Jake Diekman, Not a LOOGY; or Hock This!

Jake Diekman has pitched very well for the Phillies this season. In fact, he's pitched just as well this season as he did in his abbreviated MLB work last season. But he has continued to be dogged by a label that would restrict his use, if the Phillies front office and coaching staff cared about things like labels. That label, "LOOGY," is quickly applied to any left-handed reliever who shows a noticeable split in results between left-handed batters and right-handed batters. But not all splits are equal. While it is undeniable that Diekman is not nearly as good against right-handed batters as he is against left-handed batters, his numbers against righties don't justify restricting his use to lefties whenever possible.

First, Diekman is just as good this year as he was last year, despite the large difference between his ERAs in each season. Take a look at this fancy-dancy table if you don't believe me.

Created with the HTML Table Generator

Once you look past Diek's ERAs in the two seasons you see that his ERA-estimators are within margins of error of one another and his K%-BB% ratio has actually improved. We can attribute his worse results this season mostly to aspects of his performance that are much less in his control: hits on balls in play and strand-rate. Keep in mind that only 6 or 7 earned runs makes up the difference between Diek's ERA and his FIP this season. Over the short term that could easily be due to the random sequence of events like hits, BBs, and homers. A homer after a walk is obviously twice as bad as the reverse, but that sequence is not particularly due to Diek's talent or skill. Based on these stats over the last 90 IP, I am confident that Diek has become a top-shelf reliever, although not an elite one like Ken Giles seems to be.

[Aside: Diek's 1st and 2nd half splits from this season are worth mulling over. He had terrible results in the 1st half and has phenomenal results in the 2nd. Is this due to an adjustment? Is it just luck? I'm not sure. I'd love your thoughts.]

But you didn't come here to read about Diek's overall performance. You came here to see me defend the crazy claim that Diek is not a LOOGY, despite his quite patent large handedness splits. Well, what is a LOOGY? LOOGY, as we all know, stands for Lefty One Out Guy, but I doubt any pitcher is used in accordance with that definition anymore. I take a LOOGY to be someone who is excellent against lefties but bad enough against righties that his exposure to them should be limited drastically. So, I'll be arguing that Diekman isn't bad enough against righties. Indeed, he's perfectly average. And on the Phillies he is a good choice to face righties given certain conditions.

First, let's look at the numbers. Here are the splits for the Phillies relievers who have pitched more than 10 IP against righties, and the numbers for the average reliever in 2014 against all hands.

Created with the HTML Table Generator

Of course, small sample size caveats apply here, but it is very difficult to get an adequate sample for relievers within a time-frame in which their true talent has not altered. So, we'll just go with what we have. If we look at FIP and xFIP, Diekman is smack in the middle of Phillies relievers with regard to what he controls most in his performance. He remains excellent at generating Ks against righties but surrenders a lot of walks and a lot of hits. We might be tempted to say his BABIP is due to regress, but, as TGP's excellent prospect writers have reported, Diekman's motion, in which he sweeps his arm from out wide, gives righties a long view of the ball and adds no advantageous angle to ball's flight-path because, unlike left-handed hitters, the ball is not sweeping across the righty's body as it approaches the plate. Despite this disadvantage, Diekman still has an above average ability to suppress homers, which is probably due to how hard it is for any batter to adjust between his fastball and his slider. So, even though Diekman will let a lot of righties reach base, he is also very good at limiting the damage that ensues because he strikes out almost every 3rd righty. As the average reliever row in the table shows, this makes Diekman the run-production equivalent of an average reliever.

In the Phillies bullpen this year, there have only been two pitchers who were healthy and clearly preferable to Diekman against righties. They are Giles and Papelbon. Mike Adams would be preferable as well, but he'd have to be able to throw the ball 60.5 feet first. Aside from those three, Diek is about as good as Antonio Bastardo and Justin De Fratus. We need not even consider the dreck that the Phillies have tried to pass off as relievers behind those guys. Bastardo might have a slight edge because he has a longer history of being decent against righties and is better than Diek at getting the ball in on righties hands, but the evidence is not clearly in Bastardo's favor.

So, should Diekman be used without much thought as to whether he's facing righties or lefties? Well, no and yes. No, because, first of all, he's not that bad against righties. And second of all, no, because he is in a tie for 3rd best reliever against righties in the Phillies bullpen and trying to limit Diek's exposure to righties will require using worse relievers at other times in the game, possibly against lefties even. Yes, because there will be times when it is worth bringing in Giles or Paps to get an extra out because a righty is batting in a high leverage situation or is a really good hitter.  And yes, because sometimes he might enter to face the heart of an order who are all righties and that will turn him into a worse than average reliever. But these limitations are not the strong restrictions placed on pitchers labeled LOOGYs.

In truth, it would be better to look at the PA outcomes Diek tends to produce against righties and match those to the situations in which he faces righties. If you need a groundball don't use Diek. If you need a K, he's a good choice. If you want to limit extra base hits, Diek is a better choice than De Fratus but not Bastardo. If a walk is extremely bad, don't use Diek. If you prefer an outcome that doesn't involve a ball in play, use Diek. And we can get even more fine-grained about the criteria for these decisions by, for example, using him against righties whose profiles are more susceptible to his arsenal. At any rate, it is not helpful to think of Diekman as a LOOGY. He's better than that. Remember that the next time you want to curse Ryne Sandberg for being a bonehead who doesn't understand managerial concepts invented 20 years ago.