OPS+

Q.  You have to dock Jose Lopez a bit on his OPS+, because it's based on SLG not OBP, right?

A.  Right.   But.

... for those who haven't caught this point before, OPS and OPS+ do treat OBP and SLG as though they were of equal value.

OBP is in fact more valuable, yet you'll notice that a guy with a 300/380/420 line is considered equal to a guy with a 300/340/460 line.   The .380 OBP will get you more runs, other things being equal.

So for guys who don't walk, and/or have low AVG's, you do need to subtract, say, 5% from OPS+ to get the right "eyeball" estimate of offense (in terms of RC/27).

Sabermetricians remind us that for this or that player, you should subtract 5 or even 10 points of OPS+ for a player who skews far away from OBP.

Very true, though in practical terms the difference isn't as great as it's sometimes represented to be.

.

Q.  Is there a better way to "adjust" OPS+ for low-OBP guys, than "subtract 5 points"?

A.  Sure.  RC/27 will give you a feel for how the 300/380/420 and the 300/340/460 guys compare -- theoretically.

Jose's RC/27 for 2008-09 is around 4.5, meaning that nine Joses would score 4.5 runs a game.   The AL average is around 4.8.   So Jose's OPS+ would run around 94 by that measure, rather than the 99 it's been.

See?  Five points.  And Jose's RC/27 is not park-adjusted.

A (real) hitter with a 100 OPS+ will have a RC/27 very similar to the league average runs scored, whether or not he has a high OBP.

So it's worth considering, but of course it's not like it renders OPS+ unusable.

.

Q.  You said we should dock Jose, "But"?

A.  OTOH, anybody who wasn't a "purist sabermetrician" would put the thumb right back on the scale for Jose Lopez, the one that RC/27 took off, because of Jose's ability to hit tough pitching.

True he doesn't BB into that 105 OPS+, but it's also true that he gives a "hard" AB when the runs are leveraged.

Show me two 105 OPS+ players, one a "soft" player who does draw some walks, and one "hard" player who doesn't walk, but who does get you 100 RBI, and I'll take the hard player.

This is a battleground on which the "purist sabermetrician" and the "blended saber/scout" -- like Capt Jack is a blended saber/scout -- would differ most sharply.   "Purist sabermetricians" would be unlikely to buy into the idea that certain hitters will challenge tough pitchers, and others will cake out.

D-O-V blends 'em on this one.

And therefore I wouldn't overemphasize the OPS+ penalty on Lopez.  He's not getting the walkoff RBI's by no blinkin' accident.   He's a 90-100 RBI man in a good lineup.

.

Q.  What's the other asterisk?

A.  This (valid) rule that says, OPS+ is a little high for players who don't walk ... IMHO there is another asterisk, that being for players who hit home runs.

For those just joining us, James and Dr. D wrote a little study for the old STATS Baseball Scoreboard that compared 20 sets of historical team-pairs:

20 Teams A:  OPS 100*, more home runs & BB's, fewer 1B's and 2B's

20 Teams B:  OPS 100*, fewer HR's & BB's

The teams that hit more homers, for the same OPS, scored quite a few more runs.

Homers change the scoreboard.  RC/27 is theoretical.

...................

You take a Vlad Guerrero type, and say, well, his OPS+ is a bit high?  Not really.  Vlad changes (changed) the scoreboard probably *more* than his OPS+ suggests (suggested).

.

Q.  Is OPS+ close enough for government work?

A.  James invented Win Shares so that he could SCAN rosters, across years, and very quickly see at-a-glance some very large comparisons.

OPS+ does that for me.   Did pennant winners tend to have 60 OPS+ hitters in their lineup, or didn't they?

Did Cinderella teams tend to shore up 60 OPS+ to 90 OPS+ in the next year, or didn't they?

Is Jose Lopez getting better, year-to-year, or isn't he?  By how much?  OPS+, with its 100 scale, is simply ideal for comparing a player to himself, very quickly, like in three seconds, and then on to the next card.

OPS+ is good for moving large mounds of dirt, and good for moving through piles of information very quickly.   Personally, I try to use OPS+ to gain a sense of proportion, when comparing lots of things in a short time.

But if you want to insist that RC/27, or some derivative stat that's 2% more accurate than RC/27, is the one you want to use, that's cool.   All statistics, ERA, W/L, RBI, everything, all stats have information that can be helpful when used for the right purpose.

.

Q.  The bottom line on Jose?

A.  He's an average-solid hitter who can get you the big RBI.  Right now he'd be a lot more valuable as part of a supporting cast.

Cheers,

Dr D

Comments

1
Taro's picture

wOBA takes it all into consideration.. HRs, OBP vs SLG, SB, CS, it gives run values for every outcome. Its not perfect, but its probably a little more accurate OPS+ or RC.

2

Most hitters have trends...there are pitching families they will do better against because of their style and there are families they do worse against.  Lopez has very few such trends.
Examples:
vs. RHP: .701
vs. LHP: .731
at home: .684
road: .733
Behind in the count: 24% better than league average (though still horrible...it's amazing how bad the average ballplayer hits when down 0-2, 1-2 and 0-1)
Bases Empty: .644
Men On: .788 (this is NOT clutch skill...this is the defense shifting and Lopez getting more singles)
Leverage Index of situation: Lopez hits the same in any state more or less
Power pitchers: .665 (low BA, everything else the same)
Contact: .744 (less power, very high BA - proves Lopez adjusts to pitcher type very well)
Flyball pitchers: .743
Groundball pitchers: .650 (he likes the high pitch...his swing is too level which is why he doesn't hit for as much power as he should)
He can give you a good AB against almost any type of pitcher though it may not always include a lot of pitches per PA.

3
Sandy - Raleigh's picture

Home: .224/.240/.350/.590 (4 HRs, 15-2B)
Away: .316/.352/.538/.890  (12 HRs, 16-2B)
If there wasn't a REASON for the change, I'd think it was noise.  But, I see regime-change as very much causal.  And the real big tell that the AWAY line is likely more reliable?
Home: .230-BABIP
Away: .316-BABIP
Which one is less luck-biased?

4

...are you arguing that Lopez will continue to have his 2009 home/road split or are you arguing that it's bad luck?  First you say regime change caused the hitting funk at home...then you say his home OPS is dragged down by a very unlucky BABIP.

5

But how did French look tonight?  I can't watch the Ms where I live, but the statistics look good; 6 Ks, 10 swinging strikes, 66% strikes while getting the requisite 8-10 strikes taken away (admittedly getting one or two back this time), his fastball averaged 88.8, and topped out at 92.

8
NYMariner05's picture

French was mediocre, which is probably the best you'll get from him.  I still don't want him in the rotation next year or beyond. He's a nice guy to store away in Tacoma for starting pitching depth.

10

... remember how the old platform at D-O-V used to have you banging your head against the desk and weeping into your hands?
Gotta admit, it's pretty nice to have a more robust platform, no :- )

12

These guys at KLAT have a nice blog set-up for most things.  Their software doesn't really allow for my more meaty statistically oriented posts...no tables/charts, no clean formatting of lists other than the prepackaged options and there's something weird about the style sheet here that causes too much white space between paragraphs, but this is a heck of a lot nicer than the original DOV, and their server is definitely better (and able to handle the traffic, which is a huge plus).

13
Sandy - Raleigh's picture

My point is that prior to 2009, he showed no sign of a home/road split.  His pre-'09 approach negated much of the Safeco effect *AND* suppressed his road performance, (thereby suppressing his overall performance).
I'm saying that the .590/.890 split this year is probably more extreme than he'll produce in the future, but that based on his BABIP, his home stats in 2009 have been on the unlikely side FAR more than his road stats have been on the lucky side.  So, the likely result for the future is that his home stats will improve MORE than his road stats decline, (creating a better final aggregate).  No, I don't think he's a .890 hitter.  But I think he's probably going to be posting significant home/road skews in the future, (100+ OPS) difference.  The lines will be similar, except the road HRs and lack of safeco effects will keep his road line well ahead of his home line. 

14

If the argument is that now that he's trying to hit HRs from time to time, his power numbers will be better on the road, I can see that as a possibility.  And using idea, one might project Lopez to be a perennial .290/.330/.490 hitter once the home numbers revert to a more normal BABIP.

Add comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd><p><br>
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

shout_filter

  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.