The magic of Jeremy Hellickson

Many of you reading this article watched Jeremy Hellickson pitch over the last two seasons. If you have, you've probably seen a pitcher who doesn't strike out a lot of guys, who gives up the occasional solo homer, and tends to give up balls in play that end up up the gloves of his defenders. Although he doesn't necessarily look dominant, not blowing away hitters like Justin Verlander nor a control artist like Cliff Lee, Hellickson has posted the tenth-lowest ERA of any qualified pitcher over the past two seasons.

The big question is: how does he do it?

Over the past several years, many baseball analysts have championed the idea of DIPS, or defense-independent pitching statistics. Basically, the idea is that there are only a few things pitchers can control: strikeouts, walks, and home runs -- once a ball is hit in play, what happens is largely out of the pitcher's control. The most prominent statistic to measure pitcher performance by this methodology is called FIP, and can be found at FanGraphs. FIP is especially cool because, in the short term, it is much better at predicting a pitcher's ERA than, well, ERA is.

At any rate, FIP and ERA usually don't differ a huge amount over the course of a full season, especially in the case of starting pitchers. Some pitchers may have an FIP that under- or over-performs their FIP due to luck, or exceptional defense, or, rarely, because they can do something that actually does effect a ball in play that isn't captured in FIP. But more often than not, FIP and ERA tend to even out, more or less, over time.

Well, Jeremy is what one might consider an outlier. In 2012, Hellickson posted a 4.60 FIP, which is the kind of FIP you see from a below-average American League starter. His failure to strike guys out, limit walks, and keep the ball in the park tends to paint the picture of someone who's no star. Yet his 3.10 ERA is that of a superstar, ace-level starter.

To put things in a little perspective, no pitcher in baseball has posted an ERA that so greatly outperformed his FIP since Al Leiter in 2004. And over the past 10 years, Jeremy Hellickson's 2012 is the third-largest case of ERA out-performing FIP. The fourth-largest?

Jeremy Hellickson's 2011 season.

Somehow, it's starting to appear that Hellickson can be effective, despite not doing the most important things that a player can do to limit runs.

In his two years in the bigs, Hellboy has turned modern sabermetric thinking on it's head, as well as most projection systems -- at least for now. The big question is how he does it. Very few pitchers make a habit out of out-performing their FIP. Giants ace Matt Cain is the most often-cited example of a pitcher whose ERA outperforms his peripherals. Yet Cain's career difference between his ERA and FIP is fairly small, the difference is about a third of a run (-0.37). Hellickson's is more than a run higher (-1.44), which is more than twice as big of a difference as any other pitcher with 400 innings over the last ten years.

To put things in perspective, let's take a quick look at a number of "peripheral" statistics for Hellickson over the course of the 2012 season.

So check this out: Hellickson was worse than league-average in regards to almost every important statistical metric last season. He struck out fewer hitters than average, walked more and gave up more home runs per plate appearance. He gave up more home runs on fly balls and induced less ground balls ... all of this means that he should give up more runs than the league average.

The things that don't add up? His BABIP is low, much lower than the league average. That means that when balls were hit in play, they ended up as outs more often than usual. A low BABIP makes sense for a Rays starter, though, especially given that (1) the Rays consistently post good defensive numbers and (2) their park, Tropicana Field, suppresses singles and doubles, while raising infield fly rates.

You know, another pitcher that had famously high LOB% over his career was Braves and Mets great Tom Glavine. Glavine also was consistently able to out-perform his FIP, over the course of his career. Perhaps we can see if Hellickson does any of the same things that Glavine does? One thing Glavine was quite good at, getting opposing hitters to ground into double plays, Hellickson is -- again -- only about league-average at doing.

One of the things that Glavine did famously well, was limiting the number of home runs he gave up with runners on base. Well, Jeremy Hellickson, through luck or skill, has managed to do the same thing. Of the 25 home runs he gave up on the 2012 season, only five of them were with multiple runners on base.

Five. That's it. An 80 percent solo home run rate.

That's not very many. To put things in perspective, teammate James Shields gave up 25 home runs in 2012 as well -- and 16 of those came with men on base (36 percent solo HR rate)! Shields gave up 26 HR in 2011, and eight came with men on base (69 percent solo HR rate). Cliff Lee of the Phillies gave up 26 HR in 2012, and 18 in 2011 ... and gave up seven and eight HR with men on base (73 percent and 56 percent solo HR rates), respectively. Granted, that's only two pitchers, but that's two fairly GOOD pitchers, and their rates of multi-run HR are worse than Hellickson's. And Hellickson also had good -- but not quite as good -- numbers in this regard in 2011, posting a 71 percent solo HR rate.

Right now, we need more sample data to determine whether or not Hellickson's ability to limit HR with men on base is real. Two seasons is a fair amount of time, but it's still not quite enough to determine whether or not this is a mirage of luck, or something in Hellickson's stuff and approach that causes runners to stall out before being able to score.

Looking at advanced stats is a great way to determine just how good a player is. But in the strange case of Jeremy Hellickson, the advanced stats raise more questions than they do answers. Before the Rays know for sure whether they have a player who is just lucky -- or someone with a magical ability to strand runners -- they'll probably need more time to evaluate his performance.