Advanced Metrics

Good Hitters Strike Out More Often Than Bad Hitters Do

Say what? Strikeouts are bad right? It seems like striking out is the worst thing you can do at the plate*. Striking out is an embarrassing failure. So how can it be true that good hitters strike out more than bad ones? Don’t believe it? I’ll try to prove it to you.

*Striking out is not actually the worst thing a hitter can do. Hitting into a double play (or triple play!) is far more damaging than a strikeout. Hitting a ball that causes a runner to get thrown out at the plate is also worse than a whiff. I’m sure there are others as well. If you can think of them be sure to let us know in the comments section below.

When planning out this article I thought of two ways to test my hypothesis that good hitters whiff more than bad hitters. The first way is to sort all the hitters by their strikeout percentage (K%), then look to see if the high K% players have better or worse overall batting lines than the low K% players. The other way to test it is to sort all the hitters by their wRC+ (the best measure of hitting prowess) to see if the high wRC+ hitters whiff more or less than the low wRC+ hitters. Let’s do it!

Round 1

I wanted to avoid a small sample size that could distort the numbers, so instead of using the first 3 weeks of stats from this season I decided to use stats from the full 2015 season. I included every hitter who had 300 or more plate appearances last year. 268 hitters reached that threshold last season. The average K% for these players was 19.2% and the average wRC+ was 103.5*. I divided those 268 players into two halves: a High Whiff group and a Low Whiff group. The 134 players with higher than average K%’s went into one group, and the 134 hitters with lower than average K%’s went into the other group.

*In case you’re wondering, the league average wRC+ is 100. Each point above 100 indicates the player is 1% better than average. Each point below indicates 1% below average. The reason that the average of my sample was 103.5 is because a player has to be pretty good to be given 300 or more plate appearances in a season. Bad players are bench-warmers or get sent to the minors before reaching 300 PAs. So it makes sense that the guys in my sample are better than average as a group.

High Whiff Half Low Whiff Half
K% 23.8% 14.6%
BB% 8.3% 7.3%
AVG 0.255 0.271
OBP 0.322 0.328
SLG 0.433 0.410
OPS 0.755 0.738
wRC+ 105.4 101.6

The high K% group averaged a whopping 23.8 K%. The low strikeout group whiffed at a 14.6% rate on average. That is a BIG difference. So which group actually hit better? The high whiff group posted an average 105.4 wRC+ while the low strikeout group put up an average 101.6 wRC+. The whiffers hit better!

The low whiff group posted a higher batting average, but we as educated fans here on Redleg Nation know that batting average is an extremely poor and misleading way to evaluate hitters in modern baseball. The one surprise for me is that the low whiff group had a slightly better OBP as well. Generally speaking, players who strike out a lot also walk a lot. The high whiff group did walk more often than the low whiff group last year (8.3% to 7.3%) but in most seasons the difference is greater. Last year was a bit of an anomaly in that regard. Striking out and walking go hand-in-hand to a large degree. Hitters who go deep into counts tend to both walk more and strike out more. Both require seeing multiple pitches. If you hit the ball early in the count you won’t strike out but you won’t walk either. If you want to be a star hitter you have to walk a lot (or else you won’t have a high OBP) and you have to hit the ball hard (or else you won’t have a high SLG). Making contact isn’t the goal. Doing damage and producing runs is the goal.

Where the high whiff crowd pulls away is in slugging. The whiffers hit the ball much harder than the non-whiffers. It makes sense, the harder you swing the less likely you are to hit the ball, but when you do hit it you tend to do some real damage. Hitters who try to maximize contact tend to be the smaller, skinnier guys who can’t hit the ball very hard, so they try to compensate by at least hitting it more often.

All things considered the high whiff batters out-produced the low whiff hitters. It wasn’t a huge difference, but it was certainly a statistically significant, real difference. Striking out doesn’t make a hitter better, but it doesn’t prevent them from being better either. Don’t worry about the strikeouts, worry about the wRC+.

Fun Fact: Between Adam Dunn (career batting average .237) and Future Hall of Famer Ichiro Suzuki (career batting average .314), which batter was more likely to make an out each time he came to the plate? The answer is Ichiro was more likely to make an out. Ichiro’s career OBP is .356 compared to .364 for Dunn. Batting Average tells you Ichiro was the better hitter, but nearly every other hitting metric favors Dunn, most of them strongly. Ichiro’s career .406 SLG, .760 OPS and 105 wRC+ pale in comparison to Dunn’s .490 SLG, .854 OPS and 123 wRC+. Yet 95% of fans are convinced that Ichiro was a better hitter than Dunn. That is how behind the times most fans are even in the Information Age. Of course Ichiro was a much, much better fielder and base runner than Dunn. You can make a strong argument that Ichiro Suzuki was a better all-around baseball player than Dunn but the Big Donkey was clearly the better hitter.

Round 2

The second way to make this evaluation is to use the same 268 hitters as in the high/low whiff group. This time we will divide them into halves based on their wRC+. The 134 hitters with the highest wRC+ vs the 134 hitters with the lowest wRC+. Which group struck out more?

High wRC+ Half Low wRC+ Half
K% 19.4% 19.0%
BB% 9.0% 6.5%
AVG 0.276 0.250
OBP 0.348 0.303
SLG 0.467 0.376
OPS 0.814 0.679
wRC+ 122.7 84.3

The table shows the 134 best hitters averaged an excellent 122.7 wRC+ compared to the puny 84.3 wRC+ put up by the 134 lesser hitters. Huge difference there. The difference in K% was small but the good hitters did strike out a shade more often than the poor hitters. Not surprisingly, the good hitters smoked the poor hitters in all of the offensive metrics, especially slugging percentage and OPS. Once again, strikeout rate did not prevent hitters from hitting at stellar levels.

Scenario: You get to manage one major league baseball game. One team’s batters strike out a lot. The other team’s batters rarely strike out. If your team wins the game you get a million dollars. If your team loses the game you get a year in maximum-security prison. Which team would you choose?

If you asked 100 random baseball fans this question I bet 95 of them would choose the low strikeout guys. They would most likely end up in the slammer.

Conclusion

I think the moral of the story is that strikeouts in and of themselves are not a bad thing for hitters. You can’t look at a player who strikes out a lot and condemn him as a bad hitter. On the flip side you can’t look at a hitter who rarely strikes out and say he is a good hitter. Strikeouts are not an effective way to sort good hitters from bad. Many hitters who frequently whiff are actually extremely productive batters. Generally speaking, major league hitters who strike out a lot tend to be more productive hitters than high-contact hitters are (see the first table above). Strikeouts are just another form of out. To identify the best hitters we need to look at how many outs they make, not how they make their outs. We need to look at how much damage a batter does rather than focusing on his strikeouts.

Fun Fact: Babe Ruth held the all-time record for career strikeouts for 35 years. Then Mickey Mantle passed him and held the record for 14 years. Willie Stargell passed him and held the record for 4 years. Reggie Jackson then stole the record and has held it for 34 years. All four of those players are inner-circle Hall of Famers. How bad can strikeouts really be when some of the best hitters ever to play the game struck out the most?

 

44 thoughts on “Good Hitters Strike Out More Often Than Bad Hitters Do

  1. Good read, Nick. Thanks.

    I wonder if we looked closer if we could find some sort of break-point or asymptote where increasing strikeouts actually started becoming a major problem. I’m thinking back to Javier Baez, who famously had a 41.5% K rate in his first taste of the majors. No one doubts his on-contact authority, but the 41.5% K rate was so high is actually limited his production to the point where he was sent down. Perhaps there is a point where too much is too much. Might be an interesting follow-up! 😉

    • I think hitters who run k rates north of 30% rarely succeed in the bigs. There are some outliers, but you’ve really got to hit the carp out of the baseball when you hit it, and walk a lot.

      The Reds offense, by the numbers is very interesting this year. They’ve got the 5th best contact rate. And they are making good contact: 1st in line drive rate, lowest soft hit %, and 9th in hard hit rate.

      So why is their offense poor? Well, they’re 22nd in BABIP and 29th in walk rate. The low walk rate is partially explainable as pitchers are pounding the zone against them: 2nd in zone % and 1st in first strike%.

      But the BABIP? Well, the Reds may have become predictable in their batted ball profile. 13th in wRC with no shift on. 29th in wRC against a shift.

      • Oh, absolutely. I’ve been harping on Schebler for a while because of his contact rate (and related K-rate) and noticed that the worst guys last year on contact were Kris Bryant and Chris Davis. So basically, you need supreme on-contact power if you’re going to swing and miss that often and K 30+% of the time.

    • I was thinking that too. I think it might be interesting to break hitters into deciles based on K% and look at that. I suspect the best hitters might be in the 3rd or 4th highest deciles (of course, some selection bias based on PAs could impact this as well – if you’re striking out like Drew Stubbs (2015: 42.9% K rate in 140 MLB PA, 2016: 42.1% K rate in 38 MLB PA), or Baez in your example, you’re not going to get major league playing time for long and thus won’t meet the PA threshold

  2. Interesting analysis Nick. We’re slowly learning to appreciate what matters in all those stats that come with modern baseball and this is a compelling if slightly counterintuitive reality. The peril is that old correlation vs. causation problem. Striking out more doesn’t make a hitter better, it just comes with the territory of being a good hitter. I’ll bet there are some fascinating extreme examples at both ends of the spectrum. The Big Kluszewski comes to mind as a likely beast on the rare low strikeout / high WRC+ list. Thanks for the interesting read.

    • Feeling Skeptical, I see no significance in the K% with high wRC+ vs Low wRC+
      Naturally good hitters are going to get more AB to qualify & strike out more and have a higher OPS than good hitters that do not K as much. But what about all the awful hitters who never get a chance to meet your AB?
      I still am trying wrap my head around the BABIP evaluation that if a hitter is a high range (.330 or so)- it is unsustainable as he is lucky as opposed to a low (say .270) he is just hitting into back luck. Seems to me if you just put the ball in play you would be a near .300 hitter- so do not K and try to foul off enough pitches to walk.

      There are so many flaws in the You get to manage one major league baseball game Scenario, but it does make an interesting debate.

      Curious if you broke these into OPS or OBP to see the correlation to wRC+ and then compare the K%.

      • From my Mesoraco article, K% is the stat with the 2nd highest correlation (.32 r-squared) to HR/FB%, and HR/FB% correlates well to wRC+ since homers are the single best thing a person can do at the plate.

        I know that’s not exactly what you asked, but thought it was an interesting snippet, nonetheless.

        • I am not following this, I would expect weight would also correlated with HR/FB. I would think targeting a guy who weighs a lot as you have hopes he turns into a power hitter would be a terrible strategy but would be much better than targeting a hitter who strikes out a lot.
          Possibly there is another point than striking out does not indicate a poor hitter- I will agree. It is somewhat interesting that it is a rate and there is a correlation as I tried to find something comparable like % int thrown and it seems to have minimal correlation- Neil O’Donnel is 4th all time behind Rodgers, Russell Wilson, Brady while in the 100th range Stabauch and Jim Kelley are tied with expected guys like Kitna, Billy Joe Tolliver, Sanchez and Mirer.

          Would be curious to see how this correlates to other positive stats besides power that are indicators of a good hitter.

  3. Worse than striking out: Hitting into a fielders’ choice, which erases a speedy runner and replaces him with a guy who’s nickname is “Mo”, short for “Molasses”. Actually, hitting into one that erases any baserunner from the same base that a now slower baserunner occupies.

      • He could swing it a bit! A lot better than me (and that’s who I was actually referring to). Yeah, big Mo could hit ’em.

  4. Worse than striking out: If no one is on base and trying to increase the starter’s pitch count, then every kind of out that uses up fewer pitches is worse than striking out.

    • The times through the order penalty suggests that taking pitches just to run up a starter’s pitch count might not be a great strategy.

      I guess it just depends on how good your opponents bullpen is. Getting into the Reds bullpen is definitely a good thing for their opponents. But a team like the Yankees or Royals? If you’ve just run up the pitch count without doing much damage, then I’m not sure you’ve helped yourself by getting to those pens.

      • Definitely depends on quality of the bullpen. With typical power-arm bullpen, the pitch count stuff isn’t as important as we thought 5-10 years ago.

    • Excellent point, Steve. Nothing drives our SABRE loving friend, Thom Brenamann, than a first pitch out vs a pitcher who is laboring.

  5. I love this article Nick. This is such important information for the “thinking fan.” My favorite part is definitely when you show that Adam Dunn was a better hitter than Ichiro. I could just see a certain Reds radio voice reading that and spitting up his coffee. Great stuff.

    • Ichiro is still playing at 42 years old, though. The last several seasons have really driven down his over wRC+. If he’d have retired at the same time as Dunn his wRC+ would be right there in the low 120s with Dunn.

      Point taken, though… Dunn was criminally under-rated as a hitter because of his other deficiencies and aptness to strike out.

  6. There is the GABP factor too when it comes to comparing other players to Reds. The ball doesn’t carry in Safeco whatsoever….didn’t Votto/Dunn/Junior score 1 run there one weekend years ago? Ichiro prob hits 15 per year as a Red. Now Big Donkey was a unique dude so I doubt GABP aided him much when he connected but I have no doubt whatsoever that a guy like Rizzo would hit 40+ as a Red….and I realize our pen wouldn’t be pitching to him in that case.

    • wRC+ actually factors in ballpark. It’s based on how that ballpark played for all hitters vs other parks. It’s not perfect but it’s pretty good.

  7. I had forgotten that Reggie is still the King of K’s. It is amusing comparing Reggie and Barry Bonds K’s and BB’s. They are quite close when flip-flopped. Bonds has just about as many BB’s as Reggie does K’s. And Bonds has just about as many K’s as Reggie has BB’s.
    wRC+. Just goes to show you how advanced Joey Votto is. I’m reminded of the time in 2012 when one of the Cincy writers asked Votto what he thought the most important stat was and votto replied, “wRC+.” Not the answer many people were looking for and were confused. Just goes to show you.

    • Yep!

      And at the start of the season, Votto’s 157 wRC+ ranked tied for 10th in MLB history (min PA = 4000).

      Here are a few players right below Votto on the list: Frank Robinson, Willie Mays, Hank Aaron.

      Of course, Votto has to age… so if he can retire around 145 wRC+, that’ll put him right on the edge of Top 30 all-time. Still pretty dang good. Just about every person on that list is in the HOF.

  8. A lot of hitters are reinventing themselves by either swinging to make hard contact or changing their swing to add more loft. Most of these hitters are obtaining power at the cost of striking out more, but are offering more offensive production overall. Some names off the top of my head are: J.D. Martinez (he has made a further adjustment and lost some power?), Matt Carpenter, Stephen Piscotty, and Justin Turner. Mesoraco’s ’14 season is another example of this.

    There are a few more guys that look like they could fit this profile this year, including Ryan Gennett and Sean Rodriguez. However, the sample size is small and it is too early to tell.

    Also, Miguel Sano is an incredibly interesting hitter. Very high K rate with fewer balls in play, but he hits the ball so hard that he makes up for the high K rate…also with an insane BB rate. He’s a lot like Stanton except he actually hits the ball harder more often and walks more (so far).

  9. I don’t doubt that your data is accurate, and that home run hitters do strike out more, but that doesn’t say that all hitters that strike out more are better hitters, I’d like to see the same data for maybe Pete rose, or some of the other prolific great contact hitters who you certainly wouldn’t rule out of having on that team that might keep you out of jail.

    • If staying out of jail is your goal, then Pete might not be a great pick.

      Sorry, couldn’t resist.

  10. Looking for walks seems so Little League for your middle of the order guys. Leave to the top and bottom of the order.

    • I don’t think anyone has said a player should “look for walks”. Walks are a byproduct of not swinging at bad pitches. You take your walks if you don’t get good pitches to hit. A good hitter will have an approach like this:

      0-0 or ahead in the count – Look for one pitch in one zone and swing if you get it.
      Behind but less than 2 strikes – Look for one pitch or one zone and swing if you get it
      2 strikes but not 3-2 – Plate protection, swing if there’s a good chance it may be a strike.
      3-2 – You might take a boarderline pitch here that you you’d swing at 2-2 because it may be called a ball and then you get on base. If you put it in play, there is likely less a chance you’ll get a hit (about 30%) than if you take it (50/50 shot).

      Does that make sense?

      • Good post. My son is in Little League right now. We always have those conversations where we discuss what count to only swing at “his pitch”, versus when he needs to swing at anything in the zone.

        • That was basically my approach as a hitter. It worked very well right up until the point that I just wasn’t talented enough physically to keep up with the competition level I was at.

  11. No question that there are things worse than striking out. With that said, I would have to believe that the most difficult aspect of playing baseball is actually making contact with a bat on a thrown ball. Anyone could stand there and strikeout, they don’t even have to move their bat. Same with taking a walk. I firmly believe that both strikes and walks are more of an attribute of the pitcher than the batter. Now, actually making contact that is all skill from the batter.
    For me that is the thing wrong with some of the modern metrics. They over value the walk as an attribute of the batter when I would imagine that in reality at least 50% of the walks were an inability of the pitcher to throw strikes. For starters I would love to see all 4 ball and 4 ball and 1 strikes removed from the batter as a credit towards the batter doing something as a positive stat. Another thing to consider would be breaking down the total # of pitches thrown versus number of pitches that the batter actually swung at that were balls.
    Additionally I don’t think that other outs that the batter makes is conversely counted as being a negative. If the batter makes any out that leads to more than one out that should be a substantial reduction in the value of that at bat attempt.

    • The analytics have always over valued walks. And strikeouts as well. Analytics reduces data into static events. Its failing is measuring dynamic events. An out, of any kind, is the root of all evil and a walk is more precious than gold. Information analysis is a great thing but not the thing.

      • A walk is worth .69 runs, a single is worth .89 runs. Those numbers are based on measuring the real dynamic (not static) impact of those events at every major league game.

        “Analytics” doesn’t claim a walk is worth more than a single. But walks are *much* closer to the value of a single than an out. That’s why OBP is much more closely correlated with runs scored than is batting average.

        • not debating that I agree with all that you have said. OBP is a better measurement of “value” than batting average. but I do think that a batter with a higher batting average than another with a similar number of plate appearances has overall better skill regardless of what the OBP is. Hitting a baseball is the hardest job in all of baseball. If you can do it 3 out of 10 times and manage a base hit you are an all star. If you can do it 4 out of 10 times you become hall of fame material. Nobody who has ever played the game has even come close to 5 out of 10 times that is how difficult a task it is to hit a ball.

        • Lots of guys hit the ball 3 out of every 10 times at bat. But if you’re making poor contact (lots of weak grounders, low line drive rate), you’re not going to get a BASE HIT.

          And there is the crux of the issue with the recent Joey Votto debates. Some folks want Votto to swing at borderline pitches instead of taking them, and it seems like they assume that if he gets his bat on the ball that it means a good outcome. But it doesn’t work like that, and most sane folks would have to admit that swinging at borderline pitches would be less likely to result in a hit than swinging at a better pitch.

    • If walks were almost always dependent on the pitcher, then all hitters, every hitter, would have close to a league average walk rate; because every batter in a league and especially within a division, face mostly the same pitchers. A hitter with a good batting eye, who doesn’t swing at many bad pitches, has a higher than average walk rate; often much higher. A hitter that swings at pitches outside the zone or borderline pitches has a lower than average walk rate. If all batters were roughly the same, this would never really happen.

    • Taking walks involves skill from the batter because an unskilled batter won’t recognize which pitches to leave alone, if they aren’t obviously out of the strike zone. No question that it’s hard to hit the ball, though. First- hand experience speaking here.

  12. Something bugs me about this whole scenario? With the roid era over….more and more teams are looking for more offense at the expense of defense (Suarez, Schwarber in LF, etc. etc) but Suarez can’t boot one if you K? The Royals have done pretty well by sacrificing some power for contact….diff ways to skin a cat! Not to mention I would assume that the contact guys run much better than the high K guys (on average) so therefore would have that advantage offensively. All that being said…GABP is a launching pad so Suarez’s lift and separate for some flyballs is ok by me and if he Ks a little more then so be it? You play to your park if you want to make the playoffs!

  13. I think there’s some confusion among some of the commenters on what the article is saying. I read the article as saying that good hitters are more prone to striking out often than poor hitters are. What I’m not reading is that all high strikeout hitters are good hitters. I’m also not reading all good hitters are high strikeout hitters.

Comments are closed.