It’s Really Hard To Score A Run When You Strikeout
Posted by Steve Lombardi on May 11, 2010
File this one under "The Elephant in the Room."
Via Baseball-Reference.com's Play Index Pitching Event Finder, through last night's games, there have been 36,900 Plate Appearances by batters in 2010 - again, to date,. Here is the overall outcome of those PA:
| G | PA | AB | R | H | 2B | 3B | HR | BB | IBB | SO | HBP | SH | SF | ROE | GDP | BA | OBP | SLG | OPS |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 479 | 36900 | 32546 | 4272 | 8330 | 1766 | 181 | 905 | 3454 | 224 | 6812 | 305 | 329 | 259 | 356 | 898 | .256 | .331 | .405 | .735 |
.
Seeing this, it's safe to say that 11.6% of all Plate Appearances (this season to date) have ended up in a run being scored. Most of those runs were the result of the batter reaching base via a hit.
In fact, this season, to date, 43.1% of the time a batter gets a hit a run comes around to score. And, 2.5% of the time he walks, a run comes around to score. (Yes, I thought the "walk-run" rate would have been higher too. But, that's what it is.)
Note that there have also been 6,812 strikeouts so far this season. Here is the overall outcome of those "strikeout" PA:
| G | PA | AB | R | H | 2B | 3B | HR | BB | IBB | SO | HBP | SH | SF | ROE | GDP | BA | OBP | SLG | OPS |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 479 | 6812 | 6812 | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 6812 | 0 | 0 | 0 | 0 | 31 | .000 | .000 | .000 | .000 |
.
Seeing this, it's safe to say that 0.06% of all batter strikeout plate appearances (this season to date) have ended up in a run being scored.
There have also been 17,554 "non-SO outs" recorded in PA by batters, to date, in 2010. Here is the overall outcome of those "non-SO out" PA:
| G | PA | AB | R | H | 2B | 3B | HR | BB | IBB | SO | HBP | SH | SF | ROE | GDP | BA | OBP | SLG | OPS |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 479 | 17554 | 16982 | 488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 314 | 258 | 0 | 866 | .000 | .000 | .000 | .000 |
.
Seeing this, it's safe to say that 2.8% of all batter "non-SO out" PA (this season to date) have ended up in a run being scored.
In review, so far this season, overall, a run comes around to score 11.6% of the time that a batter steps to bat. One scores 43.1% of the time that he gets a hit. Further, one scores 2.5% of the time we wallks and 2.8% of the time he makes contact for an out. (Maybe a walk is only as good as non-strikeout out?) Lastly, and most importantly, a batter, to date this year, almost never makes a run come around to score a run when he strikes out.
Related, scoring runs in baseball is an important matter. So, would you say that batters who strikeout are not helping teams score runs as much as batters who hit, walk, or make an out via contact?
At least...that's what these queries show me today. If anyone wants to check my math, etc., you have my thanks in advance.
May 11th, 2010 at 9:10 am
Is there any way to see how many (if any) players score on a wild pitch/dropped 3rd strike?
May 11th, 2010 at 9:11 am
I'd love to know the four scenarios this season where a strikeout plate appearance led to a run being scored. Were they all steals of home?
May 11th, 2010 at 9:18 am
They probably scored during the AB which ended with a K, no? presumably on a wild pitch.
May 11th, 2010 at 9:21 am
Not sure what the point of this was but the analysis is very misleading. The data here is apparently (I think) limited to what happened on that exact play. The only way a walk can lead to a run in these circumstances is with the bases loaded whereas many types of hits can drive in runs. The "walk-run rate" for players walking is FAR higher than 2.5%.
May 11th, 2010 at 9:25 am
Andrew -
I looked at just one, and here it is:
2010-04-19 Brian Bannister KCR @TOR
Alex Gonzalez batter
tied 0-0 b3 -2- 1 10 (3-2) 1
Strikeout Swinging, Wild Pitch; McCoy Scores/No RBI; Gonzalez to 1B
May 11th, 2010 at 9:27 am
Chris - yes, that's my understanding as well - that it's what happened on that exact play. But, is that not the same, pretty much, as what Run Expectancy tables show?
May 11th, 2010 at 9:34 am
Andy you present a good theory but if it holds true we are left with the "Ryan Howard Paradox." He is striking out 180+ times a year but also leading the league year in and year out with 140+ RBIs. Interesting theory. I think this really means that if Ryan Howard can ever start making at least some more contact (not likely but he has proven to be a hard worker) he would be a beast.
May 11th, 2010 at 9:39 am
Steve, I don't have easy access to numbers to back me up but I would guess that (at bare minimum) 10% of players who walk end up scoring. I know that's different than your point but I think it's easy for some to confuse the issue.
May 11th, 2010 at 9:46 am
Steve,
I think you are misstating the percentages because you are saying that X% of at bats result in runs being scored, but from what I can tell you have formed the percentages by simply dividing total runs scored during the given scenario, by total occurrences of that scenario. This fails to account for PAs which result in multiple runs being scored, thus greatly overstating the percentage of hits and general plate appearances that result in a run being scored on that play. I would guess there are a handful of plays where the player made an out, but multiple runs scored because of an error and few, if any, multiple run scoring scenarios for Ks and BBs (though this number would be non-zero given a large sample size).
May 11th, 2010 at 9:55 am
Fair point Evan. But, still, if you look at the SO PA and the non-SO out PAs, it still shows that non-SO outs are more, for lack of a better word, productive, right? (He says, ducking, knowing that "productive outs are a dirty saber-word.)
May 11th, 2010 at 9:58 am
Ok, I've stared at it a while and can't come up with a scenario where it works out. How on earth does a batter strike out but also ground into a double play? Allegedly, it's happened 31 times. If it's a strike 'em out/throw 'em out situation (which I'd still wouldn't imagine's already happened 31 times this season), that's still not a GDP. Is this just a case of poorly defined terms?
May 11th, 2010 at 10:19 am
Tim, I think that's it - SO/CS is labeled as a GIDP. Not pretty. Agreed, if true.
May 11th, 2010 at 10:26 am
Isn't the premise that the SO isn't so bad because it is generally produced by the same type of swing that produces a lot of extra-base hits and run production? I don't think that anyone would suggest striking out in an individual PA is a positive result for that PA (with the obvious exception of positive results because of a WP/PB; or, I suppose, the extreme pessimist fan who assumes the player will hit into a double play).
May 11th, 2010 at 10:40 am
Following a college-level team one summer, I found that when a lead-off batter reached base, he scored about 45% of the time. Coaches always said "Lead off-walks score 80% of the time." That was obviously wrong. What was funnier was, lead-off 1b's and 2b's scored more often than 3b's! And, of course, lead-off HR's scored 100% of the time. Ha.
May 11th, 2010 at 10:41 am
Evan - FWIW, those familiar with the study of Run Expectancy have shared with me in the past that "going down on strikes" is merely another vanilla form of being retired - and that the "K" was no better or worse, for the most part, than any other way of being retired as a batter.
May 11th, 2010 at 10:42 am
I agree that this is very misleading.
Far more than 2.5% of batters who walk eventually score. The above just shows the results on that particular PA, which can be interesting, but it is not telling us what you seem to think it is. It doesn't really have anything to do with run expectancy. Non-K outs are not better than walks.
And of course there's no production as a result of a strikeout. That's not news. As Evan points out, teams accept the cost of a strikeout if it comes with the hitting that a big swing can provide. Unfortunately, no one has yet figured out how to homer in the same AB that they struck out.
May 11th, 2010 at 10:46 am
Steve, linear weights shows that strikeouts are marginally worse than non-K outs _on average_. If no one is on base, there is no difference. If the people on base don't advance, there is no difference. If there is a double play, the non-K is worse. Obviously, if the runner can advance on the out, that is a benefit. If you look at all PA across baseball, the run-value of all strikeouts is slightly less than the run-value of all other outs. In any one specific PA, the value of making contact may indeed be elevated.
May 11th, 2010 at 11:21 am
One offensive impact of the strikeout (as well as the walk) that we haven't discussed is effect that it has on the pitch count of the opposing pitcher. Although it is difficult to quantify the indirect impact of pitches seen on the number of runs the offense will produce later in the game, it is apparent that many teams employ a strategy of working the count in order to get into the opposing team's bullpen.
Although only tangentially related to the topic of this post, given the impact of pitch counts on today's game I think it would might be instructive to look at the average Pitches/PA for the various events that can result from a PA.
May 11th, 2010 at 11:23 am
Johnny - when you say "teams accept the cost of a strikeout if it comes with the hitting that a big swing can provide," that sort of implies that teams would prefer Rob Deer over Mark Grace...and I don't think that's true.
To me, it seems as if teams are willing to live with him whiff totals if the batter offsets that with a good to high OBA. And, if there's power too, great. But, power and K's alone are a bad match.
May 11th, 2010 at 1:01 pm
@#7 Jim: as much as I would love to take credit, Steve wrote this post, not me.
May 11th, 2010 at 1:06 pm
Take the credit or the blame, Andy? 😉
J/K - for the record, my thanks to all for the feedback on this one. Keep it coming! This is a topic that I always enjoy discussing, and, learning more about, from others.
May 11th, 2010 at 1:09 pm
It's not that simple. If someone strikes out 300 times a season and never walks, but hits 100 HR, he's gonna be a valuable player. That's not a bad match. (Admittedly, it's also not a realistic stat line.)
A player is as productive as he is, based on the entire package. You can measure that by OPS, runs created, WPA, whatever. Of course, it's easy to say that if you cut Mark Reynolds's strike outs in half, he'd be a better player. But what would be the cost of his striking out so much less? Maybe his HR would be cut in half too. Maybe he wouldn't hit any HR at all, so affected is his swing by trying to simply make contact.
I mean, I think it's obvious that teams accept strikeouts. Strikeouts keep going up. Players who strike out a lot don't get benched simply because they strike out a lot, if they are productive overall.
Run a correlation between HR and K per AB...I'm sure it's quite high. I'm sure it's quite high if you do it all the way back to 1920 (adjusting for league rates).
May 11th, 2010 at 1:09 pm
sorry, that was responding to #20
May 11th, 2010 at 1:15 pm
Also keep in mind that it's sometimes better to strike out than ground out, i.e. when grounding out leads to a GIDP.
May 11th, 2010 at 1:19 pm
All of this depends on context. As you would expect, if you look at a game by game basis the cost of a strikeout increases with the number of total bases a team has. Strikeouts on a team that doesn't reach base is not nearly as detrimental on one that relies on station to station baseball.
May 11th, 2010 at 1:50 pm
~~sometimes better to strike out than ground out, i.e. when grounding out leads to a GIDP.~~
I hear that a lot. But, it doesn't hold water with me. How many times are we told that RBI are dependent on batting order and the OBA of those in front of you - and that RBI% is a better stat. Same thing applies to GIDP vs. the K. The former is more dependent on your team, line-up, etc., than it is on your ability as a hitter.
May 11th, 2010 at 1:53 pm
~~All of this depends on context.~~
Agree, 100%.
Sheldon Ocker recently wrote on this. A snip:
The problem is that hitters who are prone to striking out can't pick their spots. They can't turn off their strikeout swing just because a runner is on second or third. Theoretically they could, but most free swingers are so accustomed to hacking away at will, they probably would strike out trying to make the adjustment.
So isn't it more sensible to frown on all strikeouts and teach hitters in high school, college and the minor leagues to temper their swings? I'm not suggesting that baseball return to the dead-ball era, when the primary goal of a hitter was to reach first base by slapping pitches all over the field. What we have now is the polar opposite of that approach, and it's far from perfect.
Is there something to gain from striking out? Unless a batter swings at strike three on a pitch that ends up bouncing to the screen, and he reaches first, there is no useful purpose to a strikeout. If a hitter makes contact, even bad contact, something beneficial might occur.
http://www.ohio.com/news/break_news/93200084.html
May 11th, 2010 at 2:28 pm
GIDP do depend on having runners on base, but that doesn't mean they aren't costly. And certain types of hitters do hit into them more, regardless of runners on base (slow, righties, groundball hitters).
I completely disagree with that Ocker excerpt. I'd assume most players who reach the major leagues rarely strike out in high school. Why would they temper their swings if they already make contact all the time? If there are other players in the minors who take shorter swings and strike out less, they will get promoted if that is what teams value. If those players are not getting the major league jobs, it's because they simply don't hit the ball hard enough, unless you think baseball as a whole is complete inefficient in its hiring and deployment practices.
If you want a game with fewer strikeouts, you need to change the game so that becomes the preferred strategy. Deaden the ball, enforce thicker bat handles, enlarge the fields. Players/teams didn't just decide to start striking out more because they don't care. The game evolved to this point because these are the types of skills which are most useful.
The question is not "Is there something to gain from striking out?" It is "Is there something to gain from taking the types of swings that may make it more likely you will strike out?"
May 11th, 2010 at 2:31 pm
Or more precisely, Is there something to gain from taking an offensive approach that may make it more likely you will strike out? (this may include taking more strikes to wait for a perfect pitch and to draw more walks)
May 11th, 2010 at 2:39 pm
Looking at the rest of the article quoted by Steve, Ocker focuses entirely too much on swinging hard and the third strike as the causes for a strikeout, ignoring the the first two strikes entirely. A player who wants to minimize his strikeout total should swing at every pitch he perceives to be a strike (thus maximizing his opportunities to "put the ball in play"), but this is a terrible approach to hitting. Strikeouts accompany increased production not just because the batter is swinging harder, but because the batter is letting hittable strikes go by, waiting for a pitch that is more to his liking. Taking a lot of pitches increases the frequency of strikeouts and walks (e.g. Nick Johnson), but it also gives the batter the opportunity to get hits on mistake pitches and pitches where he guessed correctly or looked for a particular pitch, knowing that he would let anything else go by.
May 11th, 2010 at 3:17 pm
Johnny stole some of my thunder a bit, but I'm not sure I agree with him regarding rules/parameter changes to decrease strikeouts. Giving the batter incentives to hit the ball is the best way (in my opinion) to reduce strikeouts, e.g. reducing the size and quality of gloves, shortening the distances between bases, but these help turn strikeouts into hits.
Counterintuitive as it might seem, increasing the size of the strike zone might decrease strikeouts. It would decrease the likelihood that the batter would receive a walk and decrease his expectation of receiving a "fat pitch" on subsequent offerings.
May 11th, 2010 at 4:17 pm
One of the more notable runs scored while a strikeout took place (from my vintage point) took place on October 2, 1974. The Pirates needed a win to clinch the NL East, but trailed the Cubs, 4-3 in the 9th. They had Manny Sanguillen on third and two outs when pinch-hitter Bob Robertson faced Rick Reuschel. On a 2-2 pitch, Reuschel fanned Robertson--except that Steve Swisher, the Cubs' catcher (and father of current Yankees' OF Nick Swisher) let the ball get past him. As Swisher threw to first to complete the K, his throw hit Robertson in the back. The ball got away from the Cubs' fielders and Sanguillen scored the tying run. Then, in the 10th, an infield hit by Sanguillen scored Al Oliver with the run that gave the Bucs the win and the NL East title. Though the Pirates would go on to fall to the Dodgers in the NLCS, it is interesting to talk about how a strikeout led to a rare run, or even microscopic, a division title.
May 11th, 2010 at 4:23 pm
Correction: it was not a Reuschel wild pitch; it was a passed ball charged to Swisher. Nevertheless, a strikeout on a WP/PB does not often lead to a run scoring, much less a division title as the '74 Pirates discovered.
May 11th, 2010 at 4:47 pm
The more famous passed ball on a strikeout was the Dodgers vs Yankees in the 1941 World Series.
May 11th, 2010 at 5:45 pm
I've joined this conversation late in the day. I thought that these were at-bats that led to the batter eventually coming around to score. So, the dropped third strike situation would be one in which the batter reached first, whether or not anyone scored from third (or even further) on that pitch. Or maybe that batter would eventually be caught stealing second, as happened to Jayson Werth in the Phillies' game against Atlanta on April 20.
May 11th, 2010 at 5:57 pm
As someone who has watched Wilson Valdez GIDP in EACH of the last five games the Phillies have played, I can assure you a non-strikeout out is not preferable. It's CERTAINLY not preferable to a walk. That's just asinine. You have 27 outs to work with, so you want to get on base without making an out as often as possible.
May 11th, 2010 at 8:53 pm
[...] Valdez has quite a GIDP streak going. As Rich, a reader, mentioned in the comments of another thread, Valdez has now hit into a double play in 5 consecutive games (Tonight's Phillies [...]
May 11th, 2010 at 9:34 pm
The issue seems to be people conflating the idea that "strikeouts aren't as bad as often said and probably no worse in the big picture than other outs" with the idea that "strikeouts are, in the absolute, good!" Clearly, the latter is not true while it's quite likely that the former is. Coupled with some particular modern players who strike out a ton and IN SPITE of these strikeouts (or maybe indirectly because of them with regards to general approach as a hitter) are still remarkably productive, there seems to be a general misconception that the modern SABR approach embraces strikeouts as a positive. Which I'm pretty sure is NOT true.
May 11th, 2010 at 9:45 pm
A few more things:
1. Runner on first, one out, I'd much rather see the batter strike out than hit into a double play. Of course, the next batter may strike out, too (or hit a long fly ball to the warning track that the outfielder makes a great catch to keep from being a home run or even get a double on which the guy who had been on first gets thrown out at the plate to end the inning).
2. An at-bat that ends with a strikeout involves at least three pitches. If there are runners on base, a lot can happen when these pitches are thrown. Good things for the offensive team include stealing a base, a balk being called, a passed ball, or a wild pitch. The last two can come on one of the earlier pitches, before strike three. Of course, for any of these situations that doesn't cause a run to score, if that strikeout is the third out, it was only a temporary good thing. But if it was only the first or second out, the runner(s) who advanced is/are that much closer to scoring.
The stat of batting average with runners in scoring position is not always meaningful. Getting a hit with a runner on second doesn't always score that runner. Conversely, making an out (not the third out, of course) that has a charged at-bat (grounder to the right side) sometimes scores a runner from third. The latter situation also includes some errors and fielders choice situations in which everyone is safe.
May 12th, 2010 at 2:43 pm
Dear Johnny Twisto,
It's true that a passed ball did result in a series of runs in the 1941 World Series. However, when Hugh Casey fanned Tommy Henrich and the ball got past Mickey Owen, there were NO men on base. Hence, Henrich went to first base. Even with the passed ball, the Dodgers still had a chance to end the game with a 4-3 win. Then, Charley Keller singled, and Henrich went to second. Next, Joe DiMaggio doubled and THAT'S when Henrich and Keller scored. Everyone knows what happened after that. When I did the 1974 Pirates segment, the passed ball directly resulted in a run scored. When Owen dropped the ball, the Yanks didn't score immediately. The Yanks just took advantage of their second chance.