Lots of Whiffs, but where are the Ks?

Sort among batters for swinging strike rate over the past three years, and (unsurprisingly) you’ll get a list that features most of the strikeout kings at the plate. Mark Reynolds (17.5% swinging strikes) and Ryan Howard (15.3%) obviously belong based on their respective career strikeout rates (38.7% and 32% respectively). But track a little further down the list and you’ll notice that a high swinging strike rate is not always a harbinger of a high strikeout rate.

You won’t have to go far. Even third place on the list, Josh Hamilton, has a decent strikeout rate (20.6%) despite his high swinging strike rate (13.7%). Delmon Young even pairs a better-than-average strikeout rate (18.6% career) with a top-ten worst swinging strike percentage (12.7%) over the last three years. Clearly, the tendency to swing and miss at offerings is not the sole determinant of a batter’s strikeout rate.

Follow the list down to the way 20th spot to where a possible reason for these outliers might become clear. Yes, Vladimir Guerrero comes in as the 20th-highest swing-and-misser by swinging strike rate over the last three years, and yet his career strikeout rate (12.2%) is well below average (usually around 20%). Now you’re looking at three players (five if you count Jeff Francoeur and Kevin Kouzmanoff) that pair decent strikeout rates with poor swinging strike rates, and all of the players have something in common.

Yup, all five players have below-average walk rates and are known for being free swingers. The lowest three-year reach rate of this crew was Kouzmanoff’s 34.5%, so they were all well-above average in that category, and though Hamilton’s career walk rate (8.3%) is decent, he has dropped that number to 6.6% and 7.5% the last few years. It looks like aggressive free swingers can avoid the strikeout even in the face of poor swinging strike numbers.

And in a way that’s completely understandable. Swing often, and early, and you may never get to that two strike count where the whiff means a K. By take more pitches, the count gets deeper, and the likelihood of a strikeout rises. A lower strikeout rate is a strange ‘benefit’ to having a free-swinging approach, but it looks to be true.

The effect seems to hold. Let’s set the three-year benchmarks so that a player has to have swung at more than 34% of pitches outside the zone and walked less than 8% of the time while showing a swinging strike percentage over 10. The highest three-year strikeout rate given that collection of statistics? The 20.8% put forth by Adam Jones. (Brennan Boesch qualified and put up a 21.3% strikeout rate last year but doesn’t have three years of data yet). Obviously, there’s some selection bias here- a player that reaches often, doesn’t walk, and has a poor strikeout rate might not survive long enough to show in a three-year sample. But there’s a long way from an average strikeout rate (around 20%) to a poor-but-passable one (around 30%), and none of these free-swingers ended up in that zone either.

It’s a tricky spot to be in – a player with contact problems may be best served (at least in the arena of strikeouts) by gripping and ripping it. An aggressive approach could help him lower his strikeout rate as we see here. Then again, that same aggressive approach will remove some of the player’s value by cutting his walk rate. Screwed if you do, screwed if you don’t?

(H/T to Paul Bourdett and Sky Kalkman for the twitter convo that inspired this post.)




Print This Post



Graphs: Baseball, Roto, Beer, brats (OK, no graphs for that...yet), repeat. Follow him on Twitter @enosarris.


30 Responses to “Lots of Whiffs, but where are the Ks?”

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

    i really think this effect partially explains the performance of ryan howard this season. i had looked it up in the middle of the season, not positive if the results remained at the end of the season.

    according to zone%, he saw alot fewer good pitches this year. and yet his walk rate was a career low, as was his % of 2-0 and 3-1 counts. i don’t think its a stretch to say he was swinging at more bad pitches this year. the question for me is why. are people simply learning to pitch to him and exploiting his weaknesses such that he can’t recognize pitches as well as he did in the past. or (the more interesting option) is it a conscious effort on his part to cut down on his oft hated strikeout total. if that is the case, someone in the PHI org needs to tell him that tradeoff is not worth it. he’s not finding hitters counts and his power has suffered more than enough to outweigh any benefit less strikeouts.

    Vote -1 Vote +1

    • John says:

      Over the offseason Ryan Howard worked on his batting stance so that he would be closer to the plate, hoping that it would prevent him from getting beat on pitches low and away and that he could become a better contact hitter. This might explain some of his problems this year. Standing closer might have messed up his strike zone. That combined with his efforts to be a better contact hitter might explain why he swung at more bad pitches.

      Vote -1 Vote +1

  2. Locke says:

    Constructive criticism:

    This article could be summed up in 1 sentence, and it still wouldn’t have much impact. I think we can all guess that guys who swing early in counts, even if they whiff more than most, aren’t going to put up the highest K rates. And I’m sure we could’ve guessed who the culprits were, too (Frankie and Vlad immediately came to mind.)

    What would’ve really been interesting to me is if you had dug another foot deeper, and looked at pitchers.

    Many of us look to SwStr rates for pitchers as an additional measure of predictive K ability. How much does an aggressive/contact pitcher hurt his K rate by attacking hitters despite a high SwStr. And then identify some pitchers who potentially are using improper game theory by attacking the zone too much, and take a whack at projecting what their K rate would or should be according to their SwStr %, and how many walks they might add if they did.

    If you’re going to write an article without any actual analysis, just spewing stats onto a page, you better actually have something to say along with it that’s interesting to read. This was almost as ill thought-out as Remington’s Winter Classic article. Maybe you should be more judicious in your twitter convo inspiration.

    -9 Vote -1 Vote +1

    • R M says:

      Very interesting concept for an article….but isn’t that a completely different idea from this one? I don’t see the need to rip this one apart.

      Vote -1 Vote +1

      • Eno Sarris says:

        Different article, but perhaps another interesting way to discuss the relationship between SWSTR% and K%.

        .. as for impact, I’m not saving the world here. I actually found this to be a surprising discovery.

        Vote -1 Vote +1

      • Jeffrey Gross says:

        Eno,

        Check my comment below. I did some of the number crunching for it already

        Vote -1 Vote +1

    • R M says:

      Plus, if an article CAN’T be summed up in one sentence, that means it lacks focus.

      Vote -1 Vote +1

    • Albert Lyu says:

      FWIW, I thought this presented a pretty groundbreaking idea that I haven’t really realized. Goes to show how hackers with high swstr% can still ‘work the count’ well (or low swstr% and work the count poorly).

      Vote -1 Vote +1

      • Locke says:

        Goes to show how hackers with high swstr% can still ‘work the count’ well

        Albert, as one of the sharper FG writers I thought you’d kind of box this idea up quicker, also you I think you may be confused, as your statement above is not what he talked about. You certainly don’t walk more by simply having a higher SwStr%, all he said is they can avoid K’ing by swinging early.

        (See my other comment below)

        There are two major factors that determine how often a hitter strikesout:

        - SwStr%
        - Plate discipline

        All this article said was… that the latter existed as a factor. Clearly you can avoid striking out by swinging earlier, no matter how bad your SwStr% is. It’s… ridiculously simple.

        Surprising twitter discovery!!!!!!!!!

        Vote -1 Vote +1

  3. Ben says:

    It’s a tricky spot to be in – a player with contact problems may be best served (at least in the arena of strikeouts) by gripping and ripping it. An aggressive approach could help him lower his strikeout rate as we see here. Then again, that same aggressive approach will remove some of the player’s value by cutting his walk rate. Screwed if you do, screwed if you don’t?

    It seems logical to me that you should only try to be more aggressive if your ISO exceeds your walk rate. The extra bases you get from putting balls in play seem like they should be more valuable than the lost walks. The interesting thing at that point would be: does ISO drop as batters become more aggressive?

    Vote -1 Vote +1

    • Ben says:

      Rats, I meant to quote that first paragraph of my comment.

      Vote -1 Vote +1

    • Eno Sarris says:

      That is interesting, the relationship of ISO to walk rate surrounding this swSTR/K% discussion.

      Also: Albert Lyu asked if Reynolds found himself in more two-strike counts than Hamilton. Answer: yes, he does (4/9/16/37% for Reynolds in two-strike counts to 4/6/12/33% for Hamilton in 2010).

      Vote -1 Vote +1

    • Dan says:

      “It seems logical to me that you should only try to be more aggressive if your ISO exceeds your walk rate. The extra bases you get from putting balls in play seem like they should be more valuable than the lost walks.”

      I think your point assumes that taking a free swinging approach reduces walks and strikeouts in a proportionally. That may be the case for some players, but intuitively, it would seem that is not a linear relationship. Even if it is, the amount of benefit depends on the slope of the line…

      Also, with men on base, a single is more valuable than a walk.

      Vote -1 Vote +1

      • Chris says:

        But weak contact in that situation could lead to a double play, which is worse than a BB or a K. The concept here seems to be to shorten up the swing and just try to make contact with anything once the count reaches two strikes, and that makes sense for a guy who doesn’t walk a lot leading off an inning, or a more of a power hitter with a guy on second or third (even a deep AO gets the runner home) but with runners at the corners, runners at first and/or second? Not so much. Poor contact on a curve or a sinker could mean the inning is over.

        Vote -1 Vote +1

      • Ben says:

        Chris: I would think for higher ISO players the really good outcomes (extra base hits) would probably balance out the really bad outcomes (DPs). I have no hard data to back that up though. I think that because power hitters hit into DPs relatively frequently and are still very valuable players.

        Vote -1 Vote +1

  4. BillWallace says:

    Tough crowd, I think it’s an interesting insight

    Vote -1 Vote +1

  5. James says:

    That’s not constructive criticism, Locke. It’s you being a dick. Your comment was almost as ill thought out as your birth.

    Vote -1 Vote +1

    • DIVISION says:

      You settle down, James.

      Locke is correct in his downplaying the article.

      Free-swingers will inherently have lower K-rates.

      It’s actually common sense and I don’t see the point of writing an article for a concept that should have been summed up in one sentence.

      I concur w/ the Lockester……..

      Vote -1 Vote +1

      • Locke says:

        Thanks. Didn’t mean to go all douchebag about it, which I did, but it’s pretty simple stuff. There are several things that affect how much you strikeout…

        - pitch selection/patience/swinging early in counts
        - how often you swing and miss.

        All this article said is that you can be really bad at one of these, and good at the other, and strikeout an average amount.

        Duhhhhhhhhhhh

        Vote -1 Vote +1

  6. Jeffrey Gross says:

    Eno,

    Here was the relationship I found between Whiff and K for pitchers:
    http://www.hardballtimes.com/main/fantasy/article/the-sum-of-their-parts-a-deeper-look-at-peripheral-data/

    The analysis may be a bit flawed, but the relationship b/w SwStr% and K/9 (which should have been K%…) seems quite robust.

    Vote -1 Vote +1

    • Locke says:

      Now this is great stuff. Kind of shines the light on how shitty some FG stuff is compared to throwaway (no offense) articles on THT.

      Vote -1 Vote +1

      • Eno Sarris says:

        I was with you until this one, seriously. Didn’t think I was aiming super high with this one, just pointing out a tidbit I thought was interesting. Through the self-admitted douchebaggery I saw a legitimate critique.

        But I think you kinda jumped the shark here. Jeffrey graphed the correlation between K% and SwSTR%, I talked about the outliers on that graph from a hitting perspective. I fail to see a big difference here, unless it’s the graphs. It’s just outliers for one and main correlation for the other.

        Just as you could ask more questions after mine – like how the free-swingers fare with two strikes, or how their ISO fares when they are more / less selective, or how much value they lose when they eschew the walk to avoid the strikeout – you could ask more questions after Jeffrey’s piece – like why some pitchers have high swSTR% and low K%, or how they could use game theory or pitch sequencing to take better advantage of their raw stuff. I’d say both are good jump-off pieces.

        Vote -1 Vote +1

  7. Paul says:

    I’m sorry our Twitter conversation was beneath you, Locke. But I think the comments section here (save for James — although that was a doozy) proves you wrong in more ways than one.

    Not only is the ratio of people supporting/discussing the topic to people bashing the topic/author at least 10:1, but the fact that this post inspired just a single person (you excluded) to engage in the discussion proves its worth.

    You should be more judicious with your words. Your last paragraph took you from intelligent debate to peanut gallery.

    Vote -1 Vote +1

    • Locke says:

      The profundity of the idea that swinging earlier in counts makes you strikeout less is overwhelming. 140 characters was I needed to figure it out, sorry you needed more.

      Vote -1 Vote +1

      • Someanalyst says:

        Children… I guess we can only really love our own…

        I hope you’re a teenager Locke and that you still have engineering school ahead of you to provide some intellectual polish.

        Vote -1 Vote +1

      • Paul says:

        No need to apologize.

        By the way — the idea wasn’t that swinging earlier in counts makes you strike out less. The idea was that swinging AND MISSING a lot doesn’t necessarily mean you strike out a lot.

        Reading comprehension obviously isn’t your strong suit. Try reading past 140 characters next time.

        Vote -1 Vote +1

      • Locke says:

        You still don’t get it. That’s cool. Baseball is hard.

        Vote -1 Vote +1

  8. wily mo says:

    every early-january baseball blog post definitely needs to be a massive game-changer or i walk

    Vote -1 Vote +1

  9. Jimbo says:

    Eno, first off…LOVE your mock results on RotoHardball. Dunn, Kinsler, Tulo, Arod, Kemp, Cruz, Victorino in a competitive 12-team draft?? I’d take that in a heartbeat, despite my loathing of Arod.

    I pulled a bunch of stats into xl, but in the end couldn’t get what I was lookinng for…swinging strike % after a x-2 count. Compared strikeout %, but that doesn’t necessarily mean they struck out swinging. Interesting how ISO changes for some and not others. Hamilton is awfully dangerous with 2 strikes!

    Would ‘ability’ to foul pitches off come into play? How would it be captured if Howard fouls off 2 pitches before swinging through one…versus Reynolds who may not foul many off at all. Might get into the whole 2-strike approach. Maybe some sluggers swing a little less hard?

    Vote -1 Vote +1

Leave a Reply

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

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