Productive Out Percentage is Crap

ESPN has concocted a new stat, which Buster Olney defends with a bunch of anecdotal evidence on ESPN.com.

The problem with Productive Out Percentage is pretty easy to spot right away, but let’s state the obvious:

What is a productive out?

A productive out occurs when …

  • A baserunner advances with the first out of an inning.
  • A pitcher sacrifices with one out.
  • A baserunner is driven home with the second out of an inning.

What is the formula for productive out percentage (POP)?

Productive outs divided by the total number of outs. For instance, if three of Player A’s 10 outs are productive, his POP is .300.

POP is pretty easy to dismiss, because here’s what it doesn’t measure: rate of success versus opportunity to have that success. Go back and look at the italics above, and then read on.

Look again at the three conditions that produce a Productive Out [PO]:

  • A baserunner advances with the first out of an inning.
  • A pitcher sacrifices with one out.
  • A baserunner is driven home with the second out of an inning.

All three conditions require a baserunner.

PO’s are as team-dependent as Runs Batted In [RBI] are. RBI’s are a function of two things: runners on base and the batter’s ability to drive them in with a hit. The two factors driving it are, then, OBP(team) and SLG(player), since extra-base hits are more likely to drive in runners than singles. Remember: more bases are better, because more bases get you closer to home plate, no matter how many runners are on base.

If you are a great slugger—say, Barry Bonds—and don’t have a ton of runners on base, you can’t get a lot RBI. Barry Bonds was unquestionably baseball’s best hitter in 2003 [and 2002, and 2001, but I digress], but as you can see, Bonds wasn’t in the top 40 players in terms of RBI in 2003. Why does this happen? It’s entirely due to Bonds’s teammates not being very good at getting on base in front of him.

I am not going to say that there is nothing to the PO concept itself; I do want to lambast the POP concept totally, because it divdes PO’s by total outs, regardless of whether or not there were runners on base at the time. If they reformulated POP to be POmade/POpotential, you might be able to make a case for an ability being present.

Of course, you would have to see that said ability was a “true” ability–that is, one that is demonstrated from year to year. If there is a strong year-to-year statistical correlation between a player’s ability to make Productive Outs in Year N and Year N+1, then ESPN may be on to something.

Instead, they have formulated a “stat” that’s as useless as RBI. All that serves to do is muddy the water.

Thanks a lot, fellas. All that money, and all that ability to have great resources at hand, and what do you do with it? Bupkis.


Alex pointed out that I didn’t discuss home runs or sacrifice flies when discussing RBI. I should have, given that I brought up Bonds. You do get credit for an RBI when you make both, but only one is easy to track in rate stats–the homer.

This reminds me of a study I started back when I was in college, but never finished: Actual Runs Created Percentage. The goal was simple: for each plate appearance a batter had, note how many runners were on base at the time. The maximum number of runs that a batter may cause his team to score in any plate appearance is four–the team has the bases full, and the batter drives in all runners as well as himself. The minimum number of runs that a batter may cause his team to score is, of course, zero—he can make a “non-productive out” that doesn’t cause his team to score, or he can reach base in a way that doesn’t create a run.

I think my idea was to count up Actual Runs Created—that is, the number of runs produced by each batter’s at-bat—and try to find some mean, median, and mode for it. One way to do it would be to create a percentage: divide ACR by the number of plate appearances. [One uses PA here rather than AB’s because you do not get an AB for a walk, sacrifice, or hit batsmen, even though all three can produce a run.]

That would be a true rate stat; whether it would matter a damn is worth investigating. Perhaps it has been done; I have never looked to see.

6 comments

  1. First… you really are bored at work today.

    Secondly, I agree that it would make a LOT more sense if it were a POmade/POpotential number.

    I understand that the pitcher is asked to sacrifice on a regular basis (it just makes sense), but why do they get special consideration here? So this will be another stat for the NL pitchers that doesn’t really apply to AL hitters (see all batting stats for the most part)

    These are the guys who decided out of the blue to start calling it a ‘Walk-off Homer’ rather than a ‘Game-winning Homer’. Why? Because they’re ESPN and they feel like they can (and they’re mostly correct) set the way people watch and talk about sports.

  2. Yeah, I am bored at work [it’s a slow week, and my only job is to work on IT problems we’re having, of which I can only control what happens here, and all the problems are in Houston], but this has been pissing me off ever since I read Olney spouting this crap in ESPN The Magazine‘s Spring Training issue.

  3. Ok. I just read the article (imagine that, read the actual source) and it appears that what we both would have liked to see is true.

    “Through Monday, the Tigers led the majors in POP, at .430, with 37 productive outs among 86 made in those situations.”

    Why is it that the teams don’t play a more bunt n’ run style anymore? I remember when I first noticed baseball there were the guys like Henderson and Butler (Dodgers fan, so I’ve gotta get him) running all over the place. Is it just that the guys are so focused on the almighty dinger (and the steroids to make it happen) that they don’t want to mess up their precious multi-million dollar bodies by running the bases and helping the team rather than their own stat sheet?

  4. Well, but POP is either that or it’s not, and I’m not sure that it is. Either their definition is poorly worded or it’s a stupid statistic.

    Why do teams not play bunt-and-run? It’s pretty simple: those plays waste outs. In baseball, you get three outs an inning: each is precious. Those plays can hurt the team and also can hurt the player’s stats. Is there value in moving the guy over with a bunt? Sure! Is there more value in moving him over with a double? YES!

    You have to do a risk-reward analysis. Pitchers are asked to sacrifice a runner over because they aren’t very likely to get a base hit, much less an extra-base hit. Better hitters are less likely to have the value of the sacrifice be higher, and as baseball has shrunken the strike zone, the advantage has gone to the batter.

  5. I just think that you’d win more than you lose if you could consistently eek a run or two out of any given inning through use of all aspects of the game. Does that take away the chance of the big inning… sure. But if you’re in a situation where you have good pitching (see Red Sox this year from everything I’ve heard), a few runs is all you need, so why gamble it all going for the big inning. This would relate very closely to the idea in football that defense coupled with an offense that doesn’t make mistakes wins championships. But then again, I’ve always thought in baseball that extra-base hits are overrated. Everyone should just slap singles and let the runs start flowing in.

Comments are closed.