OPS Against

by elsandito on Aug 24, 2009 10:56 AM EDT Comment 13 comments

I had been scouting around the ESPN website looking for information and noticed that it contains each team's OPS. Also, it contains OPS against for the pitching staff. I found that when I subtract OPS against from a team's OPS and ranked the result within division, the ranking within the team's division bears a very high correlation to that team's ranking in its division. Interestingly, Chicago's ranks ahead of Detroit's differential.

0 recs | Comment 13 comments

Story-email Email Printer Print

Comments

Team OPS is fine…but the whole point of OPS is to measure an individual’s contribution toward run production, and when you’re talking about a whole team, you can just look at the actual runs. There are still issue like correcting for park effects (or strength of competition if you’d like), but using OPS instead of runs doesn’t help that.

by dgcambridge on Aug 24, 2009 12:32 PM EDT reply actions 0 recs

Wouldn’t OPS be less luck-dependent than total runs?

by Ryan Kelsey on Aug 24, 2009 12:49 PM EDT up reply actions 0 recs

Let’s just say it would be independent of luck and other situational factors, e.g., blowouts.

OPS differential should roughly correlate to Runs Created differentials.

OPS under-weights OBP relative to slugging. (Notably, OPS+ basically corrects this issue.)

OPS also does not take into account any type of baserunning, neither stolen bases nor extra bases taken nor outs made on the bases.

Well, that's just, like, your opinion, man.

by Jay on Aug 24, 2009 1:47 PM EDT up reply actions 0 recs

RC and RC/9 might be my favorite stats. Much though I like the ideas behind EqA and wOBA, I really struggle with stats that are supposed to look and work like idiot stats. I mean, EqA is “keyed” to bAVG, great. What’s that mean? Whatever its faults, RC explains itself perfectly at a glance, which is probably the most important thing to me.

by fleerdon on Aug 24, 2009 2:47 PM EDT up reply actions 0 recs

But wOBA isn’t supposed to look like an idiot stat, right? Unless you count OBP as an idiot stat, because (as far as I know) wOBA is simply OBP weighted by the value of each hit. It’s not scaled to look like BA like EqA.

by Buckeye Brad on Aug 24, 2009 5:16 PM EDT up reply actions 0 recs

Fudging the explanation, true. wOBA, I’ll grant you, may sink in for me eventually, but I’m not an early adopter.

by fleerdon on Aug 25, 2009 1:12 AM EDT up reply actions 0 recs

I’ve never warmed to wOBA, for just this reason. There is only one hitting stat where I’ve been willing to learn an arbitrary, meaningless scale, and that’s OPS.

Well, that's just, like, your opinion, man.

by Jay on Aug 25, 2009 8:38 AM EDT up reply actions 0 recs

But isn’t wOBA better than OPS? It properly weights each hit according to its run-scoring value. OPS doesn’t give enough value to OBP.

by Buckeye Brad on Aug 25, 2009 11:11 AM EDT up reply actions 0 recs

Yes, but the difference is pretty marginal between any of these stats. OPS is easily understood and really very nearly as accurate, and OPS+ basically eliminates the OBP weighting issue altogether.

Well, that's just, like, your opinion, man.

by Jay on Aug 25, 2009 11:16 AM EDT up reply actions 0 recs

plus, you have to double all of garko’s rbi.

by Brick. on Aug 24, 2009 12:50 PM EDT up reply actions 0 recs

The whole point of OPS is to measure an individual’s contribution toward run production. The mirror image to this is to measure an individual pitcher’s contribution toward run prevention. OPS and OPS+ are widely recognized as valuable metrics. It seems that OPS against should receive an equal recognition as a pitcher’s metric.

featuring the curses of Bobby Bragan and Rocky Colavito

by elsandito on Aug 24, 2009 2:20 PM EDT up reply actions 0 recs

But they’re not mirror images at all. Batters proceed through the game in a lineup. Pitchers much less so. Is your concern correcting for runners inherited/left? Obviously that is much less of a problem for starters vs relievers.

I can think of those two reasons to go backwards from “runs allowed” to OPS against. One is the inherited runners correction. The other is because of the specific correlation between baserunners/hits/extrabase hits and the resulting runs, and a worry that too much “luck” is involved.

But with that second part, I worry that we are throwing out useful information also. Maybe some pitchers are more effective with runners on base, or at controlling the running game, or inducing double plays. Maybe some are better at challenging hitters without runners on base.

And when we have enough innings, don’t we expect the specific kind of luck here to wash out? (remember this isn’t even BABIP luck, which is present also in OPS against). This is especially true if we’re talking about a whole teams’ innings pitched, with very large IP(and in which case we lose the first problem I listed above.) The OPS —> runs correlation isn’t perfect.

Again, my whole point is that, with an individual hitter, we don’t even have a direct “runs created” stat, because of the lineup. But with these teams stats, we have those runs. The runs are the truth. So we need a good reason to go backwards.

by dgcambridge on Aug 24, 2009 2:59 PM EDT up reply actions 0 recs