Thinking out loud: Sherm and I had a quick exchange about linear weights in the comments section here a month ago, and for whatever reason, I thought about it this afternoon.
For years, I’ve argued that a big reason more people haven’t been exposed to more advanced offensive metrics is a simple matter of the language involved: We have easy verbs at our disposal that describe the standard, back of the baseball card stats that so many of us grew up with.
If I say, “David Wright hit .307 last season,” you know that I technically mean, “David Wright got base hits in 30.7 percent of his at-bats last season,” and you take it on faith that the average I’m presenting is correct and don’t bother looking up his at-bats and hits and doing long division.
I can attest that when writing about baseball, it’s sometimes tempting to rely on batting average — even if it’s an imperfect measure of offensive performance — for that reason alone. Saying “he hit .307” is easier and less awkward, in the course of a 500-800 word column, than writing, “he posted a .390 on-base percentage” or “he had an .837 OPS.”
I have to imagine there are baseball writers out there — ones much more widely read than I am — who would be more willing to incorporate advanced stats into their work if only there were more convenient language in place.
For a while, I searched for a verb that could convey on-base percentage. I e-mailed back and forth with John Peterson of Blastings! Thrilledge about this back in the day, but I don’t think we ever came up with a reasonable answer. To say “he based .390” sounds like some sort of drug terminology. “He reached .390” sounds like it was something he was striving for. “He safed .390”? Just weird.
Regardless, perhaps linear weights provide potential for a breakthrough. Though in concept, they are a bit abstract and somewhat difficult to grasp, they attempt to assign specific run values to every possible offensive outcome, relative to zero (making an out).
The stat wOBA — an attempt at a single, context-neutral universal offensive metric — relies on those linear weights. A good primer can be found here.
But instead of making the stat an average of linear runs produced per plate appearance, the stat’s creator, Tom Tango, made it scaled to the league-average on-base percentage to make it easier to digest.
That’s cool, and as someone who has been digesting on-base percentage for a while now, I appreciate it. Still, it adds another layer of complexity to an already esoteric metric, and one I doubt will help it earn any converts among the multitudes who weren’t already using OBP to measure offensive players.
This is almost certainly wishful thinking, and I’m probably missing something here, but I wonder if the stat would be easier to grasp if it were a simple, unscaled average of linear runs per plate appearance.
In my imagination — which is far removed from reality — that could solve the verb problem, since I could write “David Wright produced .320 last year,” or whatever it was, and you could know I meant “David Wright produced offensive outcomes worth .32 runs per plate appearance last year.”
Still too abstract for general consumption? Now that I think about it, yeah. That’s a really broad stretch beyond batting average.
Plus, like I said, I’m sure I’m missing something somewhere. Step up and tell me how I’m wrong, Internet.
Hear that? An angry mob of restless Mets fans, torches lit, are beating their drums. The drums are getting louder now as their crowd swells, and with the pulsing rhythm comes a cacophonous chorus of chants:
As for the excerpted bit, what a typically annoying and presumptuous thing for Roger Clemens to do. Imagine you’re a jewelry designer. You’ve spent years training and honing your craft, and you know that your big contract — the World Series ring — is one of your best shots to publicize the fruits of your labor.
Holy moly.
Also, the first commenter suggests giving them the finger. The ninth suggests Granderson greet fans by learning how to hit lefties.
Tedquarters 