Baseball Musings
Baseball Musings
April 30, 2004
Strikeouts

Bryan at Against the Grain brings strikeouts into the unproductive outs argument.


Strike outs are the worst thing a hitter can do, because even if you ground out weakly or hit a soft-fly you have somechance of getting on-base. Watch a game and see how many jam-shot singles there are. Not every hit is a line-drive. Strike outs are an evil byproduct of drawing walks, because just by nature the more pitches you see the more times you're going to hit with two strikes. So strikeouts aren't as bad as old-school baseball guys may have believed, but they are still probably a lot worse than many SABRmetricians or SABRquixotics or whatever you want to call them believe.

The thing that's bothered me for a number of years is that strikeouts appear to matter much more to a pitcher than a hitter. In other words, high strikeouts are one indicator that a pitcher is very good, but high strikeouts are not an indicator that a batter is very bad. I believe the basic reason for this is power; it's hard to hit HR off good pitchers, but players who strike out a lot tend to hit for power, too. As I've written before, I believe there are situations when trying to put the bat on the ball is much for important than trying to hit one out of the park. But in general, it's okay for hitters to swing hard, even if they miss a lot, because the results when they don't miss are so impressive.


Posted by David Pinto at 03:27 PM | Strategy | TrackBack (0)
Comments

Would there be any research to support a manager pinch hitting a contact hitter for a power hitter who strikes out alot, say down by a run, ninth inning, man on second, two outs? A strike out is the absolute worse thing that could happen in that situation. In the long run you want the power but in individual instances, you want the ball in play and take your chances. I wish sabermetrics would move more into this direction of studying the best play for each situation. That is where the stolen base question always comes down.

Posted by: Scott at April 30, 2004 03:59 PM

I'd say the worst thing a hitter can do is ground into a double play, myself.

Posted by: Yuda at April 30, 2004 04:10 PM

The problem I have with productive outs is the same problem I have with over-patience: it assumes a certain kind play is categorically "positive," without any context. I just don't think that's the case in baseball. Every play in baseball is context-sensitive, and to jump to conclusions about any of it betrays what I think is a lack of a basic understanding of baseball.

On the patience side, at some point too much patience will kill you, because pitchers will stop respecting you. After a while, they'll start throwing strikes, and if you can't make contact and burn them, you're finished (see: Giambi, Jeremy). So not every patient hitter is productive for his team.

On the productive out side, let's say you have Juan Pierre on second, Alex Rodriguez at the plate, James Baldwin pitching, and spaghetti-armed Bernie Williams in right. Do you want A-Rod bunting to the first baseman to move Pierre over to third? Or would you rather him crank one out to right and dare Bernie to throw out Pierre at the plate. Obviously the latter. But under the "productive out" paradigm, bunting Pierre over would show up as a positive on the scoresheet, even though it was a ridiculous move at the time.

I think that same logic holds true to strikeouts and double plays as well. First and third nobody out, bottom of the ninth, tie game. A double play (6-4-3) is a better play than a strikeout, because it wins you the game. If there's one out, the strikeout is better, because it keeps the inning alive.

The point of all of this is that I think Dave is right: this is about politics. They lump together a bunch of plays that have no business being together and call it a stat, all with the aim of debunking another point of view. I prefer the power/patience approach because it has less downside to it (there aren't many Jeremy Giambi's out there). But I know there are flaws to the approach. This dumb stat takes a different approach (which is fine with me), but completely ignores the enormous flaws it contains.

Posted by: Daniel Shamah at April 30, 2004 04:46 PM

The issue is not the relative value of Ks to other outs for pitchers, but for their predictive value. An out is an out, but guys that strike people out can be relied on to get more hitters out in the future.

The Jeremy Giambi example is dead on and you may question the wisdom of not throwing meatballs to Mark Bellhorn (.194/.402/.323). The other example (Yuda is dead on) I would cite is how Red Sox fans would root for Tony Clark to strikeout, in order to stay out of the DP.

Posted by: dg at May 1, 2004 07:17 AM