11 November 2014

Miguel Gonzalez in Perspective

This post was written by Ryan Romano. Follow him on Twitter.

When Miguel Gonzalez debuted for the 2012 Orioles, and posted a 3.25 ERA in 105.1 innings, sabermetrically-inclined fans knew it wouldn't last. After all, he backed up that ERA with a 4.38 FIP and 4.63 xFIP -- numbers that scream regression. And yet, he came out in 2013 and largely repeated the story: In 171.1 innings, he pitched his way to a 3.78 ERA, 4.45 FIP, and 4.31 xFIP. With a 3.23 ERA to go along with a 4.89 FIP and 4.46 xFIP, 2014 brought more of the same.

So, is this who Gonzalez is now? Perhaps; according to Lewie Pollis, three straight seasons of luck give Gonzalez a 44.3% chance of being an overachiever. In other words, the larger the sample, the greater the chance that Gonzalez can consistently defy DIPS. In the coming weeks, I'll take a closer look at him to see if any of his traits might suggest overperformance; for now, though, I'd like to do a comparison.

Gonzalez's career ERA and FIP currently sit at 3.45 and 4.59, respectively. Those translate to park- and league-adjusted figures of 85 and 114, respectively. This level of luck, over that much time, struck me as unusual, so I decided to look at history. I searched for people who had done the same -- i.e, pitchers with an ERA- below 90, and an FIP- above 110, over a three-year span of at least 400 innings.

Other than Gonzalez's streak, this has happened 28 times in baseball history. Seventeen men have done it, so a few guys have multiple ones. The average ERA- and FIP- during the years were 84 and 114, respectively, so they certainly matched Gonzalez's good fortune. But the thing that mattered to me wasn't how they did during their three-year stretches, but how they did after that. So I tracked down the results of each of them for the rest of their careers, and what I found didn't paint a positive picture.

The average ERA of the group came in at 16% worse than the MLB average; the average FIP, 21% worse. Only four of them owned league-average or better ERAs for the remainder of their time in the show. By comparison, four of them didn't even pitch after the three years. (Remember, pitchers get hurt.)

The most recent example pretty perfectly encapsulates the fate of the group overall. From 2010 to 2012, Jeremy Hellickson had a sterling 79 ERA-, which obscured a 115 FIP-. Since then, the latter mark has improved marginally, to 113, but the former has flown up to 133. This certainly doesn't bode well for his future with the Rays, or in the majors period.

Does this mean anything for Gonzalez? Probably not; this is hardly a scientific study. But his peripherals still make me worry, and Steamer (which foresees a 4.58 ERA and a 4.94 FIP in 2015) doesn't help. Although this merits further investigation, the initial impression is clear: In the majority of cases, luck ends, and it often goes out with a bang.

Photo via Keith Allison


Pat Holden said...

Nice post, Ryan

Statistics Don't Lie said...

So 3 options for the Orioles:
1) Trade him, selling high, if any team has faith in his continued abilities, or had a ballpark and defense that suits him.
2) Keep him until he loses it, because nobody will pay what he's worth and his salary is well under control. If he loses it, see option 3.
3) Convert him to a reliever where he won't be overexposed and open up a rotation spot for a candidate with more upside potential than downside risk.

Eric said...

Good article, and I actually agree with the last comment. All 3 wouldnt be bad options at all. Even if we get a quad-A player in return or a possible DD special of a buy-low get a good season guy it might be worth it. It opens up a rotation spot for someone, which could helps since we somehow have more than 5 at least semi-qualified starters, and maybe more on the way if things go great. Worst bet is more relief work which is something teams always have in flux so it isn't the worst. He still is Arb-eligible, so if anything they just cut him and that is that.

Matt Kremnitzer said...

I'm not sure why the Orioles would go into the season with the plan to use Gonzalez as a reliever. Even if he's pitching above his abilities, he's demonstrated that he at least deserves to be in some team's rotation. And they also apparently offered him to Boston in exchange for Andrew Miller last year, though they later reluctantly parted with Eduardo Rodriguez.

There's nothing wrong with having starting rotation depth, but the O's have a collection of back-end rotation types in the minors already (along with higher ceiling pitchers Dylan Bundy and Hunter Harvey, who are not ready yet/recovering from injury).

Even if Gonzalez is only an average or slightly above average starter, he's only now reaching arbitration, so he's under team control through the 2017 season. He certainly holds more trade value than the other starters the O's may consider dealing -- Chen and Norris -- who are both scheduled to be free agents next offseason.

I think the O's should trade one of their starters. If they want the biggest return, that pitcher should be Gonzalez. Perhaps there is some team out there who thinks he's really this good.

Ryan Romano said...

I agree with Matt -- at this point, they should just roll with Gonzalez as a starter. With that team control, he could offer a decent return in a trade, and who knows, maybe he'll even keep this up.

Also, in case anyone cares, here's a spreadsheet with the data: https://docs.google.com/spreadsheets/d/1-hGnGToCa9YvQ-L9Z8hNZqqUhe_qDDg9o1plBXY6vUY/edit?usp=docslist_api