rays love the 1st and 3rd with 1 out squeeze play and bunt a ton in that situation, I would bet more often than anyone else in the league (and maybe more than everyone else combined). My mostly uniformed guess, based upon just watching most of the games, would be that a large percentage of the Rays “sacrifice” bunts are of this variety. The Rays have a lot of hitters who strike out a lot and hit a lot of ground balls (e.g. Guyer, Brandon; Jennings, Desmond), so they figure the odds of scoring are better by laying down a bunt than they are trying to swing away and facing the possibility of either a k or a double play. No excuse for being so bad at it though.

Quinn
Guest
Quinn
1 year 10 months ago

This is exactly what I was going to say. Based on what I’ve seen it is likely the first and third push bunt up the first base line is the reason for this number to be so high.

nomoredevil
Guest
nomoredevil
1 year 10 months ago

It’s also probably a big reason for high number of bunts they don’t get down. The area the hitter is attempting to push the ball is much smaller than usual.

AK7007
Member
AK7007
1 year 10 months ago

Is that a play that is going to lead to positive WPA added often? Lemme go look at the run expectancy tables…

Hmm, RE tables say that on average, with runners on first and third, and nobody out, teams on average score 1.8 runs by the end of the inning, and 1.2 with one out. So, a successful squeeze would put you with a runner on second, one run scored, and either one or two outs.

Those situations see teams score .7 and .35 runs, respectively. One run has already scored, so you went from expecting to score 1.8 runs down to expecting 1.7 runs in the no outs first and third (negative runs expected with a successful squeeze), and went from expecting 1.2 runs to 1.3 runs when you get a successful squeeze with one out. (positive runs expected)

That’s with a successful squeeze mind you – failure to execute looks terrible, a free out and putting yourself in a poor base/out state.

So, seems pretty close to just letting players swing away if it’s successful, but terrible if it doesn’t. Some situations might see this boosting a team’s WPA, depending on how close the game score is and the strength of the batter – but the lesson is that those situations aren’t going to be all that likely.

Mike
Guest
Mike
1 year 10 months ago

Can you calculate wpa with individual hitter’s characteristics considered? If not, I’m not sure how it proves the decision is right or wrong. For example, if the batter does not strike out much and hits a lot of balls in the air, bunting is almost certainly the wrong decision. If the hitter strikes out in 20% of plate appearances and hits the ball on the ground 40+% of the time he puts it in play, i’m not convinced the decision is wrong.

AK7007
Member
AK7007
1 year 10 months ago

I’m pretty sure its possible, but more complicated/makes very little difference in the grand scheme of things (fractions of a percentage point difference between an average batter and your hypothetical batter).

Mike
Guest
Mike
1 year 9 months ago

I would be willing to bet the Rays have run the numbers and concluded that for certain hitters against certain pitchers, bunting in the first and third, one out, situation provides a better run expectancy. The more I think about it, the more it makes sense. Take Jennings and assume he swings away (using rounded career numbers):
20% of the time he will strike out;
9% of the time he will walk;
71% of the time he will put the ball in play.

Breaking down the BB data on balls in play (rounded):
13% of the time he will hit a line drive (LD% x 71%);
33% of the time he will hit a ground ball;
26% of the time he will hit an fly ball;

breaking down the fb%, he will hit an infield fly about 3% of the time (26% x 12.6%).

So if he swings away, there is about a 56% chance he strikes out, hits a ground ball, or hits an infield fly. Most of that 56% will result in an out, and some will result in 2 outs and inning over with no runs in. If he bunts, the odds of scoring at least one run would seem to be higher, even if he only converts the sacrifice 69% of the time. I don’t know if your calculations consider bunts that result in base hits as sacrifices in determining your %, but if not the actual success rate is much higher. Obviously the numbers would need to be fleshed out much further to reach any sort of definitive conclusion, but I would be shocked if the Rays haven’t done so and concluded that the bunt play is the correct decision, at least when certain hitters are at the plate against certain pitchers with runners on 1st and 3rd and 1 out. Its also not clear if a hitter attempting to bunt changes the chances he walks; if not, the bunt may create even more expected value.

walt526
Guest
walt526
1 year 10 months ago

When I started reading the article, I assumed that Molina would have accounted for far more than just a half dozen of the attempts. When you have a position player who is as terrible a hitter as Molina, sacrificing is probably the smart play–even if it’s just a 50% success rate.

I would imagine that about half of Jennings and Guyer’s sac bunt attempts were actually bunt for hits with runners on base that didn’t work out. They’re both league average hitters with speed. They’re still doing it too much given the results, but it’s probably not a straight up sacrifice.

Whether a sac bunt is appropriate (or at least defensible) is incredibly context dependent. Average WPA is too blunt a measure, IMHO.

Maddoning
Guest
Maddoning
1 year 10 months ago

Since it’s Maddon, my first assumption would have been that, since most teams have now jumped into shifting on defense, that Maddon would typically be ahead of the adjustment by having his players bunt heavily against the shift.

Not sure if that’s the nature of the bunts or not – anyone know the bunts vs. the shift on those plays? If that’s it, maybe it’s just that hisplayers aren’t doing a good job of it (i.e. still bunting softly so the pitcher/catcher can still field it vs. bunting hard/pushing the ball toward 3B if left uncovered by the shift, etc.).

Might be worth more research?

Jacob Eisenberg
Member
Jacob Eisenberg
1 year 10 months ago

Maybe looking at this from a game theory perspective would help explain both the league-wide trend in WPA and the surprising tendencies of Maddon/Francona. In the chapter on the use of the bunt in ‘The Book’ they showed how important the positioning of the defense is when it comes to bunting, and that sometimes you should bunt to keep the defense honest. The fact that league-wide bunting has decreases may suggest that more teams are accepting or at least aware of the sabermetric belief that bunting is generally a bad choice. And expecting their opponents to do the same, they may not position themselves as aggressively on defense to defend the bunt. Presumably, this effect would be most pronounced on two of the most ‘anti-bunting’ managers in the game, leading Maddon and Francona to take advantage of the fact that teams aren’t playing them to bunt.

AK7007
Member
AK7007
1 year 10 months ago

Smart thought, but then why have the results been so crappy?

Maddoning
Guest
Maddoning
1 year 10 months ago

There, you’d have to look at the individual bunts. I can see a LH hitter being shifted against and just not pushing a bunt down the empty third base line hard enough, or enough to get past the pitcher at least.

That could just be a learning-curve phenomenon, i.e. with shifting in its relative infancy, hitters/bunters haven’t quite adapted their bunting to a “harder” style (whereas in typical sac bunting, you don’t want to push the ball hard too hard ot a fielding position, as it increases the chances for a double play).

In anti-shift bunting, you’d typically want to bunt hard enough to take both the catcher AND pitcher out of it, and get it to the position where the fielder would typically be, but has now shifted out of that position.

Jacob Eisenberg
Member
Jacob Eisenberg
1 year 10 months ago

Good point, that certainly doesn’t help my argument. However the sample size is still small, and maybe Maddon’s players aren’t used to bunting because he hasn’t asked them to bunt much in the past. Also, the organization probably doesn’t emphasize bunting as a skill at the developmental level or in terms of which players they try to bring to the team.

AK7007
Member
AK7007
1 year 10 months ago

There’s probably a fair amount of truth to this. See: any of Jeff’s articles about how bunting to beat the shift would be really smart, but turns out to be really hard too.

Eric Feczko
Guest
Eric Feczko
1 year 10 months ago

I think your OP is an interesting suggestion, but your response to the point contradicts the initial hypothesis.

If the players that are bunting are not skilled in its execution, then bunting remains a poor option, even when the opponents are unprepared. Therefore, from a game theory perspective, one should continue to not bunt.

The simplest explanation for the phenomenon is that Maddon doesn’t always do the things he critiques, just like other “sabr” inclined managers. It’s probably a lot harder to make analytic decisions when one is in managing during the game.

That Guy
Guest
That Guy
1 year 10 months ago

That’s not what “begs the question” means.

Dan
Guest
Dan
1 year 9 months ago

It does now, because most people use it that way. Right or wrong, that is how language works.

Bobby Bonilla
Guest
Bobby Bonilla
1 year 9 months ago

Most people are idiots.

some guy
Guest
some guy
1 year 10 months ago

Tippett found back around 2004 that when you account for ROE, etc. bunting isn’t per se a bad play. It’s only the static “runner on first no outs vs. runner on second one out” analysis that makes bunting seem completely irrational.

HawaiiFO
Guest
1 year 10 months ago

Run scoring environment seems like it could swing things a bit. Should you normally bunt against an ace in the playoffs but against a #2-5 starter you should normally not?

Joe Strummer
Guest
Joe Strummer
1 year 10 months ago

Well, I believe in this and it’s been tested by research: He who f*cks nuns will later join the church.

Jason B
Guest
Jason B
1 year 9 months ago

lolwut

LHPSU
Guest
LHPSU
1 year 10 months ago

I protest Jose Molina’s inclusion in the “non-pitcher” category. He takes his craft of working with pitchers so seriously that he emulates their hitting as well.

Eminor3rd
Guest
Eminor3rd
1 year 10 months ago

This does NOT take into account how bad the hitters are though, right? Or does it?

RaysDaddy
Guest
RaysDaddy
1 year 10 months ago

I suspect most of those were players attempting a bunt for a hit.

Jon L.
Guest
Jon L.
1 year 10 months ago

I don’t think WPA is taking the quality of the hitter into account. It could be that the sacrifice bunt has a better payoff structure than swinging away when it’s a bad matchup, such as Jose Molina against a Major League pitcher, and/or a fast runner, such as Desmond Jennings against a lefty.

Turk's Teeth
Guest
Turk's Teeth
1 year 10 months ago

There are two other outcomes possible from a bunting strategy besides a “productive” or “unproductive” out. One is a hit on the bunt attempt. Another is the bunting batter reaching on an error (often induced by a surprise bunt and/or speedy runner).

Let’s take the simplest outcome: bunting for a hit. Some players are actually pretty good at it. Say, Erick Aybar on the Angels, who has a 36.5% career success rate on *bunt hits* per bunt attempt, and is at a 47.4% success rate this year. If you’re talking about low-power guys who have strong bunt hit success rates, this might be an activity you want to encourage against certain pitchers and certain defensive alignments.

A bunt hit is just an infield single by another name. And obviously, a successful sac bunt has more value than a strikeout or infield pop-up that moves the runner(s) not-an-inch.

So it seems any analysis of bunting value has to look at full-credit bunt hits + partial credit sac bunts over all bunt attempts. Some players are more successful at bunting for a single than swinging for one.

Jennings of the Rays, for example, has 20 bunt attempts this season. Nine of those resulted in a sac bunt that advanced the runners. But five of those resulted in hits. So you’re really talking about 14 of 20 successful outcomes, for some definition of success.

This assumes that the ‘BU’ metric on FanGraphs is inclusive of all bunt attempts, regardless of outcome. (This could be a bad assumption, which could negate the Jennings example, but the general point of qualitatively different outcomes on bunt attempts holds nonetheless. It ain’t all outs.)

Eric
Guest
Eric
1 year 10 months ago

what do the bunting habits of the Moneyball A’s look like? Fewest in the league, with just 12. Part of that is due in part to their league-worst 44% success rate, but they’ve also attempted just 24, the sixth-fewest in the MLB.

Maybe I’m confused about the definitions going into this, but how does 12 successful bunts out of 24 attempts give 44%?

B N
Guest
B N
1 year 10 months ago

Maybe Orioles Magic stole one of them?

MGL
Guest
MGL
1 year 10 months ago

Right, what is the definition of a “successful” attempt? Does that include hits and ROE? The key to a successful attempt is getting lots of hits and ROE and minimizing the outs with no base runner advance. All of these things need to be looked at. Are you including bunt attempts that go to 2 strikes (and thus don’t end well even when the batter swings away with 2 strikes)? Those muct be included as well. IOW, you must include:

1) batter starts bunting, goes to 2 strikes and then swings away – all the results that ensue, which will generally be a far below average PA of course.

2) DP.

3) Out, no runner advance.

4) Out, runner advance.

5) Hits and ROE on a bunt.

6) Walks when attempting to bunt, before 2 strikes.

Unless all of those things are included in the WPA or any kind of analysis, it is impossible to say whether a group of bunts were “correct” or not.

And, as someone pointed out, the results in 58 attempts are going to be very noisy. Even given the results, it is impossible to look at 58 attempts and conclude that they were warranted or not (as a whole).

Tim
Guest
Tim
1 year 10 months ago

For 1) we should presumably use the run expectancy from the count rather than the actual results. Actually, we should probably just look at all missed bunt attempts and assign them the run expectancy of a strike.

MGL
Guest
MGL
1 year 9 months ago

You could do that.

B N
Guest
B N
1 year 10 months ago

I am also quite confused about what this sample size says. We know the following to be true:

1. The league, in general, had a 71% success rate
2. The Rays had a rate only around 60%
3. The sample size was 58 AB

We can interpret this one of a few ways:
1. The true-talent Rays success rate is actually 60%, in which case they should stop all that stupid bunting business.
2. The true-talent success rate of the Ray’s should probably be vastly regressed towards the league average, given that we’re dealing with a tiny sample size. We then need to check if, given league-average success, the Rays should be bunting or not.

The second case is probably much more true, and I am kind of surprised not to see it. If somebody did this kind of analysis with a single player’s first 58 AB and made claims that they should be swinging/not-swinging based on the outcomes, we’d certainly be regressing-to-mean the heck out of those results.

Helladecimal
Guest
Helladecimal
1 year 10 months ago

Maybe get granular and look at defensive alignment tendencies of opposing teams when Maddon calls for a bunt. He should understand shifts well, after all.

Mr baseball
Guest
Mr baseball
1 year 10 months ago

You have to look into game context to truly determine if a team is bunting too much.

tz
Guest
tz
1 year 9 months ago

Including specific adjustments for the next few hitters scheduled to bat.

It was one thing to take the bat out of Longoria’s or Zobrist’s hands when they were hitting great. But for much of the first half this year, neither guy was even league average (Longoria’s still not himself). Maybe Maddon has been trying to take pressure off his struggling stars by working to manufacture more runs than usual.

Bradley Woodrum
Member
Member
1 year 9 months ago

As someone who’s seen just about every Rays game this season and knows the team well, I can say they’ve attempted about only 8 or 9 sac bunts this year. That’s 6 Molina sac attempts, and maybe one or two other attempts.

That list of players who’ve attempted bunts are the five fastest players on the team, plus Jose Molina (the slowest runner and worst hitter). Just about any time those guys have dropped a bunt it has been for an attempted hit or a squeeze play. The team rarely attempts a basic sac attempt.

I think that makes sense, too, to try to bunt hit more often. As bunting becomes more rare, I think we’ll see more Matt Garza-type fielding pitchers. The sort who guarantee either a hit or a RBOE.

AK7007
Member
AK7007
1 year 9 months ago

So, should there be a minimum .gif quota for each article? Would show this phenomena. Jeff would approve!

Mike Green
Guest
Mike Green
1 year 9 months ago

Is every bunt with a runner on base a “sacrifice bunt”, for this purpose? Which players have significantly positive WPA? My guess would be that Anthony Gose leads the league.

Mike Green
Guest
Mike Green
1 year 9 months ago

Ack. I missed Bradley Woodrum’s incisive comment, which goes to my first point.

Udntknowhatuthink
Guest
Udntknowhatuthink
1 year 9 months ago

when is he bunting? what situations does he bunt? is he bunting with runners on 1st and third with one out? this info could of been useful

RichW
Member
RichW
1 year 9 months ago

It’s somewhat misleading to compare sacrifice data to MLB averages. The number of attempts in the AL and the success rate are quite different than the NL. That being said TBR’s SAC attempts are close to the NL average. Interesting to note that TBR are 19-14 in the 33 games where they had successful sacrifices. Also interesting that team records in games with 1 or more sacrifices in 2014 are 575-383 or 60% wins! League wide team records over the last 6 seasons with one or more sacrifices are 60%+ wins. Lousy teams like Arizona are 23-16 in such games this year. Why is that?

MGL
Guest
MGL
1 year 9 months ago

That is because teams tend to sacrifice when they are ahead by 1, 2 or 3 runs and never when they are behind by several ones (more runs than base runners). It is also because when teams sacrifice, it is often in tie games. When you have a runner on first or second in a tie game with no outs, you are a favorite to win whether you sac or not.

So you have an obvious and classic example of correlation with absolutely no causation…

RichW
Member
RichW
1 year 9 months ago

Thanks for answering my question.

passerby
Guest
passerby
1 year 9 months ago

Hypothetical (and imaginative) proposition: imagine you are ahead by 1 run. It would be really nice to be up by 2. You’d get a lot more relative value out of going from a 1 to 2 run read than you will out of 2 to 3, 3 to 4, etc. You look at the odds, you think that risk (don’t score any, lose an out, still ahead) vs reward (gain 1 additional run to pad your lead) makes more sense day in, day out than waiting for that extra-base hit. Additionally, by picking your batter/pitcher combinations, you feel like you are gaining additional edge as well.

My primary question is this: would such a strategy (sac bunting to improve your lead from 1 to 2 runs) be able to be differentiated from an essentially different strategy (say, sac bunting when down by 1 run only) by evaluation of WPA alone? If not, how could we do it? These strategies are meaningfully different in terms of risk/reward; how could one go about evaluating which one is “better”? Furthermore, how could we further analyze Joe Madden and Terry Francona’s decision making to identify the presence of some level of context-specific strategy?

I anticipate that some will reply along the lines of “WPA is WPA, whether are ahead or behind.” And I am willing to consider (and perhaps even admit!) that this is the final answer. If this is your view, could you try to show me what you think I am missing?
I

passerby
Guest
passerby
1 year 9 months ago

note that my description above implies that a sac bunt automatically leads to a run scoring; this is of course an oversimplification, it simply increases the odds of scoring 1 while decreasing the odds of scoring 2+….I feel this is intuitive (to perusers of these comments, at least), and does not affect the underlying illustration.

angelo
Guest
angelo
1 year 9 months ago

Okay, the larger scale effect noted is that bunting is getting rarer and more successful; given that knowledge, what types of bunt situations are getting more or less common?

wpDiscuz