The Royals Bunt to Not Win, Succeed

The main story of the last night’s installment of the Battle for Grass Creek between the Royals and the Mariners was Hisashi Iwakuma. Iwakuma shut down the (admittedly less-than-intimidating) Kansas City bats with an eight-inning effort, during which he allowed only four hits, no walks and no runs, and struck out seven. The Mariners needed all Iwakuma could bring, because the Royals themselves only allowed the Mariners one run. The Mariners’ lone run came right after Royals’ manager Ned Yost made the questionable decision to have left-handed starter Danny Duffy walk the left-handed hitting Robinson Cano to pitch to right-handed Corey Hart. The problems with that decision have been discussed elsewhere. My own short summary: some intentional walks might make sense, but this was not one of them.

Yost made another interesting decision in the bottom of the ninth. With none out, a runner on first and the Mariners’ closer, Fernando Rodney, on the mound, Yost had Norichika Aoki lay down a sacrifice bunt. Aoki did so successfully, but after an Eric Hosmer walk, the Royals made two more outs and it all came to naught. After the game, Yost explained his decision:

Because I want to take a shot at tying it. My ‘pen was strong enough where I felt like I could go ahead and go for the tie. Some nights you don’t. Some nights you play for the win.

Like intentional walks, not all bunts are bad. Sometimes they are the smart play, sometimes they are not. It is not always easy to say one way or the other. Yost’s teams have sometimes bunted in situations where it made sense. Was this one of those situations?

Just about any managerial decision can be subject to analysis from a myriad of angles, and involve many things that cannot be covered in a blog post, even without considering information to which we are not privy — and said information does exist. Of course, this is also true of many things in the world about which we laypeople have opinions, and that does not mean we cannot have and express such opinions. But it must be admitted that we probably do not know everything.

Still, Yost laid out his thinking in a pretty clear manner: he was going for the tie. It is easy and fun to pick on the “Some nights you play for the win” line, but we know what Yost was really trying to say: get the tie in ninth with one run, then give the Royals’ good bullpen a chance to hold the Mariners down while the offense tries to scratch out the win in extras.

Saber-friendly analysts do think there are appropriate times for sacrifice bunts, and in close and late situations one-run strategies are sometimes the right call. This would seem to potentially be one of those situations. Without getting into every angle ourselves, this is a situation in which Win Probability Added (WPA) might help us out. Unfortunately, according to WPA the bunt did not increase the Royals chances of winning, the WPA for the play was -.049.

Now, WPA does not take into account the skills of the individual players involved. It is worth looking more closely to see if it makes a difference. The most relevant player here is Aoki, naturally. Is Aoki a particularly bad hitter, such that he would be better off bunting? Aoki has been on something of a cold streak lately, and (after last night) is hitting just .277/.331/.346 (89 wRC+) on the year. Even if that is poor, his on-base percentage is still above average. More importantly, we know that observed performance is not the same thing as true talent, and for his career Aoki is an above-average hitter (107 wRC+). While ZiPS projects his current true talent as 93 wRC+, Steamer has him as a 110 wRC+ hitter. It is fair to say he is probably about average.

Aoki, a left-handed hitter, also had the platoon advantage versus the right-handed Rodney. Now, some will point out that for career in the North American major leagues, Aoki has a slight reverse platoon split, with a 110 wRC+ versus lefties but a 105 wRC+ versus righties. That is a pretty small split, though, and given the sample sizes of major league plate appearance needed to estimate platoon skill, we are a long way from being able to say that Aoki has enough major league plate apperances versus lefties (he has 430, and at 1000 we regress halfway to league average). As for Rodney, for most of his career he has, as one would expect, fared a bit worse versus left-handed batters. At least on this broad view, platoon consideration do not really help Yost’s case for a sacrifice bunt.

A good case can be made, however, that a bunt might be defensible if it is not just a sacrifice bunt. Aoki might be a decent hitter, but he is clearly no Giancarlo Stanton. Hitting one out was probably not going to happen. What Aoki has shown himself to be good at is bunting for hits. Since he came into the league in 2012, Aoki is third in the majors with bunts for hits with 19. Almost 37 percent of this bunts have gone for hits. Depending on the position of the fielders, bunting for a hit might not be a bad idea.

Aoki’s skill at bunting for a hit is is not relevant to this discussion of tactics, though. Aoki squared up early as if sacrifice bunting, and Yost himself did not suggest that it was anything other than a called sacrifice bunt.

It is impossible to get into every little factor, but a couple more are worth briefly mentioning. Neither make the decision to sacrifice look any better. For one, Aoki is pretty fast, and the numbers suggest that he is less likely to ground into a double play than the average hitter, so that was not a reason to sacrifice (as it might sometimes be).

Finally, it is not as if Rodney looked especially unbeatable. Indeed, before Aoki came to the plate, Rodney walked Alcides Escobar on four pitches. I was not in the Mariners’ dugout, but I would guess there was little talk about not allowing Alcides Escobar to beat them. Rodney’s first pitch to Aoki was a ball. Rodney was clearly not doing well with control.

The Royals very probably would have lost whether or not Aoki had bunted. And they might have won after he sacrificed. Maybe if he had swung away, he would have grounded into a double play. Maybe Billy Butler or Salvador Perez would have gotten a hit after Hosmer walked and tied the game or even put the Royals ahead. “Can’t predict baseball” is as obvious as it is trite. Overall, though, the decision to have Aoki sacrifice bunt with one on and no one out seems to have hurt the Royals chances, and a deeper examination does not really help Yost’s case.




Print This Post



Matt Klaassen reads and writes obituaries in the Greater Toronto Area. If you can't get enough of him, follow him on Twitter.


38 Responses to “The Royals Bunt to Not Win, Succeed”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. Robo says:

    Yosted! I don’t see them on TV often so most of my Royals info comes from checking articles and game recaps while doing fantasy things/killing time. Any Royals fans care to comment whether Ned Yost is as bad day in and day out as his reputation? Because most indications I have are that he’s pretty terrible.

    Vote -1 Vote +1

    • Crono says:

      He’s worse than your wildest nightmares.

      Vote -1 Vote +1

    • KCDaveInLA says:

      Yup. The Royals are to Fangraphs what the OJ trial, Monica Lewinsky, and Lorena Bobbitt were to Jay Leno – and Ned Yost is definitely not helping that.

      Vote -1 Vote +1

    • Careless says:

      As a Cubs fan, I miss having him in the division.

      Vote -1 Vote +1

      • SleepNowInTheFire says:

        And yet somehow the Cubs always seemed to finish behind the Yost-led Brewers.

        A fair point though. Yost was BAD. And as a Brewers fan, I’ll say this: it appears, judging by minor league talent, that the Cubs won’t suck that much longer. And the Brewers? Well, they better start drafting and developing better if they want a half-decent future. Their farm system is a wasteland.

        Vote -1 Vote +1

    • Pedro The Sea Lion says:

      I think that his reputation means that anything Yost does to validate that reputation will especially stand out.
      Our ability to measure Yost’s strategic decisions does indicate that those decisions are often not value added, but I don’t believe suggest that he is significantly below replacement level as a manger. (replacement level in this case involving bad decision-making) His greatest asset, admittedly more subjective, is his ability to manage his players in a professional sense. This is not measurable and less visible than what is done on the field and what is said in press conferences, but is widely believed to be true from what I have read by various reporters and people in the industry. In my opinion, Yost doesn’t make strategic mistakes that his predecessor probably wouldn’t, and his ability to manage people from a business standpoint probably makes him an average manager in terms of overall value to the organization.

      Vote -1 Vote +1

  2. Ruki Motomiya says:

    I’ve always wanted to see teams, with a man on first and a fast hitter/one good at bunting for hits, have their guy try to lay one down for a hit with the sacrifice bunt essentially being a consolation for if your Aoki or whatnot doesn’t make it on base. Not what happened here, but it is an idea I like.

    Vote -1 Vote +1

  3. Kazinski says:

    Where is the other half of the article? I don’t see how you can talk about the events in the 9th without bring up how the Mariners scored their only run in the 3rd here is the play log with WPA, and RE24 added:
    Mike Zunino doubled to center (Fly) .070 0.61
    Michael Saunders sacrificed to pitcher (Bunt Grounder). Mike Zunino advanced to 3B. -.013 -0.16
    Stefen Romero struck out swinging. -.069 -0.57
    Robinson Cano was intentionally walked. .013 0.13
    Corey Hart singled to center (Liner). Mike Zunino scored. Robinson Cano advanced to 3B. .122 1.00

    Now admittedly there is a very good chance that Zunino would have scored on Hart’s single from second anyway. So maybe Saunders sacrifice didn’t help, but who knows, Zunino is not the swiftest of runners and it certainly would not have been a lock. It looks like Saunders was bunting for a base hit too, the play was very close and he’s already got 3 BUH this season, and a 42.9 BUH% for his career, so attempting a bunt doesn’t actually lower his odds of getting on base.

    But it certainly would have made one point about sacrifice bunts: you win some, you lose some. The Mariners scored their only run with a sacrifice bunt in the mix and won the game. The Royals blew their chance to tie in the ninth with a bunt, and lost.

    Vote -1 Vote +1

    • Aaron says:

      “If you play for one run, that’s all you’ll get.” -Earl Weaver

      Vote -1 Vote +1

    • Matt says:

      There is also a difference between bunting a man from second to third and bunting a mam from first to second. With a man on third it opens up the ability to score on a sacrifice or a ball getting away from the catcher which you can’t score on from second.

      Vote -1 Vote +1

    • KDL says:

      “Like intentional walks, not all bunts are bad.”
      -the article you didn’t read very closely

      Vote -1 Vote +1

    • ColonelTom says:

      That’s a very special inning, managerially speaking. The Mariners reduced their run expectancy with Saunders’s sacrifice, and the Royals gave it right back by walking Cano with two out in the third freakin’ inning. Brilliant.

      Vote -1 Vote +1

  4. Richie says:

    Saunders bunted for a hit, and wound up with the consolation prize. Aoki squared early, so pretty much no chance of reaching 1st safely. This is all clear from the article, which you give the impression of only skimming after you once got the ‘Waitaminute!’ idea in your head.

    Vote -1 Vote +1

    • Kazinski says:

      Did I miss something? Even with my limited reading comprehension I still don’t see the words “Saunders” or “Zunino” in the article, so I don’t think its “all clear”. And while I allow Saunders was bunting for a hit, even though it was scored a sacrifice, it was with the idea of getting Zunino over to 3rd whether Saunders got to first safe or not. Besides if you evaluate both bunts by the results they both ended up with an out and advanced the runner. One bunt was part of the winning run sequence. The other was not, but according to Tango, the Royals chance of scoring at least a run at any point in the inning were never higher than 42.9%.

      Don’t get me wrong, I am not a big fan of the sacrifice bunt, especially in the early innings. I just think it should be noted, it does work often enough it shouldn’t be totally scoffed at because the differences in outcome expectantcies are often not so stark that unique circumstances like whos on the mound, who’s coming up next, how the team has been playing, lighting conditions etc, have a enough of an impact to erase the narrow percentage advantage of letting a player swing away.

      Vote -1 Vote +1

      • Wally says:

        I think if you were actually pay attention to Tango’s points on bunting, clearly turing around to give away all chance of bunting for a hit is pretty much never a good idea. In some situations if you try to bunt for a hit and end up with a sacrifice great, but that’s not what happened, and I think that’s a big part of the point here.

        Vote -1 Vote +1

        • Kazinski says:

          The reference to Tango is not on his bunting work, but on chances of scoring at least one run on the RE24 charts. RE24 is usually expressed in terms of run expectancy, and of course that makes sense because the more runs the better. But there are some situations where the 2nd and 3rd runs are much less valuable than the first run. And it maybe that Yost considered this one of those times.

          The point I was trying to make though was looking at the chance that a run will score at some point in the inning and tie the game in that situation, the Royals never had the odds in their favor at anytime in the inning. The odds were always in the Mariners favor to hold on to the lead and win. The bunt by Aoki was marginal, and it marginally hurt the Royals chances. But its not like according to WPA and RE24 the Royals had a golden opportunity and Yost’s bad managing blew the whole thing up.

          Vote -1 Vote +1

        • Wally says:

          “The bunt by Aoki was marginal, and it marginally hurt the Royals chances. But its not like according to WPA and RE24 the Royals had a golden opportunity and Yost’s bad managing blew the whole thing up.”

          The flaw there is you have to resort to the extreme, “blew the whole thing up,” to have a case. Of course it didn’t take the chances from 42% to 0%, but it wasn’t a good move, especially when giving up on the possibility of bunting for a base hit by not squaring around so early. Tango does describe how with no out and one on moving a guy to second in exchange for an out can increase your chances to score at least one run, but its close enough that things like quality of hitter, double play chances with the given hitter, platoon split, squaring around early can easily take it from a slight positive to a clear negative. The point being made here, is that this seems to be case where it was negative.

          I’ve seen worse, probably very recently if I thought about it. Probably without walking Cano in the 3rd inning this isn’t even discussed.

          Vote -1 Vote +1

      • Schuxu says:

        This article is about the bad Royals descicion to bunt in the 9th inning. What does the Mariners bunt that happened earlier to do with it?
        Or are you suggesting that the Mariners bunt influenced Yost descicion to let Aoki bunt? That would be even worse on Yost’ side.

        Vote -1 Vote +1

  5. jim S. says:

    Rodney’s best pitch is his change, which breaks sharply away and down from lefties and would be tough on Aoki. You can’t use career splits because his control improved greatly when he came to TB. In 2012 lefties batted .166 off him and last year they batted .248. Give Yost a break. He might have been right.

    Vote -1 Vote +1

  6. shthar says:

    Studying what the royals do, or why they royals do it, is just wasting your time.

    Just put it under the ‘what not to do’ category and move on.

    Vote -1 Vote +1

  7. MGL says:

    “Aoki’s skill at bunting for a hit is is not relevant to this discussion of tactics, though.”

    That is 100% wrong. His skill at bunting for a hit and his speed are critical to this discussion.

    I may sound like a broken record, but any analysis or evaluation of a sacrifice bunt attempt should have almost nothing to do with whether the WE increases or decreases after an out and a base runner advance. Now, that might be fine if that is what happens when a manager orders a sacrifice attempt (the base runner is simply given the next base and the batter is declared out), but, that is not nearly what happens. Sometimes the batter is safe on a hit or error, sometimes he bunts into a force, sometimes he ends up striking out, walking or hitting a home run after failing to get the bunt down within the first 2 strikes, etc.

    The reason the batter’s bunting skill and speed are critical to the analysis is that the better and faster the bunter, the more he ends up safe on a hit or an error, even when he is clearly bunting for a sacrifice. And there is a break even point at which the sacrifice bunt attempt is better than swinging away. And that break even point involves how often the batter is safe on a bunt attempt, how often he advances the runner with than out, and how often he does not advance the runner with an out (as well as a myriad of results when he is forced to swing away after 2 strikes.

    Can we declare a permanent moratorium or injunction against any sacrifice bunt “analysis” that assumes that the only result is an out and runner advance? This is not the 1990’s. That kind of analysis is 100% worthless. That is not even close to what happens when a batter attempts a sacrifice. And what does actually happen varies considerably depending upon the bunting skill and speed of the batter, such that a fast batter who is a good bunter can easily increase his team’s WE by attempting a bunt, especially if he is not a great hitter when swinging away, and especially against a tough pitcher, although tough pitchers can also be tough to bunt against.

    Here is an example of what happens when a batter attempts a bunt in the late innings of a close game when the defense is expecting a bunt:

    Batter out, runner advances: 54%
    Batter out, no runner advance: 10%
    Force out: 9%
    DP: 4%
    Batter and runner safe on a hit or ROE: 12%
    Walk: 3%
    K: 8%

    As you can see, the assumption of an out and runner advance is ridiculous.

    As far as Matt’s contention that the bunting ability of the batter when he attempts to bunt for a hit does not matter, here is some data that suggests that that is completely wrong:

    Bunting proficiency of the batter as measured by their number of bunt hit attempts per PA.

    Bad bunters

    Batter out, runner advances: 41%
    Batter out, no runner advance or DP, including K: 33%
    Batter and runner safe, including hits, ROE, and BB: 26%

    Good bunters

    Batter out, runner advances: 42%
    Batter out, no runner advance or DP, including K: 28%
    Batter and runner safe, including hits, ROE, and BB: 29%

    So good bunters make outs with no runner advance 5% less often, and reach base 3% more often. In fact, their total RE is .040 runs better than the poor bunters.

    In addition, fast bunters yield .034 more runs than slow bunters, so being fast is also a good pre-requisite for attempting a sacrifice, for obvious reasons.

    Again, please, no more sacrifice bunt attempt analyses unless you want to talk about all the things that can happen when attempting a bunt, and you can estimate their frequencies based on the speed and bunting ability of the batter and the position of the defense. The latter part, the position of the defense, is a huge determinate of the result of a bunt attempt. When the defense is not particularly attempting a bunt, or is agnostic, there are many more hits and ROE’s and fewer forces on the runner than when the defense is nearly certain that the batter is going to bunt and is crashing hard on the corners.

    Vote -1 Vote +1

    • Richie says:

      Not every Fangraphs post can qualify as a ‘The Book’ chapter, nor should it. You show enough evidence that this was a dumb move by Yost, then move on with your professional life. You certainly don’t “talk about all the things that can happen when attempting a bunt” unless you want to go back to driving the school bus for a living.

      Your requirement would rule out ever writing about a managerial bunt decision on this site given the proper space requirements. Which would be a really lousy business decision. Only thing I can see Matt maybe should’ve added given those constraints is noting at about what depth the infield was playing.

      Vote -1 Vote +1

      • MGL says:

        “You show enough evidence that this was a dumb move by Yost, then move on with your professional life.”

        The entire point of my post, which you are apparently intellectually incapable of understanding, is that Matt has presented zero evidence that the bunt attempt was either good or bad, because in order to do that one needs to consider the speed and bunting proficiency of the batter as well as the position of he defense, and how those things affect the frequencies of the various possible outcomes of the bunt attempt. Without addressing those things there is no way of knowing whether a bunt attempt was a good strategy or not.

        Vote -1 Vote +1

        • Richie says:

          Systemic evidence Matt offers that the bunt there was a poor strategy:

          WPA is lower after a successful sacrifice.

          Situationally specific evidence Matt offers that the bunt there was a poor strategy:

          Aoki squares up.
          Aoki is an above average hitter, average at worst.
          Aoki has the platoon advantage.
          Aoki’s speed makes a double play less likely.
          Rodney’s inability to find the plate on the previous weak hitter.

          Vote -1 Vote +1

        • Richie says:

          To quickly sum up, it’s been demonstrated that bunting is a poor idea other than in certain specific situations. (which is why managers seldom do it) Bunting with your leadoff hitter down a run in the top of the inning? Just about everything situationally specific needs to say ‘bunt’ to overcome the general lousiness of that notion. Matt shows we’re nowhere close to ‘everything’.

          Vote -1 Vote +1

  8. McKingford says:

    I can’t believe all this discussion on bunting completely missed, from the very same game, the McClendon shenanigans. He inserted a pinch hitter in the *7th inning* for the express purpose of then sacrificing a runner to second. I know we talk about how precious outs are, which generally makes sacrificing stupid (you only get 27!). But what’s even more precious than outs are bench bats. You’re going to bring a guy into the game, with almost a third of the game left to play, just so he can deliberately make an out!?

    Vote -1 Vote +1

    • Kazinski says:

      The substitution was not only for a sacrifice. Jones is the fastest man on the team, and was going to come in defensively in the late innings of a close game anyway. Both players were recently called up from the minors and neither has proven MLB ability, so bunting with a replacement level batter is not that bad of an idea there. The move was like having a pitcher bunt in the NL, with the bonus of having a speedier baserunner and getting your best defensive player out on to the field.

      Vote -1 Vote +1

  9. david says:

    the way things were going for billy butler that night, rodney would have walked the bases loaded and he would’ve grounded into a triple play.

    Vote -1 Vote +1

  10. Is joe madden always right? says:

    I am not a huge numbers guy but I obviously know that giving up outs is bad. But I found myself wondering when watching the last game of the orioles and rays series the other night if a bunt would of actually been the best play for the Rays when Jennings came to the plate men on first and second and with no outs in the bottom of the ninth down two. Not that it’s relevant but he hit into a double play. What is relevant is that Darren O’Day was on the mound, a pronounced ground ball pitcher with huge platoon splits. On deck they had Joyce, a leftie, followed by an elite right handed bat, longoria and another leftie in Loney had it gone that far. I wouldn’t know where to start trying to use numbers to find the right move so would be interested in any numbers anyone could give me to support the move or otherwise! Thanks

    Vote -1 Vote +1

  11. Richie says:

    ‘The Book’ is a good starting point. After that, well, you probably just have to crunch the numbers yourself (or find someone else who’s done it for you) in any specific situation. But ‘The Book’ will give you the basis for doing so.

    In the simplest sense, Tarzan can strategize the bunt. “Defense back? Bunt Good. Defense up? Bunt Baaaaaadd!” So the main factor in what you cite would’ve actually been how far up the corners were. Also how good a fielder O’Day is, as pitchers wind up fielding pretty many bunts. Another factor here is that advancing then scoring the second runner only extends the game for you, doesn’t end it in your favor.

    Vote -1 Vote +1

    • Richie says:

      Also in your particular case, there’s more variability. Is the third baseman hanging back some to hopefully force the runner there? Then more room for a bunt. Or is he charging with the shortstop perhaps scurrying over to then cover third? Then lots and lots of holes in the infield. And of course you can’t be sure as to which it’ll be until the pitch is on its way and the batter is committed to one or the other. So you have a game-theoretical guessing game going on too.

      Vote -1 Vote +1

  12. Mr Scout says:

    Data geek orgy. Face it, Yost has more data than you do mr data geek. I’ll vote for more data than using less data.

    Vote -1 Vote +1

  13. Schuxu says:

    The way Rodney was pitching at the beginning of the inning, Aoki just standing there with the bat on his shoulder would have yield the same chance to advance the runner to second (via an Aoki walk) as the bunt.

    Vote -1 Vote +1

  14. waynetolleson says:

    I watched that game, and I second-guessed the bunt immediately. It wasn’t so much the decision of bunting in that situation is always terrible, but Rodney seemed rattled, like he had no idea where the ball was going. It seemed in that situation, giving Rodney an out was really doing him a favor.

    Different closer, or same closer pitching differently, I might have a different opinion. In that situation, the way Rodney was going, he might have walked five straight, of laid a cookie right over the middle of the plate.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>