April 21, 2014

SABR-Toothed Triber: Justin Masterson’s Regression

Just a few weeks back, I started to question how much longer Masterson could stay in the Indians’ rotation.  Early in the season, I knew it had the potential to get ugly, but I was all for sticking with him for the year to see if he could figure it out.  But each time Masterson took the mound, things seemed to get worse: his control looked off, lefties were still pounding him, and he just couldn’t catch a break.  For his sake, I started to wonder if Masterson didn’t need a break from the rotation.  Further, it looked like Laffey’s demotion to Columbus signaled that the front office had seen enough of the experiment as well, and they were on their way to shifting Masterson back to the ‘pen, as soon as Laffey could get stretched out.  I can’t say I disagreed.

But then a funny thing happened on the way to demotion.  Masterson went on a three game tear that looked something like this:

@ NYY 6.2 3 8 1
@ CHW 5.2 1 1 6
v BOS 9 0 6 2

Over that span, Masterson’s ERA dropped from 6.13 (prior to his start against the Yankees) to its current 4.74.  Sure, not every number in that chart is pretty: against the White Sox he walked six and only struck out one.  But he “controlled the damage” and got out of Chicago with his first win in nearly a year.  He absolutely dominated the Red Sox, shutting them out while getting 17 of the 21 balls in play to go for groundballs.  And three runs over seven innings is nothing to sneeze at when you look at the Yankee lineup.

So does this mean Masterson has this starting thing all figured out now?  Should we rest easy?  I figured I’d take a look at some of his career numbers and compare them to this season.  I was surprised at what I found:

2010 8.03 4.61 0.66 4.74 0.353
Career 7.80 4.25 0.85 4.15 0.306

What differences do we see from his entire career to this season?  He strikes out slightly more this season than his career rate would suggest, and that’s good news—especially considering his shift from the bullpen to the rotation.  Yes, he’s walking a few more batters than usual, but that’s to be expected with the demands of starting, and regardless, he’s letting up fewer homeruns, so the walks shouldn’t kill him.

So why should his ERA be .60 runs higher this year than it is for his career?  Well, look to that last column.  Masterson’s BABiP (batting average on balls in play) is an unsustainable .353 this year.  Why do I call that figure unsustainable?  Well, for his career, it’s been only .306.  And on top of that, if you’ll remember from one of the first pieces I wrote here, BABiP is largely affected by luck, but most pitchers regress to .290 to .310.  For his career, Masterson is slightly above .300—which would make sense, since groundball pitchers obviously will have higher BABiPs than flyball pitchers.  That .353 BABiP this season is just flukey.  And we can attribute a good portion of his rocky start to balls dropping in for hits, when they just as easily could have been turned into outs.

In other words, Masterson’s had some crummy luck.  But over the course of a season, we’d expect that luck to balance out (or regress) back to his career norms.  What happened in the last three games is recompense for what had been happening to that point in the season.

Still don’t believe me?  Let’s look at another chart:

4.74 4.07 3.88 4.21

You all know what ERA is, but what about those others?  Well, they’re all ERA estimates that try to take “luck” out of the picture.*

*Quick and dirty descriptions: FIP is “fielding independent pitching”.  It counts only walks, strikeouts, home runs, and HBP to approximate what a pitcher’s ERA should be given average luck and defense.  xFIP is “expected FIP”, and it is exactly the same as FIP, except that it believes pitchers don’t have control over the rate at which they allow HR, so it just uses an average for all pitchers.  tERA is a bit different, as it accounts for the various hit-types a pitcher can allow (line drive, fly ball, ground ball) and assigns run values to all of them.


So what do I take from this chart?  Every ERA estimator is below Masterson’s actual ERA!  That means no matter how you skin the cat (FIP, xFIP, tERA), Masterson has been seriously unlucky this year.  xFIP says he’s pitched well enough to garner a 3.88 ERA, while tERA suggests it should be 4.21, but either way, it shouldn’t be anywhere near 4.74.  Why is it so high?  Because his BABiP has been an unlucky fluke so far.

What we’ve seen over the past several weeks has been a correction of sorts.  Masterson was never as bad as his early numbers suggested, and he isn’t quite as good as he looked against the Red Sox last week: in that start he gave up 21 batted balls, so it would be fair to think that about 30% (six or seven) would be hits.  In actuality, he gave up only two.  That night he was lucky.  But most other nights, he hasn’t been.

I’m still not sure Masterson will be a great starter.  But that was never the goal.  The goal was to slot him in the middle of the rotation, eat some innings, and get some experience.  And that’s exactly what he’s done.  We know more about him now than we did at the beginning of the season.  We’ll know even more by the end of the season.  And what else could we want from this rebuilding season than to figure out about our young players going forward?

  • JNeids

    I’ve been seeing that BABiP stat flying all over the place this season but never once stopped to read what it actually was. Thank you for this write-up that someone as lazy/stat-ignorant as me could understand and enjoy. I now feel better about having just picked him up for my fantasy team yesterday.

  • http://www.kidcleveland.com Mike B.

    The defense doesn’t really do the staff any favors either.

  • http://www.waitingfornextyear.com Jon

    @ Mike B.

    Right on. When I was putting this together, I looked up our defensive ranks. According to UZR, we have the fourth worst defense in the Majors at -18.4 runs so far. Biggest offenders: Trevor Crowe (-8.2 in CF) and Jhonny Peralta (-5.8 at 3B). FWIW, Choo is our best defender at +4.3 runs in RF.

    Luck is part of it, but if your fielders are below average, balls in play don’t turn into outs as often as they should. Good point.

  • Tommy

    1. Love the article as always

    2. Masterson’s FIP numbers haven’t changed much over the last 3 starts at all. I know you made note of them before, but why are we so much more comfortable using them as justification now?

    3. I am not convinced Masterson is really a .300 BABIP pitcher. Between 2009 and 2010 he is nowhere close to that number, and 2008, which is the season that drastically brings his career BABIP down, he was mostly a reliever and faced a much higher percentage of righties. Also, since he has come to the Indians, his GB% rates have been much higher. Whether this is due to being a full-time starter, or natural maturation/progession as a pitcher or something else entirely, it is probably a better indicator of his future. And if he is going to be a league leader in GB%, he most likely will have a very high BABIP to go along with it.

  • http://www.waitingfornextyear.com Jon


    To your second point: “we” aren’t more comfortable with them now. Smart people would have been comfortable with them all along, but I was doubting them because I was believing my eyes more than my brain. My fault.

    To your third point: You’re essentially arguing that the FIP model doesn’t apply to Masterson, because he’s destined to have higher than average BABiPs.

    Two things in response to that.

    First, you may be right. In 2009, Masterson had a .321 BABiP. That’s higher than normal, and as a result his FIP (4.00) was less than his ERA (4.40).

    On the other hand, BABiP just doesn’t correlate very well from year to year for any pitcher. The top five last season are nowhere near the top five this season. I’d be more likely to assign his .320+ BABiPs to the Indians’ bad defense and bad luck than to Masterson’s groundball-tendencies. After all, Fausto–a notorious groundballer–has a career BABiP of .298. There’s got to be something else at work, and I’m calling that something “crummy luck.”

    Would it be safe to say that Masterson has a “natural” BABiP in the neighborhood of .320? Maybe. But .350? I don’t think so. No pitcher has sat with a BABiP that high for his career.

    And to your first point: well, thanks man.

  • http://www.mattyfos.blogspot.com MattyFos

    These SABR-Toothed Triber articles make by brain hurt..