Steven Katz ed.
Not Wins Against Replacement. That’s a subject for another
day. I’m talking about another kind of war - one I’ll start here against OPS,
the combination of On Base Percentage (OBP) plus Slugging Average (SLG). This
sum is exalted for its ability to reduce a player’s skills at the plate to a
single number. The idea, of course, is to measure the ability of a player to
both get on base – On Base Percentage - and hit for power - Slugging Average.
According to the conventional wisdom (try Bill James himself), by combining
statistics with established and unassailable track records for measuring two
important hitting skills, a reliable measure of a player’s offensive value
emerges. According to this thinking, for every player, an OPS of .800 or
higher is very good and an OPS of .900 is great.
With W.A.R. still perceived by many people as too
complicated and convoluted, OPS has overtaken Batting Average as the single
most important index of all-around offensive prowess. Given inside-baseball’s
near-universal acceptance of the stat, my war was probably lost before it starts.
But my memory of the failure of the Game Winning RBI stat keeps hope
alive.
Let’s examine the flaws of this superstat:
1. OPS mixes apples and oranges. Some things belong together: Peanut butter and jelly; hydrogen and oxygen; hot dog and mustard. Some things do not. Yes, OBP and SLG are easy to understand, easy to calculate, and time-tested measures of a hitter’s ability. But that doesn’t mean they should be clumped together. On Base Percentage measures the ability to reach base, Slugging Average measures power, and the two have little to do with each other.
What do Tony Gwynn and Reggie Jackson have in common? Their
OPSes were almost identical - Jackson at .846 and Gwynn one point higher at
.847. They’re both in the Baseball Hall of Fame for their achievements at the
plate. But they were polar opposites in terms of the actual skills they brought
to their respective clubs. Gwynn got on base a lot. Reggie Jackson hit a lot of
home runs. By most measures – balls put in play, runners moved over, run
production, whatever - the two players had little in common. OPS might be the
last stat a general manager would look at in trying to figure whether a player
was the right fit for his club.
2. OPS is not a balanced offensive measure. It is not “transparent,” if you'll
excuse the overused term. Since Slugging Average tends to be 50-150
points higher than On Base Percentage, OPS will always be top-heavy on the
slugging side. (Consider the difference in the two denominators; the maximum On
Base Percentage is 1.000 while the maximum Slugging Average is 4.000.) All
skills being equal, the scale is tipped heavily in favor of long-ball hitters
against the guys who get on base more. With Slugging Average carrying the day,
OPS is not a balanced measure of overall offensive ability. Thus, OPS may be
the only stat where deconstruction becomes an essential tool of analysis. Only
when baseball fans break down a .767 OPS into a .310 OBP and a .457 SLG, or
into a .380 OBP and a .387 SLG, can they gauge a player’s real
value.
3. OPS is actually a throwback
statistic. Statistics are the lifeblood of the game. The range of ways to gauge
performance has been expanding since the first diamond was laid out. The recent
explosion of new ways to measure success and failure is breathtaking. A
Nineteenth Century record book would fit neatly into the palm of your hand. In
the mid-Twentieth Century, the career records of several players could fit on a
single page. Now, the statistical life of even the most obscure player consumes
several pages. Like most everything else in the real world, overall achievement
is judged by many standards over an expanse of time. The worth of everything
(stats included) lies in context. Is OPS an irony? Or is OPS a return to a
simpler time? Are we actually wearying from the abundance of new measurements
of performance?
4. Simply put, it just isn't useful.
So let's replace it with something that is. The number doesn't mean anything.
The number seems benign, but it doesn't actually represent anything tangible.
Sure, it's easy to calculate, but that's really the only thing going for it.
And compared to the mathematics involved in modern advanced statistics,
modifying the OPS equation into something that represents meaningful, concrete
information about a player's performance should be a breeze. In place of OPS, I
propose a Total Bases Average (TBA) that would present a number signifying how
many bases a batter earned per plate appearance. Total bases, including base on
balls and hit by pitches, divided by plate appearances. That's simple enough.
And best of all, it actually tells you something clear and meaningful about a
player's offensive production.
I urge the baseball establishment to get rid of this irrelevant statistic. If you're itching to see how OBP and SLG look together, you don’t need a separate column on a stat sheet; simply add the two numbers on your own. I'm just not sure why you would.