Davenport on pitcher control of balls in play
As a sort of follow-up to the Franklin arguments we’ve had here lately (summary: “Pitchers have an extremely limited ability to affect whether a ball is put into play and becomes an out or a hit”), I wanted to bring this up. One of the theories on why major league pitchers differ so little in hit prevention is that they’re selected for this ability as they advance — pitchers who aren’t good at this don’t advance.
Conveniently, Clay Davenport looked at this throughout the minors today, while the argument’s still fresh.
The pitchers who made the major leagues are, not surprisingly, better than their counterparts who did not, by every measure of pitching you may desire–including giving up fewer hits per ball in play.
Pitchers who aren’t any good at getting outs when a batter makes contact (their fastball is straight, or they’re no good at changing speeds — whatever) don’t advance, not only because they get shelled continually, but because as a group, their strikeout, home run, and walk rates aren’t as good.
This doesn’t answer a new question (why is the variation in a pitcher’s ability to prevent hits on balls in play so small at the major league level, while strikeout rate, walk rate, and home run rate have a wide variation) but it’s an interesting addition to the puzzle.
Comments
44 Responses to “Davenport on pitcher control of balls in play”
It seems to me that what’s being said here boils down to something we’ve all known: That soft-tossers (i.e., pitchers less likely to have high K rates and thus have a greater PERCEIVED ability to affect their outcomes) are prejudiced against and their minor-league successes dismissed, while the flamethrowers (perceived to have more abiluity to control the game along with those higher Ks scouts love) are given chance after chance to fail (i.e.. Matt Thornton).
I remember about 15 years ago, Bill James, in one of his newsletters, took up the case of a pitcher at the White Sox’s Triple-A club named Tom Drees … a soft-tosser who threw three no-hitters in 1990 (one, as I recall, came with NO strikeouts, which to me is pretty amazing but, granted, probably says more about his defense than him) but got one brief callup, didn’t do well, and was released. There was no rush to pick up Drees. He never made the majors again, and his career drifted away a few years later. James never said that he thought Drees would be a star, but he argued persuasively that Drees should have gotten the chance to find out. “I advocate a simple philosophy: Pitchers who succeed should keep getting a chance to succeed until they fail.” His point was that there was such bias among scouts against Drees’ stuff  which, as I recall, topped out with about an 84-mph heater  that he’s lucky to have a Macmillan Encyclopedia line at all.
Makes me wonder about Franklin … if, in all the reasons to be annoyed with him, we aren’t focusing too much on his lack of strikeout ability … and not enough on whatever it is that actually makes him habitually subpar over the long haul.
Good stuff. I read Trippet’s article about 2 years ago when it first came out and have been fascinated by this topic ever since. I think Davenport’s research is a significant addition to the discussion. Frankly, I don’t think the true significance of Voros’ original finding has been fully mined, yet. Even before Voros’ discovery, many smart old schoolers instinctively knew the value of strikeouts and walk minimization. I can recall many a time when Lou Pineilla would walk to the mound and utter 3 simple words.. “throw strikes, son.”
At the major league level I think this is quite well known as strikeout, walk and HR rates are being used to make decisions about promotions, roles, etc. Davenport is probably on the right track. The most significant use of this information is probably at the minor league levels and even earlier to develop and hone a pitcher’s ancillary skills.
A likely explanation is that at MLB the differences in ability to control BABIP are much smaller than random effects.
The measured BABIP is the sum of the pitchers skill component and a random flucuation component. As low skilled pitchers are weeded out, the luck factor becomes increasingly dominant. The weeding out continues to a point where the luck factor dominates skill effects; at that point the marginal returns from further controlling BABIP diminish greatly.
That hypothesis also recognizes that there still will be some pitchers in the big leagues who are better than their counterparts at controlling BABIP, and that the differences will be relatively small (smaller than the variations among pitchers measured in the lower minors) and will often be overwhelmed by statistical noise. Which is almost exactly what is observed.
Not sure if this has been touched on yet, but are foul balls considered anything but strikes on BABIP type studies? I’m wondering a foul ball should be perceived as a fortunate outcome for the pitcher, neutral for each, etc. Any comments?
This is interesting stuff — what if you looked at first pitch strikes and swings? We know that if you pitch ahead in the count you do better than if you pitch behind.
Say, fascinating tangent: Salamon Torrez pitched yesterday for the Pittsburgh Pirates and allowed his inherited runners to score. Picked up where he left off 10 years ago, eh?
Salomon Torres made his comeback in 2002, and pitched well in his brief time in the majors. In 2002, not as much, but in 2004 he looked really good.
So as to your question — not so much.
Steve’s post @ #3 is excellent, and I agree entirely
How about the flip side of the discussion. Do hitters have any ability to control what happens when they put the ball in play?
#8
Ichiro and the infield base hit. Infielders get so nervous of his speed that they seem to make bad decisions. That is the only thing I can think of and honestly you could shoot holes through my argument. No, he cant control what happens, be he certainly can influlence it. Just thought I’d offer that sincer were brain storming here.
Re 8:
I believe that Ichiro! has proven that hitters can control what happens. If not, he has been extremely lucky his whole career, which would imply that eventually he’ll fall to earth and be a .200 hitter. I haven’t seen any signs of that happening, so I would say that it is definitely possible.
If hitters couldn’t control it, then Juan Pierre would probably hit for a lot better average than he does wouldn’t he?
And maybe guys like Rey Sanchez…
Hitters (not just Ichiro) do show an ability to influence their batting average on balls in play. Otherwise, the only differenctiation between hitters would be their prospensity for Ks and HRs, but we know that’s not true.
Poor hitters are not as likely as good hitters to roll weak grounders to 2B.
Hitters absolutely can control their own BABIP, although not as much as some people believe.
Good hitters can choose if they want to pull the ball or hit an opposite field hit. They can keep the ball on the ground or in the air. This is what makes Ichiro! such a good hitter. He can put it where he wants to. I don’t think much more can be controlled, however, besides right left and up down.
Careful– remember that it’s not a matter of “can or can’t” control what happens, but of “how much” they can control what happens– or, more precisely, whether the natural variation due to randomness outweighs the variation due to skill, and if so by how much. (These are the “two components” from Steve’s #3 above, with which I also agree.)
I took another look at the Hardball Times numbers, this time comparing hitters to pitchers. The percentage of variance due to randomness in a single season (for 162IP/502PA qualifiers), if I did the math right, looks like about:
BABIP: Pitchers 86% random, hitters 62% random.
LD%: Pitchers 63% random, hitters 50% random.
What this suggests is that, yes, a hitter’s BABIP and LD% is more reflective of his skills than the same stats for a pitcher. The above is just a quick estimate, of course– not a full scientific study. But I think I’m generally on the right track with it. It certainly makes sense that BABIP would be more “random,” relative to hitters and pitchers, than LD% would, since BABIP involves the extra components of fielding and park effects. Likewise, it makes intuitive sense that hitters would have more influence on the results of balls in play, since once the ball comes to the hitter, it’s both literally and figuratively out of the pitcher’s hands. (There’s no such thing as a literally unhittable pitch; there are only degrees of difficulty-to-hit.)
Re: hitters controlling balls in play.
I’ve always thought it was kind of funny about the arguments that strikeouts are pretty much meaningless for a hitter. Sure, it’s just the same as any other out, but if you put the ball in play, you’re not guarenteed to make an out.
Note that I agree that there are certain players/skillsets who shouldn’t worry about strikeouts. Adam Dunn, for example, gets his strikeouts as a result of being patient, working deep into counts, and drawing a great deal of walks to counter the strikeouts. If you get into a lot of 3-2 counts, that’s what’s going to happen.
Those high strikeout/low walk guys, though, shouldn’t be given the same free pass. Seems like most of the ‘stat community’ will only focus on the low walk total. When Ichiro shows us that a low strikeout/low walk guy can be productive, that tells me that walks aren’t the end of the discussion.
If the guy is going up to the plate hacking, he would be well advised to actually make contact on those hacks.
“Infielders say it doesn’t matter what the count is — Ichiro is sometimes so locked in, he can actually direct a batted ball to the gap of his choosing. That’s not supposed to be possible on a regular basis, but Derek Jeter spoke of the difficulty of defending against a hitter with such bat-control.
‘If he sees you take a step in, he’ll punch the ball right by you, like he was aiming a hit,’ the Yankees shortstop said. ‘I don’t know anyone else who can do that. And the way he gets out of the box, forget it. If you have to go more than a step to your right or left to get the ball, he’ll probably beat it out.'” – ESPN.com
I remember Hargrove mad the comment that if one were to watch one game of Ichiro’s, alot of hit hits seem to just be out of reach of the fielders and pure luck, but for some reason that seems to happen alot for him.
But anyways, I don’t have any stats. Just throwing commentary out there.
Aaron (re. #16): I suspect that there is sometimes a misunderstanding on that point (strikeouts vs. “any other out”), in both the old-school and new-school sides of the industry. Certainly it’s true that if a hitter puts up a .270/.360/.500 stat line (or whatever), it’s going to make very little difference to his team’s W/L record whether his outs involved contact or not. (Many of us have been frustrated with Seattle management’s [and media’s] difficulties in understanding this principle as it applied to Mike Cameron when he was here, or to Justin Leone now.)
However, what you’re discussing is batting instruction, which is a significantly different matter. The fact that drawing walks is a valuable (and underrated) offensive skill, for example, doesn’t necessarily mean that every hitter should be advised to try to work the count as hard as he can. Oakland under Beane has tried various approaches to try to teach patience at the plate, and so far I don’t know that they’ve really found anything that consistently works.
The key, I think, is for a hitting coach to have a good level of sensitivity to players’ individual hitting styles, so as to be able to recommend improvements in weak areas that don’t undermine the things that the hitter is good at already. If a team pushes Ichiro to be more patient, or Dunn to make more contact, there’s an awful lot of goodness being risked for a tantalizing but uncertain gain.
I think what Paul said is important… it’s linear, not threshold. I would be very wary of speaking in absolutes (“It’s not possible to direct the ball”). Not that anyone here is, but I hear it a lot.
I have no trouble believing that Ichiro has some ability to steer his hits towards gaps. If you assume there is any control over a batted ball at all, and I think you have to, then to me the obvious answer is that Ichiro is simply (like knucklers and Moyer) at one end of the scale – the extreme example of control. And still, obviously, not absolute.
By the by… did anyone see this bit in the P.I. this morning?
“Ichiro Suzuki homered in his first at-bat, then singled in the seventh, giving him nine multiple-hit games in his first 12 starts. He said he came to the plate thinking about swinging for a home run. “If the first pitch was a strike, I wouldn’t,” he said. “But if the first pitch was a ball, I would.” Garcia missed with the first pitch and Ichiro hit the second into the right-field seats, his first homer of the season…”
Ichiro is hilarious. And awesome. Awesomely hilarious.
For me, Ichiro is by far the funniest Mariner.
It’s clear that when Ichiro hits a homerun that he is trying to do exactly that. His swing is completely different. Rather than falling toward first base, he twists himself up as the pitch is being delivered, and then springs at the ball.
Now, I do recall one homerun he hit where he picked a slider off his shoes – it was his second homer of the game. That one looked different. After that game he was asked if he felt he’d been lucky to hit 2 homeruns in one game. His response: “These things don’t happen by accident.”
Thank you Derek for posting the link to another contribution on this fascinating topic. You probably also should get credit for selling the BP subscription I’ll be buying when I go home from work tonight.
This topic is fascinating because although I understand statistical analysis as an economist, and baseball as a fan, I have some work to do where the two come together. I recently have read the seminal articles by McCracken and Tippett, and so far I think Tippett presents the much-better-analyzed point of view. My interpretation of his conclusion is that the statistics don’t disprove the assertion that pitchers have some influence over BABIP. This is, at least from an academic point of view, a valid application of statistical analysis.
McCracken, on the other hand, is extremely sloppy, even negligent, in dismissing the hypothesis without applying rigorous statistical analysis. He is hardly academic in the 1/23/01 article “Pitching and Defense,†sharing little about his methodology. Nor does he even attempt, as Tippet does, to isolate the pitcher from the effect of his defense. His rejection of the hypothesis because it BABIP is inconsistent year-to-year is ludicrous – in his example of Maddux in 98-99-00, virtually EVERY statistic Maddux statistic put up varied widely (except W-L: 18-9/19-9/19-9). Furthermore, his examination of two year vs. Tippett’s 80-some is pretty sketchy. Perhaps I will find on further research that he has rigorously analyzed the statistics, but frankly, much of my current skepticism regarding “statheads†is based on the widespread (though admittedly not universal) acceptance of his assertion.
My own hypothesis is that pitchers DO have varying ability to control the outcome of balls in play, and BABIP (even team-relative as in Tippett) is simply inadequate to measure it. We would need to factor in GIDPs and SACs, for starters, and more work is required to factor out the quality of the defense and the park effect.
Derek, do you have a bibliography on this topic? Perhaps if you posted a required reading list, USSM could be an even more intelligent place to waste our time.
Thanks.
DMZ – agreed. He’s a six-tool player.
Going back to the original topic, that’s one of the (many) reasons why I don’t believe in “clutch” players; if a guy shows a propensity for buckling under pressure, he’s generally going to be weeded out before he reaches the Majors, whether it be in AAA or high school. I have trouble believing that Latroy Hawkins would’ve made it as far as he did if he really couldn’t handle high-leverage situations, for example.
A comment re foul balls and BABIP.
I believe that foul ball outs should be counted the same as strikeouts – a foul ball is not a ball in play because if it drops there is no play on the ball. A fly ball pitcher who induces foul pop ups that are caught is accomplishing something his groundball counterpart is not; he is getting “free” outs that are almost every bit as valuable as strikeouts.
I’ve always thought it was kind of funny about the arguments that strikeouts are pretty much meaningless for a hitter. Sure, it’s just the same as any other out, but if you put the ball in play, you’re not guarenteed to make an out.
I see this comment a lot. It’s simply a misunderstanding of the statement, and its ironic that you make the correct comment in your argument.
A strikeout is essentially the same negative value as a pop fly, groundout, flyout, etc. An out is an out, basically. Your point is that a strikeout is worse than a ball in play, which is obviously true, but totally misses the point.
The people defending the value of high strikeout hitters aren’t arguing if they turned 30 % of their strikeouts into hits and the other 70 % into in play outs. That’s obviously true. The point is that they wouldn’t be any more valuable if they turned 100 % of their strikeouts into other types of outs.
Among outs, strikeouts aren’t any more harmful than any other kind of out. That’s just an empirical fact.
So, since the theory is that by the time they make the big leagues, pitchers have been so heavily weeded out for skill that luck is all that’s left to differentiate (at least for BABIP). But for hitters, there’s still a significant skill component for BABIP.
And I always thought it was quality pitching that was in short supply. According to the “BABIP is just (mostly) luck” theory, we’ve got a glut of arms in the ML and a shortage of bats.
Just to be clear, I think the theory that pitchers have limited control over BABIP is bogus, but I am glad to see some effort to explore the issue more completely. Brian is right about the difference between Voros and Tippett on the subject.
And I always thought it was quality pitching that was in short supply. According to the “BABIP is just (mostly) luck†theory, we’ve got a glut of arms in the ML and a shortage of bats.
What? The balls in play theory doesn’t make any kind of statement anything like that. That’s just a bizarre conclusion to reach.
#24: But couldn’t there be the set of ‘clutch’ players who not only don’t buckle under pressure, but perform well above the norm in it?
And also, I can tell you, despite your beliefs, one player who buckles under pressure, at least last year, is Matt Herges of the San Francisco Giants. I followed the team closely and Herges would do well if put in with a good lead or a non-pressure situation… but take him out of the comfort zone, and he just bombed (got bombed, that is. I literally got scared every time he came to the mound.
I recently have read the seminal articles by McCracken and Tippett, and so far I think Tippett presents the much-better-analyzed point of view. My interpretation of his conclusion is that the statistics don’t disprove the assertion that pitchers have some influence over BABIP.
Tippett’s conclusion was that some pitchers do have a demonstrated ability to lower a batter’s average on balls in play — it’s just much smaller than anyone expects it to be, and tends (as far as we know from current research) to be tied particularly to pitchers who change speeds and most strongly to pitchers who throw a knuckleball.
The fatal flaw in the BABIP conversation (And I’ve been following it since McCracken published his article) is that the elimination of home runs from ‘in play’ is totally arbitrary and pointless. A pitcher delivers a ball and at some point later the bat might hit it. If it does, it goes somewhere. The question is whether the pitcher can impact where that ball goes if the hitter makes contact. But the BABIP statistic eliminates fouls and home runs, which are the two most significant places it can go. So you get left with this absurd statistic that says something like ‘When the ball bounces into this narrow area that represents the in-play area below the wall, can the pitcher make it more likely that it gets hit to a fielder’ to which the only reasonable responses are 1) Probably not and 2) so what. That’s a tiny fraction of a pitcher’s job. A pitcher’s job is to keep the good part of the bat off of the ball, and hopefully not let it into play in the first place. BABIP totally ignores this.
Yes, although I offered my interpretation, I agree also with your paraphrase of Tippett’s conclusion. His actual conclusion, however, omits the qualifiers “some” and “demonstrated:”
“The bottom line, though, is that I am convinced that pitchers do influence in-play outcomes to a significant degree.”
Sidereal (if I understood your comment correctly): Note that the “probably not” half of your “probably not and so what” is very much the same thing that Voros, Tippett, Davenport and the rest of us have been saying.
In answer to the “so what” part of it, however: The reason why this is significant is that people do quite commonly judge pitchers by what happens in that narrow area. If you see a line like “5 IP, 10 H, 5 R, 3 ER, 2 BB, 10 K, 0 HR” (that’s Schilling vs. Toronto today, btw), it’s very easy to think “hmmm, not such a good game– he’s got to do something about those hits.”
But that’s not necessarily true at all. The information we get from DIPS theory is that 5 10 5 3 2 10 0 is not a stable line for a major-league pitcher– giving up 10 hits, and only getting 5 (=5.0*3-10) outs, on balls in play does not fairly reflect any major league pitcher’s skill set. It is far more likely that a bunch of balls just found holes and gaps than that Schilling was basically throwing batting practice when he wasn’t striking people out.
Now, this is admittedly limited information. Line drive percentage is more useful, but unfortunately is not readily available. I don’t know whether the 10 hits off of Schilling were smoked, or whether they dribbled through. If the majority of them really were hard-hit, then I revise my view of the matter. But until such information comes to light, my view of Schilling’s game today is that, if he keeps pitching like he did today, good things will happen for him over the course of the season.
*************************************
By the way, re. Voros’ original article: Yes, I agree that the methodology was weak, and that the conclusion was overstated. Having been through the same situation myself, though in a lower-profile discussion (pitching vs. hitting prospects), I’m not inclined to be too hard on Voros for it. Even if his information didn’t quite prove as much as he originally thought it did, he still did the industry much more good than harm, simply by raising the issue, by giving enough data to take it seriously, and by creating a situation where guys like Tippett would take his work and improve on it.
When I send in my hitting/pitching prospects article to BP, I was expecting that somebody would peer-review it and get back to me with questions like, “Have you considered any other possible explanations? Are you sure that you’re not stumbling into a small-sample trap here?” Instead, it got published with only wordsmithing-type edits. I’m thankful that they published it, but it would have been better with more challenges beforehand. And a similar process might have helped Voros with his original article too.
Hi folks, Jeff here. I’m on the road and away from my login info, but thought I would remind people that tomorrow night, April 19, is our get-together for King Felix’s first home start.
We’ll be gathering at Engine House 9 4ish beforehand, and likely afterward too. Please join us!
Full info (with incorrect, pre-rainout date) here, including directions links.
Dave,
Not a bizarre conclusion at all – it follows logically from the theory.
Here’s what Steve said (and Derek supported):
The measured BABIP is the sum of the pitchers skill component and a random flucuation component. As low skilled pitchers are weeded out, the luck factor becomes increasingly dominant. The weeding out continues to a point where the luck factor dominates skill effects; at that point the marginal returns from further controlling BABIP diminish greatly.
So, ML pitchers have limited influence (compared to one another) over BABIP because, well, they’re already so darned good at it. Any pitcher who didn’t have very good BABIP skills would be left in the minors.
So I asked about BABIP for hitters. Ah, but crunching the numbers implies that hitters do have a significant impact on BABIP. It’s not just all luck for them. So, I guess they haven’t been weeded out as heavily. There must be significantly more bad hitters in the league than bad pitchers, because we haven’t reached the point where all the hitters with bad BABIP skills have been left behind in the minors.
Of course, that is a bizarre conclusion, but don’t blame me, I just asked the (leading) questions.
Personally, I think Sidereal is onto something. To make an arbitrary distinction between a 410 ft HR and a 390 ft out leaves the whole BABIP theory on shaky ground.
#31:
The flaw in your logic here is that you are confusing measurement and interpretations.
BABIP is a measurement, and in compiling those we can also measure the degree to which BABIP correlates from yeaer to year. That measurements shows there is little correlation in BABIP from year to year. Those are empiric facts; they are comparable to sticking a thermometer outside your door and measuring the diurnal temperature variation every day. Then you can compile years of data and measure whether the average diurnal variation in June of 2004 is higher or lower than the diurnal variaion in June of 2003.
Home run rates and strikeout rates are also measurements. Unlike BABIP, when we meaure them across several seasons, we do note that there is some correlation; a pitcher who has a high career home run rate is likely to have a high homerun rate in the next season. Same thing for strikeouts.
The conclusion that many people draw from that is BABIP is relatively little influenced by what pitcher happens to be on the mound, but homerun rates and strikeout rates are much more influenced by who happens to be pitching. I think those are direct and straightforward conclusions, but feel free to disagree – as you certainly do.
But if you do disagree with that conclusion, you need to come up with an alternate explanation for those empiric facts.
The case for pitchers have little control over BABIP is out there. What’s your explanation for how pitchers have significant control over BABIP, but yet are unable to consistently express that ability from year to year?
In the thread on this topic a week or so ago, I made this similar comment, which seems to have gotten buried at the end of that thread.
So, ML pitchers have limited influence (compared to one another) over BABIP because, well, they’re already so darned good at it. Any pitcher who didn’t have very good BABIP skills would be left in the minors.
Yes.
So I asked about BABIP for hitters. Ah, but crunching the numbers implies that hitters do have a significant impact on BABIP. It’s not just all luck for them.
Yes.
So, I guess they haven’t been weeded out as heavily. There must be significantly more bad hitters in the league than bad pitchers, because we haven’t reached the point where all the hitters with bad BABIP skills have been left behind in the minors.
No. You make a huge jump here for absolutely no reason.
The balls in play theory states that pitchers don’t have a significant influence on whether a ball in play turns into an out or not. Because a ball in play is a confluence of skill between the pitcher and hitter, it logically follows that a hitter would then exert a significant amount of control over whether a ball in play becomes an out. If neither the pitcher or the hitter showed any ability to control the result, then all that would be left is random variation, and thats an illogical conclusion. And one unsupported by evidence.
Re: (#s 8,9,10,14,17,19,and 21; especially #s 14 and 19) ICHIRO’s ABILITY TO PLACE THE BALL – There are five holes in the infield (six, if the defense is playing five infielders). [Two of them, 1B to RF foul line and 3B to LF foul line, are negligible.] When ICHIRO comes to the plate, he examines the alignment (including the depth at which each infielder is playing), then aims for the largest hole. (This year, he seems to hit the hole more than 40% of the time.)
Re batters BABIP.
Batter’s BABIP and hitters BABIP are two different statistics. A pitchers BABIP is recorded across a large cross section of hitters, some good and some not so good. In the end, it becomes a measurement against league average.
If a certain pitcher, however, faced only Barry Bonds, Albert Pujols, and Ichiro!, his BABIP against would likely increase. The BABIPs for Bonds, Pujols, and Ichiro!, however, probably would not change.
Another point in response to JMHawkins, et al. is that BABIP is really a lot less significant for hitters than it is for pitchers; a pitcher who’s below-average in that regard is going to have a more difficult time advancing. Hitters who are below average in this regard may still advance due to other reasons, most notably skill in the field (think Pokey Reese).
Re Dave’s post #26:
“The point is that they wouldn’t be any more valuable if they turned 100 % of their strikeouts into other types of outs.”
Is that really true? My take is more moderate: that strikeouts aren’t as bad as some people think, but they’re still not as good as ball-in-play outs.
On a strikeout the only way for baserunners to advance is if it’s a wild pitch or passed ball. With a ball-in-play out, the baserunners have a much better chance to advance. Is this plus totally negated by the possibility of a double play? I doubt it.
Josh,
See this article at Baseball Prospectus. There’s been a ton of other work on the subject as well. In case you’re not a subscriber, here’s the concluding paragraph.
On a very rough scale, a strikeout costs a team about three one-hundredth of a run. Looking at team totals from 2004, Reds batters led the league in strikeouts with 1,335 while the Giants trailed with 874, a difference of 461 whiffs. All those failures at the plate cost the Reds an estimated 13.6 runs over the course of the season, or just over one win. With individual batters who accumulated at least 600 plate appearances, Adam Dunn led the league with a well-publicized 195 strikeouts while Juan Pierre trailed with a mere 35. The 160 strikeouts–the most extreme case in the majors–add up to a difference of 4.7 runs.
So, basically, yes, the possibility of a double play essentially cancels out the value of moving runners.
Dave,
I’ve been a BP reader for three years and I missed that piece. Good stuff.
But I think it shows we were both right.
I can accept that you were being overly dramatic in your original “100” statement…the conclusion of the BP article doesn’t go that far.
If you take 100% of the Reds’ strikeouts from 2004 and make them flyball outs, they get 1,335 X .03 = 40 more runs.
40 more runs is a lot! No one at BP would argue it isn’t.
But this is a hypothetical and as the article correctly states, you can’t just trade strikeouts for flyball outs without also trading in some power and walks, too. But I’m saying if you could swap out, you would. It’s simply not true that there’s no extra value whatsoever.
Hypothetically speaking, the Cubs’ 1080 strikeouts could have been worth about 3 more wins had they been balls in play. And that’s how many games they were out of the wildcard spot. And Sammy Sosa would still be out there in right field, too! Hahaha.
Thanks for the conversation.