A question on strikeouts and home runs

by Andrew_G on Sep 25, 2009 11:09 AM EDT

I've been reading and hearing a lot in the past few days about Mark Reynolds and his strikeouts being not a big deal, which I agree with (how could you not? The guy has a 131 OPS+ and 125 runs created, so it all must be working), but it does seem counter-intuitive to me to suggest that strikeouts for batters aren't important when they are important for pitchers.

The most compelling case I heard was that his strikeouts are a result of increased bat speed, which also led to more home runs (43), so asking him to cut down on his strikeouts would cut into his homers, and then he wouldn't be that good.

If you take a look at, for example, 2008 Bobby Abreu who hit only 20 home runs but struck out only 109 times while compiling a 120 OPS+ and creating 119.5 runs, it seems to me like while Reynolds is more valuable, he isn't that much more valuable. My question then is could there or is there a "break even" point between strikeouts and home runs - that is, could you say "as long as you hit X homers per plate appearance, you can strike out Y times per plate appearance"?

Tweet Comment 3 comments | 0 recs |

Do you like this story?

Comments

I will take this challenge. Give me an hour or so

Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.

by Jeff Zimmerman on Sep 25, 2009 11:12 AM EDT reply actions

Here is my best guesstimate - maybe more later

I took all the players from 2008 that had 200 (and 400 PA, both were similar) and got their SO/HR ratio and their OPS+. The players with an OPS+ of 100 or more averaged a SO/HR ratio of ~5 while the players with an OPS+ of less than 100 had an SO/HR ratio of ~10. I played all the points and created a expotential regression curve and using the equation the SO/HR value at an OPS+ of 100 is ~6.5

This should be done with multiple years, but should give a general idea. Feel free to ask question because I am keeping it short because I have “work” to do.

Jeff Zimmerman - Protecting the world from RBI's and Wins from my mom's guest house.

by Jeff Zimmerman on Sep 25, 2009 11:42 AM EDT reply actions

I'm guessing you could do this with linear weights

Here’s the reference from Tom Tango. I’m using the highlighted 5 RPG environment, even though currently it’s a little less than that I believe.

If you wanted to take the average values of the linear weights, then the average HR is worth 4.45 times more than the average strikeout, going the other way, Of course, this is isolated from any other outs or positive offensive effects; Reynolds is not only striking out and hitting home runs. Still, you’d expect that with a strikeout total like Reynolds’ this season, he would have to have 47 HR to “even out” the runs, though producing zero runs is probably bad.

But again, this is isolated from all other versions of outs and positive effects, so you can’t really conclude anything. In addition, it would not have mattered all that much whether the outs were K’s or normal outs. If you make it so that Reynolds never struck out, the value difference would be about 2.5 runs better than he has with all these strikeouts. For Reynolds to improve his value, he needs to make less outs, not necessarily less strikeouts.

As for an explanation, obviously there are times that K’s are better than or worse than normal outs, but it all averages out to be around the same value. And yes, we don’t know whether changing his approach is going to make him a better/worse hitter because of its effect on his other outcomes.

Marlin Maniac, a Florida Marlins blog
Check me out at Beyond the Box Score as well.

by SFiercex4 on Sep 26, 2009 7:30 PM EDT reply actions