Jan
26

Do strikeout totals tell us anything about offensive performance?

By


We’ve all heard the argument before. If high strikeout pitchers are so great, then why aren’t high strikeout batters so bad? Most will argue that you want a guy at the plate who puts the ball in play when you have men in scoring position, and that’s certainly true, but it’s an oversimplified look at things. Mark Teixeira, the number three hitter for the best offense in baseball last season, had runners in scoring position in just over 31% of his plate appearances. That’s it. Miggy Cabrera, the cleanup man for a middle of the pack offense, had men in scoring position in just over 25% of his plate appearances last year. We can’t just ignore the other chunk of plate appearances because of our confirmation bias, though that’s usually what happens.

The Yankees struck out fewer times than all but one AL team last year, so we have the best of both worlds. Dis-ir-regardless, I decided to look into this a bit. What I did was take every batter with at least 400 plate appearances over the last three seasons, and plotyed their strikeout rate against their weighted on-base average (wOBA,, which Joe explained in detail here). If strikeouts are so bad for hitters, then theoretically the players with the highest wOBA’s would have the lowest strikeout rates, and vice versa. As always, make sure you click on the graph for a larger view. Oh, and current Yankees are in pink to make life easy.

So how about that. The data seems pretty spread out, no? The two data points between Chipper and Holliday/Tex are Hanley Ramirez and Chase Utley, and the other two .400+ wOBA players (between Holliday/Tex and Prince) are Manny Ramirez and Ryan Braun. I didn’t want to go too crazy with the labels and clutter things. The of the trendline is microscopic at 0.0021, which suggests there’s basically zero correlation between strikeout rate and overall offensive production.

Now don’t get me wrong, I’m not trying to say that strikeouts are good. They’re bad, we all know it. However, it’s okay to sacrifice a few strikeouts from position players in exchange for other things, like hitting for power or getting on-base at better than average rates. Just look at the graph, you can see that almost all of the players with really high strikeout rates (say, 33% and above) are generally above average offensive players. If that many of your plate appearances end in strike three, you better do other things well at the plate, otherwise you’re useless. Adam Dunn, Ryan Howard, Carlos Pena, Mark Reynolds … all those guys make up for their strikeouts by hitting baseballs far, far away.

At the same time, look at all the low strikeout players that are offensive black holes. Omar Vizquel, Cesar Izturis, John McDonald … those guys contribute nothing with the bats. If you had men in scoring position, they’re the last people you’d want up because they’re the least likely to do something positive. It doesn’t matter that they don’t strikeout much, their wOBA shows they’re offensively inept. The only reason they’ve managed to keep their jobs is because they’re outstanding glove men. It’s a trade off, just like high strikeout totals.

Now what about the other side of the coin? How do strikeout totals affect a pitcher’s performance? For that, I plotted ERA vs. strikeout rate, which I know isn’t perfect. Ideally I’d plot their opponent’s wOBA instead of ERA, but I don’t have that data handy and I’m sure as hell not going to take the time to calculate it. This will have to do for now, but yes, I’m very aware of the flaws. Same deal as above, pink data points are Yankees, click for an enlargisized view.

The two pink data points just below Burnett are Joba and Javy Vazquez, while Andy Pettitte and Chad Gaudin are a little further up the scale.

Unlike wOBA, there’s a pretty significant correlation between strikeout rate and ERA, and it’s easy to see from the graph. The R² of the trendline is 0.33, although we don’t know if that tells us anything meaningful because our sample isn’t very big (I limited it to pitchers with at least 200 IP over the last three years to weed out as many relievers as possible). However, it’s safe to say there’s a (much) bigger correlation between strikeouts and pitching success than there is with offensive success, and it’s pretty obvious in the graph.

The low strikeout guys are higher up on the ERA scale, while the higher strikeout guys are further down. You start at Sidney Ponson and Sergio Mitre then ride the slide down to Zack Greinke and Tim Lincecum. It’s not a coincidence. Strikeout pitchers are the most effective because they take their defense right out of their equation. Like hitters, there’s a trade off, you can live with low strikeout totals if a pitcher does other things well. However, there’s only so much a pitcher can do to make up for it. They can generate an extreme amount of ground balls and limit walks, but even that only goes so far. If you can’t get hitters to swing and miss, you’re going to give up hits. If you give up hits, you’re going to give up runs. It’s just the way it is.

All this post does is reinforce what we already knew: you could still be a good, even great hitter despite striking out a ton, but chances are you won’t be very effective on the mound if you can’t strike out a decent amount of batters. Oh sure, there’s definitely some exceptions, but they’re few and far between. We all hate watching players strikeout when there’s ducks on the pond, but there’s so much more to the game than that.

Photo Credit: Kathy Willens, AP

Categories : Analysis
  • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

    (stands and applauds)

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

      You know what’s awesome about graphs?

      You can see them with your own eyes.

      • Jose Offerman’s Unyielding Rage

        Plus the sweet colors.

      • Tampa Yankee

        ietcvm

        Actually, I let out a nice loud “HA!” and had about 4 people in my office turn around and look at me like I’m an idiot.

    • History Teacher

      Edited by RAB: Pretty off-topic there.

      • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

        Reply FAIL.

  • gc

    Math.

    • Jose Offerman’s Unyielding Rage

      …is power.

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

      … my ears are burning…

    • Rose

      Anybody remember this little segment on the old PBS show “Square One”?

      http://en.wikipedia.org/wiki/Mathman (SAFE)

      • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada
        • Rose

          Can’t go to Youtube, what is it? haha

          • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

            It’s a Mathman video.

            • Rose

              Gotta love Mathman…though it was a little traumatizing seeing the protagonist getting eaten and killed constantly.

              Mathnet was also great.

  • king of fruitless hypotheticals

    a few things:

    1) Damn, Mo is awesome.
    2) Damn, Tex and Arod are awesome.
    3) Damn, being a Yankee fan is awesome, and right now, I think I could play left field and the whole Yankees vs The Field bet is still distinctly Yankee leaning.
    $) Damn, Cashman is awesome.
    5) You guys are pretty damn awesome yourselves.

    oh, but Tommie…put down the coffee: coffee is for graphers.

    • vin

      Gotta give Joe Nathan some love too. He’s a great pitcher, good guy, very deferential to Mo, and doesn’t really kill the Yankees.

      • king of fruitless hypotheticals

        absolutely. if he wants to come play with mauer, he can pitch teh 8th!1! if he wants.

        surprise (for me) was K-rod. heh.

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

      oh, but Tommie…put down the coffee: coffee is for graphers.

      You know what it takes to comment at RAB?

      It takes brass balls to comment at RAB.

  • vin

    If you look at CMW’s 2006-2008 years, he’s in the Aaron Cook zone. ~ 3.74 ERA, 4.1 k/9

  • A.D.

    Tee this post up the next time someone asks/tries to make the hitter vs pitcher strikeout argument.

  • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

    The 2009 Kansas City Royals employed both Tony Peña Jr. and Sidney Ponson.

    FAIL.

    • http://twitter.com/tafkasic the artist formerly known as (sic)

      Trust the process, TSJC.

      /signs scott podsednik
      //moore’d

      • A.D.

        Given that Eaton is a FA, surprised they’re not all over him

    • Brooklyn Ed

      Pena becomes teh pitcher!!!

  • Andy in Sunny Daytona

    So, strike out pitchers are good because they take the defensive equation out of the question when it comes to making outs, BUT, strike out hitters aren’t as bad even though they don’t put the pressure on the defense to get them out?

    I’m sure I screwed this up somehow.

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada
      • Andy in Sunny Daytona

        Boversimplification on my part? I thought as much.

        • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada
          • Angelo

            That definition is one of the best definitions ever…

            • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

              I give because I love.

              • rbizzler

                Love the Lohuddite tag on the definition. It is a personal favorite of mine that I think needs a little more exposure.

                • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

                  Let’s wedge it. Let’s wedge it.

    • Jose Offerman’s Unyielding Rage

      I think another part of it is that when you strike out, you don’t give the defense an opportunity to make a play. As in, a player that strikes out with a runner on first, rather than grounding weakly into a double play.

      Obviously you want to put pressure on the defense, but one out is better than two. Not that the double play scenario is the only example, but it’s the one that makes the most sense to me.

      • Andy in Sunny Daytona

        But double plays really don’t happen all that much.

        • Jose Offerman’s Unyielding Rage

          Maybe because everyone is always striking out!?

          Just kidding. Yeah, you’re right. Like I said, it was just one practical example I could think of when striking out would be better than putting the ball in play.

          • Andy in Sunny Daytona

            The game is ever evolving. In the early part of the 20th century, pitchers didn’t worry about strikeouts, they tried to make that batters hit the ball because the ball was softer than today’s ball. The pitchers realized that you had to try to strike out these guy in order to survive. Conversely, hitters prided themselves on not striking out, they would “take a little off” with 2 strikes to try to make contact. Now, that theory is gone, and hitters, for the most part, are max effort guys on every swing.

            I guess it’s not good or bad, it’s just that the game is played differently now.

            • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

              Totally different era. No way to compare it.

            • Jose Offerman’s Unyielding Rage

              For sure. My coach in college still preaches the two-strike approach and situational hitting, but he also preaches swinging and missing. As in, you’ve got three strikes, don’t waste one by just trying to make contact.

              • Jose Offerman’s Unyielding Rage

                And there’s a definite disparity in how the “new age” players and coaches look at striking out than the old school guys.

                The guy I played for last year was very much an old school coach who had been around for years and he might have preferred you coming at him with an ax to a strikeout. Whereas my coach this year is fresh out of college, only about 26 years old and is much more about being aggressive on every swing no matter what the count.

          • pete

            also, one of the more common examples is the ever-present missed-third-strike-with-a-man-on-third scenario, which generally results in not only no out being recorded, but also a run scoring.

            In all seriousness, though, it’s true that there aren’t many situations where striking out is BETTER than making another out, but there are also similarly few situations wherein absolutely any other OUT is better than a strikeout. In each situation there is sort of a ranking of favorableness of outcomes, and the strikeout is at the bottom of that list roughly never, on account of it at the very least requiring 3 pitches. generally, it goes triple play, double play, and then a number of outs that hold essentially the same value (any out in play that doesn’t advance the runner, strikeout) other than how many pitches they require, and then the “productive” outs. Given the unlikeliness of scoring a run in any particular inning to begin with, all outs suck, and while they vary in detrimental effect, the degree to which they vary is minimal. Especially relative to the degree to which non-outs vary, and even more especially relative to the variance between non-outs and outs. IOW, the difference between guys who don’t make a ton of outs and make a lot of good non-outs and guys who make a ton of outs but a fair number of “good outs” in the process, is huge.

      • A.D.

        Else another way to think of it (though this focuses on batting avg, not the best:

        Almost all pitchers end up with a BABIP ~.300, so given this the big differences between pitchers will be their ability to record outs (or prevent hits) from balls out of play, which the big one will be the strikeout, and then of course the damage they give up on these hits in play & preventing walks.

        For hitters there is no avg BABIP that all hitters seems to be around, a given hitter will have an avg BABIP, but we don’t see the trend of pitchers, therefore a hitter can strike out a bunch, and just have the natural talent to hit for a better BABIP, and thus still hit at a high level.

        This, as I noted, really just goes towards batting avg vs the damage a hitter does & captured by the better metrics, which allows a Nick Swisher or Adam Dunn to be excellent offensive weapons, despite low BA.

  • Rose

    The Yankees struck out fewer times than all but one AL team last year

    Pretty amazing how close all those numbers are. There all pretty much clumped together in groups…no one team is out on Venus with some astronomical number on either side…

    • Bo

      Thats because they had some pretty good hitters dont u think?

  • mryankee

    Man this offseason is long. I mean these graphs are nice and all but we need some kind of interesting news to discuss,

    • http://www.riveraveblues.com Joseph Pawlikowski

      Your comments about a boring off-season are not welcome. You do it all the time. Please do not waste space in our comments section by repeating your boredom ad infinitum. Thank you.

      • Angelo

        +42

    • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

      My advice: Start drinking heavily.

    • jsbrendog officially approves signing Fernando Tatis

      then. go. to. another. blog.

  • allan

    The most amazing thing is the dot on the hitters graph with the name Pujols next to it. Let’s you see how he is on a different scale than any other batter in the majors at this moment in time.

  • OldYanksFan

    Hey Man…… look at all the dots!

  • Peter

    i’m kinda getting tired of the strikeout posts.

    just one guy sharing his opinion. No need to bash me.

    • pat

      Riiight, they typed up a whole post because you don’t like strikeouts.

    • http://www.riveraveblues.com Joseph Pawlikowski

      You could keep it to yourself, too. We’ve had two strikeout posts this month.

    • Angelo

      Waste of a comment.

      I can listen to no one caring now.

      Be a fanatic, pessimism is unneeded to this brilliant site.

  • Ivan

    Im kinda surprise the yankees are the 2nd best team in concern of being struckout the least. Anyway, its really common sense, if your gonna k alot you betta hit alot of homers and be a high on base/OPS guy. Guys like omar vizquel have no value anyway besides putting the ball in play on a consistent basis and when it is in play 90% of the time its a out.

  • pete

    I think it’s worth noting that many of the low-strikeout, low-production guys are simply incapable of hitting the ball well, so they have much lower BABIPs. In other words, you can choke up and “just try to put the ball in play” on two strikes, but that approach will probably cut your BABIP in half. If the best hitters in the league who are always swinging hard get out on 65% of the balls they put in play, someone who hits nothing but weak grounders and bloopers will get out MUCH more – probably closer to 85 or 90%.

    • king of fruitless hypotheticals

      ichiro?

      • http://twitter.com/tsjc68 tommiesmithjohncarlos a/k/a Archimedes Torquemada

        GGBG?

      • pete

        well yeah but ichiro is the exception, not the rule. typically, if you don’t hit the ball hard a lot, but still hit the ball a lot, you will make lot of outs, and typically making lots of outs renders you an unproductive player

        • king of fruitless hypotheticals

          Pete, do you know what the difference between hitting .250 and .300 is? It’s 25 hits. 25 hits in 500 at bats is 50 points, okay? There’s 6 months in a season, that’s about 25 weeks. That means if you get just one extra flare a week – just one – a gorp… you get a groundball, you get a groundball with eyes… you get a dying quail, just one more dying quail a week… and you’re in Yankee Stadium.

          • pete

            Love the quote. Lost on the relevance to this topic though.

    • RollingWave

      Your correct in that most of the really good hitters have higher BABIP in the .350ish range but your wrong in the later comment. most of these info can be looked up easily on placs like fangraphs. RAB doesn’t actualy post very deep math analysist, they’re posting information anyone can find easily.

      career BABIP

      Omar Vizquel : .298

      Cesaer Iztruis : .284

      John Mcdonald : .274

      Miguel Cairo: .295

      Yes weaker hitters ends up with less of their balls going for hits . but the degree is not nearly as high as your stating. the difference between the best and worse hitter is roughly 1% not 30%

      which puts us back into the same loop. there are basically 2 ways of not whiffing much.

  • rbizzler

    Great post Mike.

    Just another example of the value of challenging the CW. While strikeouts may be frustrating for some fans to watch, they are a necessary evil when it comes to production.

    Gotta love the productive outliers though – Pedroia, Mauer, Chipper, Albert (obviously).

  • Ed

    Nice analysis. It appears that it is OK to trade some contact for power. My questions are:

    1) How does this analysis hold up with runners on base – especially on third with less than 2 outs. Some contact, even a fly ball, seems more necessary in that situation.

    2) How do hitters with similar power numbers, but different strikeout totals compare. I wonder if a high power, low strikeout guy would have better overall stats than a high power, high strikeout guy.

    • pete

      the answers to both of your questions are in the post

  • DocBooch

    I subscribed to the RSS feed for RAB, which is awesome…you guys do a great job with your feeds.

    Anyway, I have a few problems with this analysis.

    I understand why you chose to compare wOBA with K/Ratio but after looking at your analysis it is apparent that it is an apples to oranges problem. While wOBA is awesome to see an individuals production of getting on base, a K is an out without runners advancing(99% of the time at least). It doesn’t account for driving in runs, which is what a K absolutely prevents in any circumstance.

    Also, the K/Ratio vs ERA is a perfect correlation. Therefore, if a given pitcher is striking out more batters, then an individual batter isn’t necessarily going to show this stark comparison in his stats.

    Maybe you could look at TEAM k/ratio and compare it to their total production of runs and play around with those numbers. I’m sure it’s no co-inicdence the Yanks had the most potent offense and struck out the second least. Or if you want to keep this individual, compare a hitter with a low k/ratio to one with a high k/ratio and their other offensive production stats like RBI chances and total RBI’s.

    Most of this stuff is way over my head as far as the statistics involved, but I’m at least halfway intelligent enough to see some flaws.

    • pete

      as was noted above, the team K rates are wayyy too close together to pull together a meaningful graph. And I think you’re misinterpreting the post. Mike isn’t arguing that the strikeout is productive, or that there aren’t situations wherein it’s a statistically “worse” out than others. The point is that hitters who have high strikeout rates can still be productive players, while hitters who have low strikeout rates can still be unproductive players. Also, wOBA is not just a way of observing how good a player is at “getting on base” – that would be OBP. wOBA shows us how productive a player is (although it doesn’t adjust for park factors and such, and is only a vacuum statistic – wRC+ is a better way of looking at a player’s offensive production) on offense in a statistically-average vacuum. In other words, if a players strikeouts all come with men in scoring position and his hits and extra base hits and walks all come with the bases empty, then wOBA assumes (perhaps incorrectly, but that’s another debate) that that is merely a factor of bad luck, and does not discredit the player for that. Its weights for strikeouts/groundouts/singles/etc. are based on how those events correlate to runs on average, based on 50 years worth of data. So using player wOBA vs. strikeout rate is actually a very good way of examining the issue that mike is actually examining.

      • DocBooch

        Thanks for the explanation, i’m a little familiar with wOBA but nowhere near an expert. What you said makes sense, but it’s just strange that it doesn’t correlate with the other side of the ball. It is obvious that K/ratio benefits a pitchers production. Or perhaps its just a co-incidence since better pitchers make hitters miss or put weak balls in play. Sometimes we can get numbers to prove a point even though it really has nothing to do with it.

        • pete

          well technically they’re unrelated.

          Players with good wOBAs have those #s because of what they do when NOT striking out or getting out in general, which is typically positive. What the post (accurately) suggests is that the strikeouts do not weigh down players who produce well when they don’t get out very much, and that not striking out doesn’t bring players who get out a lot up very much. Pitchers with good K-Rates, on the other hand, generally have good K-Rates on account of being difficult to hit. Thus they generally have correspondingly good ERAs. All that Mike is really saying is that K-rates are much more closely related to how GOOD a pitcher is than they are to how BAD a hitter is.

  • Greg

    It would be interesting to see what the correlation is of guys who strike out on fewer pitches vs. guys who strike out on more pitches (and also perhaps guys who strike out swinging vs. looking – or pitches taken per at bat vs. pitches swung on an missed per at bat – with some adjustment made for guys that always take the first pitch or always take on 3 and 0). I’ll bet there’s more of a correlation between number of pitches per strikeout and overall productivity as a hitter. I think a guy like Swisher, who takes strikes as part of his strategy because he’s looking for better pitches (or the walk) and may strike out because he runs out of pitches to take, will be much more productive than a guy (like Soriano used to, or better examples such as utility infielders and pitchers) who take the first pitch strike because they are told to from the bench, then swing at the next pitch without discernment. I believe strikeouts on fewer pitches MAY (although not always) be correlated to more swinging strikes and thus indicate less skill in actually contacting the ball (whether they cannot catch up the fastball – like pitchers, can’t hit a major league curve- 4A players, or just don’t have a good eye- like Soriano). The third kind (bad eye) can still be productive if they have great reflexes and/or strength, but will never produce as many runs (unless they have a lot of power like Mcgwire did during his huge years) as a guy who may strike out just as much (but does so more often looking rather than swinging…OR, strikes out, but on the 5th, 6th, 7th, pitch of the at bat).

    I think we’re comparing the wrong factors here- I think we need to separate guys who take more pitches vs. less pitches and also separate again by guys who strike out looking vs. swinging. This comparison might give us more concrete results describing whether or not high strikeout guys are, as a whole, just as productive as guys who don’t strike out as much. I think there will be a difference.

  • http://thestrikeoutisKrap godfather

    that this “issue” comes up on a site that hallows the team dimag adorned makes it even more ludicrous; i’ve heard the arguments and excuses (strawberry: “better than hitting into a dp”)….well, actually, not always; nyy scored the winning run in the 1-0 finale of the ’62 Series on a bases-loaded dp of the standard variety; the k relieves the defensive team of any responsibility…how many runs are driven in on outs? wanna study something…dwell on the asinity of a batter hitting into a dp not getting an rbi…or the insipid idea that a dp can’t be assumed even on a play your wife could have made…or the indifference rule

    • pete

      Here’s what I suggest: go back and read the article again. In the beginning, you’ll come across a term called wOBA, which is linked to an excellent explanation Joe wrote on this stat. If you are already familiar with wOBA, read on. If not, you will be unable to comprehend the rest of the article until you familiarize yourself with the statistic. Joe’s write-up is, as I said, excellent, and I recommend you read it before proceeding with this article. Now you’ll find a nice explanation of how there is very little correlation between hitter productivity and strikeout rates, backed up by evidence that is conveniently displayed in the first graph. Mike then offered some opinions on why this might be so, as have numerous commenters (myself included). Then Mike lays out how, interestingly, there is actually a fairly significant correlation between pitcher strikeout rates and overall performance. After this, Mike, and again many many commenters, have offered plenty of suggestions as to why this might be (again, myself included). From the evidence we can, however, absolutely, indisputably, unequivocably conclude, that striking out a lot does not necessarily mean you are a bad hitter, despite the fact that pitchers who strike out a lot of hitters are typically good pitchers. So please, go back, read the article again, read the comments, and, in the future, refrain from ranting about topic that has been discussed in-depth prior to your post without having read or considered the aforementioned discussion. Please. Most of the members of the RAB community are very charitable with their time, offering well-thought-out opinions and responses to objections. It is very frustrating, however, when a commenter doesn’t thoroughly read the post and comments that are there before he/she comments, because this frequently leads to asking questions or raising objections that have already been addressed, and add nothing new to the discussion.

  • Pingback: The two sides of making contact | River Avenue Blues

  • Pingback: Going for the Kill: Bring in Adam Dunn | River Avenue Blues

  • Pingback: 2012 Season Preview: The Strikeout Kings | River Avenue Blues

  • Pingback: The Supposedly Decreasing Importance of Strikeouts | Community – FanGraphs Baseball