This was inspired by a recent comment by Andrew T. Fisher, regarding groundballs and FIP, and also the significance that groundballs pose to pitching success. We all believe that groundballs matter, but really, just how much do they actually matter?
Fisher made an interesting comment that strikeouts are arguably the most significant component of FIP. To which I countered, has there actually ever been any sensitivity analysis conducted regarding how FIP relates to strikeouts, versus walks, versus home runs? And then I realized, hey, sensitivity analyses are pretty fun, in a very nerdy sort of way, and so why not do one of my own? Though, because I think FIP is still somewhat too abstract and difficult to intuitively relate to, I plan on showing how groundballs can impact something much more basic: ERA.
In order to understand impact, I think it's necessary to establish a framework of reference. In baseball, we have groundball pitchers and flyball pitchers. But what exactly constitutes a flyball pitcher versus a groundball pitcher? Because my analysis is going to focus on how groundball pitchers prevent home runs, which in turn depresses ERA, it's slightly easier to illustrate my point by looking at how "flyballish" a pitcher is to show in reverse why being a groundball pitcher is awesome. This is because we have easy access to statistics on HR/FB rates, but since these stats don't take into account line drives when being calculated, using pitchers' groundball percentages would complicate my math a lot more. And I don't particularly enjoy math.
According to Fangraphs, there were 44 qualifying NL starting pitchers. I'm going to sort of arbitrarily (though I think this makes some sense intuitively) to designate my "flyball pitcher" as the 25th percentile of flyball pitchers and my "groundball pitcher" as the 75th percentile of flyball pitchers. Somewhat ironically, our designated "flyball pitcher" is Rodrigo Lopez, at a 41.0% flyball repertoire. Our designated "groundball pitcher" is Mike Pelfrey at a 32.0% flyball repertoire. So essentially, we say that the difference between a "groundball pitcher" and a "flyball pitcher" is that the flyball pitcher's batted ball profile leads to 9% more flyballs.
Now, here is where the analysis gets a little bit sketchy. I don't have a good way of calculating this, but just by looking at some sample pitchers, I feel fairly confident that a standard pitcher will put around at least 540 balls in play in 200 innings. Last year, Tim Lincecum (highest strikeout rate in the NL, and therefore, intuitively, lowers balls in play rate in the NL) put 573 balls in play in 212.1 innings, which comes out to 540 balls in play per 200 innings. And let's face it, most pitchers just aren't going to be as good at preventing balls in play as Timmeh.
But for the sake of showing the significance of groundballs, let's use 540 as the lower limit of balls put into play. 9% extra flyballs. NL league average of 10.6% HR/FB rate. All of this together, means that the flyball pitcher can be expected to throw 5.1516 HR more than the groundball pitcher. HR have a 1.4 run value above the average outcome of an at bat. So the flyball pitcher can be expected to give up 7.2 runs more than the groundball pitcher, all else being equal. In 200 innings, that amounts to an increase in ERA by 0.32.
Now, maybe it's just me, but in my mind there's a pretty significant difference between the performance of a 4.00 ERA pitcher and a 3.68 ERA pitcher. Add to this fact, that we basically looked at the best case scenario (pretty much every pitcher in the league will put more balls into play than the Tim Lincecum comp used above), and we haven't even looked at the effect of preventing doubles and triples resulting from flyballs, and this tells me that being a good groundball artist can significantly affect your pitching results.
So basically, the moral of the story is that groundball pitchers are awesome. Now, let's go find ourselves some worm-burners!