Lancaster’s game story today (obviously) focuses on Eric Milton, and raises some interesting points.

Since his first start off the disabled list, a rough outing May 20 in Detroit that saw him allow 13 baserunners and five runs in five innings, Milton has been nearly untouchable.

“He’s never walked a lot of people, he’s always made guys put the ball in play,” said Reds manager Jerry Narron. “What hurt him last year, I thought, more than anything, was he threw a lot of fastballs in fastball counts in poor locations. So far, he has been mixing his pitches real well, he has been changing speeds real well. He is just not going to throw somebody a cookie just to keep from walking them.

They made plenty of contact and got the ball in the air more often than not, as only three of the 18 outs he recorded on balls in play came on the ground, but nothing had any oomph coming off the bat. It was like shooting popcorn out of a cannon.

“He jammed a lot of guys, I thought, with some fly balls,” said catcher David Ross. “Some blooper hits, too. The ones he gave up weren’t really hard-hit.

Milton attributes his success to the Mario Soto-taught changeup.

The nine million dollar question is whether Milton can sustain these results. I’m always worried when a pitcher’s success is tied directly to “better results on balls put in play.” I’m not going to re-open the whole Voros McCracken debate, but Milton is having remarkable luck in that regard.

In those last three starts, his opponents’ batting average on balls put in play (BABIP) is a miniscule .191. From the Prospectus Statistical Glossary:

Batting Average on balls put into play. A pitcher’s average on batted balls ending a plate appearance, excluding home runs. Based on the research of Voros McCracken and others, BABIP is mostly a function of a pitcher’s defense and luck, rather than persistent skill. Thus, pitchers with abnormally high or low BABIPs are good bets to see their performances regress to the mean. A typical BABIP is about .290.

Miltion’s strikeout rate in that stretch is around 5.1, which is okay, if not good. (Though if we’re going to parse an already tiny sample, he’s only struck out 4 in his last 2 starts (14.2 IP).

3 Responses

  1. al

    i don’t know if what i’m going to post is reopening the debate you said you didn’t want to reopen, but, well…

    the problem with the explanation of BABIP that you posted from BP is that it’s outdated and overly simplistic.

    Take this quote from the Hardball Times’ David Gassko who reworked the DIPS stat (which uses forms of BABIP) with Voros:

    “We know that a pitcher has some control over his BABIP. We even know how, for the most part, he can control it.” http://www.hardballtimes.com/main/article/another-look-at-batted-balls-and-dips/

    why that stat was reworked was that it was clear that pitchers could control what types of balls were put into play against them except for line drives, and that the different types of balls in play give different results (obviously). you can read the article for more info.

    the point is that while milton may be luckily giving up a few less line drives of late, which may change, the main reason he’s pitching so well isn’t luck resulting in a low BABIP, it’s the type of BIP that he’s allowing.

    Everyone has reported that he is changing speeds much more effectively and mixing up his pitches more, which has led to more weak BIP (like infield pop ups of which he had 6 yesterday).

    Weak BIP are hits less frequently and extrabase hits less frequently, and not a function of luck (ask mariano rivera about that), so i have some confidence that milton will continue to succeed, even if he has a nominal increase in BABIP.

  2. al

    his infield fly rates have increased though. and while i like stats as much as the next guy, all you have to do to see the difference, is see the difference, with your eyes.

    Whether it’s health, pitching coach, meditation, flaxseed, soto’s change-up, browning moving him over on the rubber, or if it’s sustainable, i can’t say.

    but i can say that right now, he doesn’t look a thing like the pitcher we ran out there last year.if the stats can’t pick that up, then they’re not effective.

  3. al

    sure sure, i get that our eyes can’t be trusted to fully pick up what’s going on in a complex game. my point i guess is to warn about data mining, because we live in a relational world, and if you look hard enough you’ll find a relationship that demonstrates just about anything. That doesn’t mean it’s significant.

    we have seen some empirical data over his last three starts: 1.19 ERA, .62 WHIP, .40 HR/9, 13/0 k/bb, .170 opBA, .170 opOBP, .244 opSLG.

    so finding one suspect stat that indicates a regression is what it is: one piece of information amongst many, not a harbinger of doom.

    last year with clemens, if you looked hard enough maybe you could have found that more grounders were taking good hops for him than “normal,” or that hitters didn’t swing at middle in fastballs as often as “normal,” and made the case that regression was going to come any day because of it.

    but at somepoint it’s just fishing, and the standard peripherals (along with observational data, which isn’t useless) has to take over.