It seems appropriate that on the day in which George Bush was trying to sell the U.N. on the stability in Iraq—sorry, I know, too political—Major League Baseball decided to announce the creation of a new statistic. It's called the O Zone Factor and like the volatile gas for which it's named and which breaks down into simple oxygen when exposed, the basis for this stat crumbles as one exposes it to the air—how's that for stretching an analogy to the breaking point?

[O Zone] measures a team's success at scoring runners from second or third base, as well as its ability at preventing opponents from doing so.



The O Zone calculates the percentage of runners in scoring position (second or third base) that ultimately cross home plate. Take a team's offensive O Zone percentage, subtract its defensive O Zone rate, and the result is the O Zone Factor - a measure of a team's success with runners in scoring position on offense compared to when they are pitching.

Well, that sounds nifty, doesn't it? But does it mean anything? Here's MLB's justification:

While there are many team statistics which help explain a team's winning percentage and its playoff chances, recent history has shown that a positive O Zone Factor is important : from 1999-03, 34 of the 40 playoff teams finished the season with positive O Zone Factors (meaning they had more success batting than in the field with runners on second and third). Last year's World Champion Florida Marlins had the ninth best positive O Zone Factor and in 2002 the World Champion Angels had the best O Zone Factor of all 30 teams.

You just can't beat anecdotal rationalizations. So why try? Let's look at the basis for the O Zone Factor—Sounds like a new reality show, right?

The basic idea is that we take the offensive and defensive ratios of runs scored by a team's players in scoring position to the number of men to get into scoring position. Ratios are great, whiz-bang things, but subtracting ratios is just plain dumb since the denominator or divisor is different in each ratio. Here's an example: A team gets 1000 runners in scoring position and scores just 500. Its opponents get only 750 runners in scoring position but plates 400 of them. Its O Zone would be 500 divided by 1000 minus the result of 400 divided by 750 or -0.0333 (negative .03-bar). They would have outscored their opponents by 100 runs in these situations but still would have a negative rating.
Besides, who's to say scoring runs when runners are in scoring position is more important (i.e., correlates to winning) better than scoring runners when the bases are empty or there's a runner at first? A run is a run after all. Is a solo home run less valuable to a team winning than a run scored by a pair of doubles? Well, if you say that there were two hits in the second case, what about a run scoring on a single followed by a double? How about a single followed by a home run? Think about a run scoring on consecutive doubles representing a 100% offensive O Zone when a two-out, bases-empty single, followed by a triple, followed by an out results in a 0% offensive O Zone.

Okay, but maybe somehow the stars align and O Zone Factor really does capture the essence of winning. It would be easy to test. Unfortunately, given that the stat is "available exclusively" at MLB.com and its team sites and they only have stats for 2004 posted, our analysis would be based on a small, era-centric sample size.

However, let's forge ahead and see what we get. I'd like to run RF-RA or Pythagorean winning percentage (which is based on runs for and against after all) versus O-Zone to see which correlates best to winning percentage.

Here goes for the 2004 stats available. MLB for some reason refers to O Zone here by its non-marketing friendly name of NET-RS-RISP-%, Net Runs Scored Per Runners in Scoring Position Percent. I added Runs Scored Minus Runs Against and Pythagorean Expected Winning Percentage:

There's our raw data, and now for the correlations…O Zone's correlation coefficient (to winning percentage) is .7368. Both RS-RA and Pythagorean PCT correlate significantly: .9516 and .9512. Even MLB's own made-up intermediate stat NET RS-RISP (net runs scored minus runners in scoring position) correlates better: .8967. Maybe they should have stopped there.
So my conclusion is that O Zone Factor 2004 equals Game Winning RBI circa 1987. It's a stillborn idea that's D.O.A. but MLB just doesn't know it yet. GWRBI took nine seasons to die (it was recorded from 1980 to 1988). How long will it take the oh so ludicrous O Zone to expire?