Trying to Make Sense of Phil Hughes' Odd Peripherals

So far in 2012, Phil Hughes has a K/9 that is 0.72 points higher than his career average, and 0.86 points higher than league average. Great, right? Well, it might be if he didn't also have a HR/9 almost a full point higher than the league and a batting average against that is .017 points higher than it has been for his career. These numbers just don't make a ton of sense.

Been saying that Hughes has the oddest peripherals that I've seen in a while. High K's, low BB, but tons of homers and hits. Well....

— Bloggie McBlogger (@yagottagotomo)

Using PI, one starter with at least 50 IP has had a season with a K/9 over 8, a BB/9 under 2.5, a HR/9 over 1.5, and H/9 over 9.

— Bloggie McBlogger (@yagottagotomo)

Moshe Mandel tweeted this yesterday, which puts into perspective just how odd Hughes' 2012 has been. The only other starter with those peripherals that Moshe referenced was James Shields in 2010. Ridiculous. What is causing that to happen this year with Phil Hughes, though? How do you begin to explain how he has such a good K/9 ratio, but such terrible HR/9 numbers?

Nailing down one specific reason why Hughes is having success with strikeouts but giving up a ton of home runs is not an easy task. We know that all the Yankee starters have been giving up home runs at a rate that everything in the world says is unsustainable, so there's no reason to believe that Hughes isn't just part of that bad luck statistic that will eventually normalize into something that doesn't continue to make everyone throw up their arms in disgust.

Home runs and strikeouts have just been weird for Hughes so far in 2012. If the home runs eventually normalize, we are left with the improved strikeout numbers to explain; and since we are banking on regression for the former, trying to figure out what is causing the latter seems like the natural direction for this post to go.

(via Joe Lefkowitz)

The bulk of the swinging strikes that Hughes is getting are on the high fastball, as opposed to having batters chase the low pitches that he intentionally tries to bury. His whiffs on both his curveball (8.1%) and changeup (8.6%) are below the league average of 11.63% and 12.63%, respectively, but still higher than they have been in the past for Hughes. The fastball has been thrown with better comparative results, resulting in an above league average (6.03%) whiff rate of 8.0%. That's when batters miss, though. When they don't miss, you get what we all know has always been the downfall of Hughes' attempt to pitch deep into games: the foul balls. Hughes' fastball in particular has been fouled off 25.9% of the time, sailing above the league average of 19.93%. Stop doing that, Hughes. Just stop.

Swinging strikes on Hughes' elevated fastballs alone aren't really the entire story, because he did basically the same thing with that pitch in 2010. If you break down Hughes' pitch selection, the one thing that really stands out is the fact that he is throwing the curveball a little less and the changeup a bit more (10.7%) than he has in the past (4.3% for his career), mainly to left-handed batters. In 2010, Phil only threw 86 changeups, and he's already thrown 116 of them this season. Hughes' changeup has been the most valuable that it has been in his career (probably because he's decided to actually throw it) with the exception of 2010 so far this year, and his second most valuable pitch this season behind his fastball. It is the only one of his pitches that hasn't been hit for a home run in 2012.

That may be insignificant for all I know, but when someone has better strikeout numbers than they have since his days in the bullpen in 2009, you can't simply write off a change like that as nothing. His curveball, while still below league average at swing-and-misses, has generated a decent amount more whiffs so far than it did for him in 2010, and a better than league average strike% in 2012. Maybe that combined with the added quantity and effectiveness of Hughes' changeup has batters swinging a little late on his fastball, generating more strikeouts all around than in the past.

Having another effective pitch that batters know you aren't afraid to throw keeps them just a little more off-balanced and leads to a little more guessing at which pitch is coming at any given time. Weaker contact because of incorrect guesses and off-balanced swings generally yields good results for a pitcher. The right-side up red triangles below show that even when batters aren't swinging and missing at the changeup (blue right-side up triangles), it's inducing a good amount of groundballs. Groundballs are good, because those don't end up in the seats. Usually.

(via Joe Lefkowitz)

Below is a break down of Hughes' results for 2012 (left) and 2010 (right), via Texas Leaguers:

More strikeouts, more groundouts, less lineouts. All good. The home run numbers? Well, regression, please come soon. Granted, the sample size is an entire season for 2010 and only one-third of a season for 2012, but at least a good number of the statistics that tend to make a difference seem to be trending in the right direction.

It's impossible to say with certainty why, exactly, Phil Hughes is striking out more batters this year than before. Maybe it's voodoo magic, his curveball being thrown for more strikes, or maybe it's the fact that his changeup is being thrown more often and better than it has been in the past. Whatever is helping his effectiveness, it seems like he's been doing it better lately than he was at the beginning of the season, and that's all we can really hope for. People, including myself, have noted that Phil Hughes has been given a ton of chances to succeed in the Yankee rotation, but when he turns in performances like he did against the Detroit Tigers on Sunday, all those chances seem like they might be worth it if Hughes can really put it together, because we know the kind of pitcher Hughes is capable of being. He's completely maddening at times, but only because you know he has it in him to be great.