Examining First-Inning Sacrifice Bunts
The other day on our internal message board, the shockingly handsome David Laurila pointed out that the Reds had dropped down a sacrifice bunt in the first inning the other day, and that this had struck him as strange. In exploring the subject further, I found that the practice has not yet evaporated but this there is some hope that it is lessening this season. Or at least, there would be if not for the San Diego Padres, who piled up five sacrifice hits in the first inning during April.
Let’s break down the five plays, shall we?
| Date | Opp | Batter | Pitcher | Situation | WPA | Run scored? | Gm Res |
|---|---|---|---|---|---|---|---|
| 1-Apr | NYM | Everth Cabrera | Jon Niese | 0-0, no outs, runner on 1st | -0.016 | No | L 2-11 |
| 4-Apr | NYM | Will Venable | Dillon Gee | 0-0, no outs, runner on 1st | -0.016 | No | W 2-1 |
| 15-Apr | LAD | Will Venable | Chad Billingsley | 0-0, no outs, runner on 1st | -0.015 | No | W 6-3 |
| 16-Apr | LAD | Everth Cabrera | Chris Capuano | 0-0, no outs, runner on 1st | -0.015 | Yes | W 9-2 |
| 19-Apr | SF | Everth Cabrera | Madison Bumgarner | 0-0, no outs, runner on 2nd | -0.011 | Yes | L 2-3 |
In each situation, the sacrifice bunt came after the leadoff batter reached base. This is the situation in which we would expect to see a first-inning sacrifice bunt, so that fits. The only other typical early-inning bunting situation would probably be a sac bunt from the pitcher, and since the pitcher doesn’t usually get a chance to hit in the opening frame this is what we would expect. However, just because it is what we would expect to see doesn’t make it proper strategy.
Keith Law’s article this week is the latest to reinforce the notion that a team’s number-two hitter should really be its best hitter. Now, neither Cabrera nor Venable are scrubs offensively, especially when you account for the position they play. But neither is San Diego’s best hitter.
Among Padres position players with at least 20 plate appearances this season, Cabrera’s 117 wRC+ ranks sixth, while the slumping Venable sits 12th out of 14 with a 72 wRC+. Venable is obviously better than that — he has a career 104 wRC+, and has had at least a 99 wRC+ in all five of his major league seasons — but he’s certainly not the team’s best hitter. That would be Chase Headley. Now, he wasn’t available for most of these five games, but then the mantle probably falls to Yonder Alonso or Carlos Quentin.
What the Padres did in April borders on historic. Thanks to some help from the always gracious Jeff Zimmerman, we can see that few teams have laid down as many first-inning sac bunts in any month of the season (dating back to 1974) as the Padres did in April:
| Team | Month, Year | # Sac Bunts |
|---|---|---|
| Angels | May-75 | 6 |
| Braves | Aug-82 | 6 |
| Braves | Jul-74 | 5 |
| Dodgers | Jun-95 | 5 |
| Expos | Aug-75 | 5 |
| Mariners | Jul-80 | 5 |
| Padres | May-79 | 5 |
| Padres | Apr-13 | 5 |
| Pirates | Aug-92 | 5 |
| Rangers | Jun-78 | 5 |
| Tigers | Jun-80 | 5 |
| Twins | May-76 | 5 |
You’ll notice that the most recent team before the Padres here was back in 1995. Teams just don’t bunch together sac bunts in the game’s opening frame like that anymore. Perhaps Black was nervous about the Padres’ ability to score runs with Headley out. Perhaps Black was really apprehensive about the pitchers he was facing.
That could be particularly plausible in the last game against Madison Bumgarner, and the bunt in that situation wasn’t as bad considering there was a runner on second. But that still doesn’t make it the correct decision. In what was also sure to be a close game, it was imperative to try and jump out early on Bumgarner. Especially since Bumgarner has historically been at his worst in the first inning — both his 4.74 ERA and .798 OPS are his worst marks per inning (ninth inning excluded since he only has three career IP in the ninth). Yes, the Padres did plate the run, but they would go on to lose 3-2, and perhaps missed a golden opportunity to jump out to an early two-run lead.
The potential good news is that first-inning sac bunts have dropped precipitously in the early going this season, particularly in the AL. There were only 11 first-inning sac bunts in April — the five from the Padres, and one each by the Giants, Indians, Orioles, Pirates, Rangers and Reds. It’s still a little early to call this a victory against early bunting, especially since the practice has not vanished entirely from the game in the past 20 years. To wit:
After spiking in 2004, a four-year downturn carried through 2008. In the past four seasons however, first-inning sac bunts had ticked back up. At the moment, the total has dropped to its lowest point since 1996, and it might be lower had the Padres not ripped off their near-historic month. It’s still early though, and even if the Padres taper off going forward, other teams may fill the void. But the practice doesn’t seem like it will evaporate entirely. Even in ’96, the other low point of the past 20 years, there were still 56 first-inning sac bunts. In the past decade, the range has been anywhere from 80-129 first-inning sac bunts a season. That is a lot for the so-called Moneyball age, and it is a number that didn’t really fluctuate at all from the decade previous to it.
Now, you’re probably thinking, 100 or so bunts, big deal. And perhaps you’re right. After all, on a percentage basis, sac bunts in the first inning occur less than during any other inning:
| Inning | # Sac Bunts | % Sac Bunts |
|---|---|---|
| 1st inning | 3,756 | 6.0% |
| 2nd inning | 5,758 | 9.3% |
| 3rd inning | 8,372 | 13.5% |
| 4th inning | 5,264 | 8.5% |
| 5th inning | 7,780 | 12.5% |
| 6th inning | 5,780 | 9.3% |
| 7th inning | 7,418 | 11.9% |
| 8th inning | 7,088 | 11.4% |
| 9th inning | 6,754 | 10.9% |
| Extra innings | 4,216 | 6.8% |
But to me, the fact that it is even close to the percent of bunt attempts in extra innings — which is the ultimate time to be employing a single-run strategy — shows that these bunts are occurring far too frequently.
Perhaps this year will be the year where first-inning sac bunts go the way of the dodo. Perhaps the Padres’ April will be an isolated event, and the game will see 50 or fewer sac bunts. But perhaps this year just started slowly, and we will see this number rise to the levels of the past four seasons as we push deeper into summer.
Paul Swydan used to be the managing editor of The Hardball Times, a writer and editor for FanGraphs and a writer for Boston.com and The Boston Globe. Now, he owns The Silver Unicorn Bookstore, an independent bookstore in Acton, Mass. Follow him on Twitter @Swydan. Follow the store @SilUnicornActon.
Does your data set include pitcher bunts? If it doesn’t that was unclear to me. Those really need to be taken out to formulate a proper analysis…
I did not have time for that, as that is a split of a split of a split of a split. As I mentioned, for the purposes of examining the first inning, I don’t think it’s an issue, as the pitcher rarely bats in the first inning.
Granted. I was more curious about the last data table, comparing frequency across later innings.
i’d infer that pitcher bunts are included in the last table, but i’m inferrous by nature.
You should see a doctor, that sounds like a problem with your iron levels.
Interesting data on bunts by inning–I suppose the 3rd inning spikes to 13.5% as it marks the inning most likely to see the 9-spot hitter; the fifth inning spike must be because the 9-spot is coming up for a second time?
Yes, that is how I interpreted it.
The only question I have is how you distinguish sac bunt and failed attempt to bunt for a hit. Cabrera and Venable both were among the leaders in successful bunts for hits last year (10 and 8 respectively) so it is not clear to me that these were just misguided sacrifices. The logic may have been to guarantee advancing the runner while still having a legitimate shot at reaching base safely as well.
I simply used SH’s, and then went back and looked at the games in which those occurred.
I looked back at all the bunts. It was actually rather easy since they were all in the top of the 1st.
I would say that neither of the Veneble sacrifices were real attempts at a hit. He totally squares up his body and offers very early. As a lefty, if he was trying to bunt for a hit, I think he would have just dropped the bat and almost drag bunted a little.
The Cabrera ones are debatable. 4/1 and 4/19 look close enough to hit attempts, but he does square a little earlier than if no one was on IMO. However, I think 4/16 is definitely a base hit attempt.
So, in conclusion, I would say it is between 2 and 4 actual sacrifice bunts without a real intention of a base hit.
Maybe I am missing something, but Cabrera has attempted 9 bunts this year and hit safely on 3 of them. Successful bunt hit % seems like a legitimate part of this conversation.
Given that the WPA changed by 1.6% or less in all situations, I feel like even a relatively small chance (20%) to successfully get a bunt hit or ROE means that a bunt can very easily break even. A bunt hit or a ROE results in a 5% swing upward.
Look at the Oakland/Baltimore game last week; Bob Melvin tried to sacrifice with a runner on 1st, and Strop threw it away. Then he sacrificed with runners on 1st and 2nd, and Machado threw it away and the winning run scored.
That situation is obviously extreme, but yes that is exactly what I am saying.
With guys like this who can reasonably bunt for a hit, it’s not fair to examine any specific bunt (in this base/out/score state) without the video. If they’re squared early and/or the defense is up their butt, it’s bad. If the defense is at a reasonable depth and it’s an attempt for a hit (getting a running start out of the box instead of making sure to put it in play, etc), it’s fine.
You might want to explore the starting pitchers of the bunting teams. Last year, while at a Toronto vs Tampa game, Joe Maddon had his number 2 hitter sac bunt a runner to second base. My brother’s response was “why?”. My response, “David Price is pitching; Maddon wants the team to give Price at least one run.”. Price proceeded with six shutout innings as Tampa took a 9 – 0 lead before eventually winning the game 9 – 4.
Oh, if Joe Maddon does it, it can’t be stupid.
There’s a good argument to be made for a manager not to be consistent in his decisions in close situations. Thus, the other team cannot analyze his past and prepare for bunts, hit and runs, steals, etc. in certain situations, but must guess. Perhaps this is what Joe was doing–throwing in a random bunt.
Looks like as has been mentioned by most of these comments that a lot of the sac bunts were attempts at bunt singles that didn’t work out. Not necessarily a bad strategy with runner on as at least the runner advances.
OTOH hard hit ball right at a player could result in an out and no advancement or even a double play.
As others have said, the proper universe for this sort of study is really “all bunt attempts”. And that has to include squaring to bunt at ball 1 (and missing or fouling it off), then swinging away, because by making that bunt attempt you change your expected BABIP on a groundball (or liner) as the infield comes in.
In fact, even at-bats without a bunt attempt are influenced by your previous attempts – teams will know that Cabrera or Venable might drop the bunt, so the infield may be a step or two closer than they would otherwise.
I recall reading that Jay Bell had 40 or more first inning sac bunts with the Pirates in ’91 or ’92.
I really liked the article, though I’m not sure the “sac bunts in inning one” being close to the “sac bunts in extras” is really comparable. Presumably there have been the same number of 1st through 4th innings, and almost that many of 5th through 8th innings. But how many extra innings (and batting 9th innings) have there been? I think the percentage assignment in that last chart breaks down after 8, at which point it only makes sense to look at frequency.
But that’s a tiny quibble. Otherwise, good stuff.
That’s an important quibble, though. The % of sac bunts in extra innings (and the 9th inning, to a lesser extent) has to be taken in context. Many fewer PAs happen in extras than in regulation, simply because most games don’t go to extra innings.
I do not think Bud Black is an idiot. I really think we are all failing to look at context here.
I did not see any of these games, but I suspect most or all of these may not have sacrifice bunts in the strictest sense. Official scorers are supposed to score bunts as ground outs when the intent of the batter is to reach base himself, but if a runner advances scorers almost always rule the sacrifices. Cabrera and Venable are both extremely fast and excellent bunters. Collectively they are 46-126 on bunt attempts throughout their careers, and Cabrera is 3-9 this year.
Two soft hitting speedsters who are, at worst, just as good at getting on base by bunting than not. Is the small chance they get an extra base hit really more valuable than the extremely high chance they advance the runner even if they are thrown out? I don’t honestly know, but just based on some quick math with the run expectancy matrix it is hard for me to say Black made the wrong call. (Unless we are talking about whether they should have been batting in the two-hole, because that is probably the wrong call).
Sorry, it feels like we’re beating a dead horse here, but I feel like I should explain myself further. My math is rough but makes even with the most conservative estimates it seems like bunting in these situations is very much the right decision, not just a “close” or “not that bad one”.
I went to Washington State vs. Oregon yesterday and WSU attempted to bunt with the second batter of the game, he failed to get it down and singled after the bunt got taken off. Then the 3 hitter also attempted to bunt and couldn’t get it down. He too singled to load the bases.
Oh man, the 1992 Pirates. I’m shocked every season of the Leyland/Jay Bell batting second era wasn’t listed.