Human League

• Posted by Adam Warner
• on July 21st, 2010

So Human Element Strikes Again!

Here’s the situation behind this picture. Mets closer Walk-Rod (formerly known as K-Rod) has already blown a 3-1 lead and yet another great outing by Johan Santana. It’s now 3-3, bottom of the 9th, ground ball to David Wright, he throws to home and the runner slides into the plate before the catcher gets the tag down and…..of course he’s called out. Giants then don’t score, Mets score in top of 10th and win. And thanks to The Human Element, they are 1-5 since the All-Star break, and not 0-6!

Hat tip Brooklyn Mutt via Metsblog.

Anyway, not here to write about bad ump calls that we could easily correct. Nope, it’s silly statistics that Baseball Geeks use to show you that what you see is not *real*!

That and that “mean reversion” is a tricky concept.

Case in point, Joba Chamberlain. Yankee fans, Joba Fantasy owners, anyone actually seeing him pitch can’t help but notice he’s essentially an inneffective journeyman reliever right now.Bloomberg Sports? Not so much.

On the surface, it isn’t hard to see why both Yankees fans and fantasy owners have found Chamberlain frustrating in 2010. The big right-hander’s ERA stands at 5.77, his WHIP’s at 1.51, with opponents now hitting a robust .290 against him. He has not had more than three consecutive scoreless appearances since the middle of May. Of the 11 appearances in which he has allowed a run this year, he has allowed either 3 or 4 runs in five of them, a trend that makes it all but impossible for him to consistently lower his ERA.

In reality, Chamberlain’s luck has been the biggest difference this year from years past. Chamberlain’s FIP (Fielding Independent Pitching–a version of ERA that strips away all luck-related variables) stands at 2.66, more than three runs lower than his ERA. He is not walking his way into trouble, as his 3.23 walk rate is his lowest since 2007 and about league-average. His live drive rate is up a little bit, but a 21.7% line drive rate–compared to a career rate of 19.4%–doesn’t account for more than three runs of difference between ERA and FIP.

Chamberlain’s sky-high .391 BABIP (Career: .327) and microscopic 58.7% strand rate (Career: 73.1%) are the main reasons for his struggles. One minor change, a decrease in pop-ups (2.9%, Career: 9.6%), hasn’t helped. But the rest of his profile’s mostly unchanged.

The stat behind this all and the one that truly drives me nuts? BABIP.  The theory is that pitchers can control their walks and their K’s, and their home runs allowed (sort of) but will more or less mean regress to allowing a little under 3 in 10 balls in play becoming hits. So if he allows, say a .210 BABIP, he’s extremely lucky, not actually skillful in generating weak contact….as a more mortal like myself would see it .(Mariano Rivera incidentally has a .194 BABIP this year and has stranded 80% of his baserunners) And if he’s Joba and has a .390 BABIP? Must be bad luck.Oh and strand rate? That’s the percentage of baserunners he strands on base, the league average is something like 70%. So where you and I maybe see an inability to to bear down and get key outs, the stats borg sees another stat that should mean revert.

So why hate BABIP?

*it’s distorted by survivor bias. You and I couldn’t pitch in MLB to a .290 BABIP, more like .990. That “mean” number assumes major league level skills. A guy pitching to a .390 BABIP may very well not have the skills necessary to pitch in MLB. Of course I’m not saying that about Joba, just the flaw in using that stat as a crutch to expect his performance to magically improve. Higher BABIP may imply bad luck or bad defense behind you, but it also may imply bad skill at inducing weak contact. I would think I could tell the difference with my eyes better than with this number.

*Pitches have a group mean in the .290 range, whereas batters have “personal” means. A guy who zips to first like Ichiro or a guy making powerful contact on a more regular basis will have high BABIP’s. How can it possibly make sense that a pitcher has relatively no influence on his BABIP while a batter has large influence? At the very least, a pitchers BABIP has to relate to the batters he faces.

*It just defies common sense as a viewer. Watch a game. If a pitcher is getting lit up with hard contact, he’s PITCHING BADLY. If he’s Mariano and he’s sawing guys bats in half and inducing a bunch of soft grounders, he’s PITCHING WELL. Obviously there are days when popups and grounders find holes and other days where line drives go right at fielders and fly balls get caught at the warning track. But again, you can actually see that by watching, not a randomized stat.

I don’t watch the Yankees all that much, but my friend Dave tells me Joba throwing batting practice these days. I’d sooner take his unscientific scouting opinion than these stats. If Joba turns it around, it will be because he pitches better, not some better luck.

The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.

blog comments powered by Disqus