Thursday, May 26, 2016

The Reinvention of Richard Hill

Blind player resumes is such a common thing in fantasy analysis, it's almost a cliche. But there's an old cliche about cliches: "They're only cliches because they're good." So let's kick off this breakdown with some blind player resumes that you probably know the answer to if you read the title of this post:

Pitcher A:
1.97 ERA, 10.49 K/9, 2.70 BB/9, 48.5 GB%, 10.8 SwStr%, 2.57 FIP, 3.23 xFIP

Pitcher B:
4.72 ERA, 8.62 K/9, 4.34 BB/9, 35.1 GB%, 8.6 SwStr%, 4.42 FIP, 4.45 xFIP

Which pitcher do you want?

Don't think too long, because, as you might've surmised, both pitchers are Rich Hill, the 36 year old breakout who's current run of dominance resembles the best stretch Max Scherzer has ever put together.

In 84 career starts, Hill has ten games with double-digit strikeouts. Five of them have come in his last 14 starts. In the last calendar year, only Clayton Kershaw and Jake Arrieta have lower ERAs than Hill. Only Kershaw has a lower ERA and a higher K/9. That from a guy with a career 4.72 ERA over 70 starts from 2005-2013. That from a guy who bounced up and down from the minors more times than I can count.

What's been behind the amazing, late-blooming renaissance?

(Unlike most of my breakdowns, I'm not comparing Hill's 2016 to his 2015. Instead I'm comparing his 2015-2016 with the rest of his career. For our purposes, "Old" Rich Hill is from 2005-2014. "New" Rich Hill is from 2015-present.)

First, he's managed to improve in all four of the "Holy Quaternity" stats (K%, BB%, GB%, PU%). That's not easy to do; when you improve your walk rate, your strikeout rate usually falls as well as you pound the zone and get more balls in play. When you increase your grounder rate, it usually means you're lowering the launch angle of balls hit against you, thus eliminating some pop-ups. Usually, you make trade-offs within the four stats. So, obviously a pitcher improving in all four catches my eye.

To validate K rate spikes, I first look at where the extra strikeouts are coming from; what's driving them, in other words.

For Hill, I believe that answer is multi-faceted. First and foremost, he is getting ahead in the count far more often. "Old" Rich Hill got to two strikes on 54% of batters he faced. "New" Rich Hill is doing so on 64%. This is partially supported by a two percentage point increase in first-pitch strike percentage. I suspect it is also supported by a decreased Contact% (if one of the first two strikes is put in play, he can't get to two strikes).

Once he gets batters in a hole, he's also been more efficient at sitting them down: "Old" Rich Hill converted 40.7% of his two-strike counts into strikeouts. "New" Rich Hill is converting 45.9% This also is supported by the drop in Contact% as well as SwStr%. I believe it is also supported by a Fool% (percentage of pitches that are in the zone and taken plus percentage of pitches that are outside the zone and swung at) that has improved from 30.0% to 33.1%. It's interesting to note that Hill's pitch mix with two strikes has hardly changed; he still has about a 40/60 fastball/curveball split with two strikes.

Last but not least, it's impossible to ignore the improved whiff rate as a source of additional strikeouts. "Old" Hill had a 8.6% swinging strike rate, whereas "New" Hill has a 10.8% rate. Obviously, more swings and misses leads to more strikeouts, since the vast majority of strikeouts are of the swinging variety.

In the past, Hill could rely on his curveball for whiffs but his fastball wasn't good for much of anything. But now, he's getting a ton of whiffs on his fastball: it's 12.4% swinging strike rate is nearly double the league average, and it's whiff/swing ratio is the best in baseball, and by a wide margin: his 35.59% ratio tops David Price's second place mark by a whopping six percentage points, which is about the same as the difference between Price and fourteenth place Noah Syndergaard, or the difference between Syndergaard and 45th place Cody Anderson.

How is a pitch that averages 91.5 mph getting more than 1 1/2 times as many whiffs per swing as a pitch (Syndergaard's fastball) that averages 99.3 mph? And whatever skill allows it to get such gaudy whiff rates, why is it just now appearing?

I had a few hypotheses, and to examine them, I put together the following table of info about his fastballs over the years:

There's a handful of interesting observations we can make here. Obviously, there's the aforementioned spike in swinging strikes. There also appear to be three distinct periods, which I separated out. I think of them as "The Four-seam Period" (2007-2009), "The Two-seam Period" (2011-2014), and "The Reinvention Period" (2015-2016). 2010 is an outlier, mostly because Hill only tossed 32 fastballs that year.

In the "Four-seam Period", Hill was throwing a four-seam fastball, obviously, and was getting very good vertical rise, but mediocre horizontal movement. In the "Two-seam Period", that was reversed.
 He got plenty of arm-side run, but his vertical movement was below average.

Now, however, Hill is combining the two, throwing what PITCHf/x classifies as a four-seam fastball, but that has similarities to two-seam fastballs or sinkers (it's most comparable pitches are Santiago's sinker and Miley's two-seamer). He's combining above average vertical rise (40th out of 103 qualified pitchers) with elite horizontal fade (3rd).

Not only is more movement a good thing by itself (as long as you can still control it), but the direction of Hill's added movement is in the exact opposite direction of his curveball. Hill's fastball has some rise and great arm-side run, while his curve has the 12th most vertical drop among qualified pitchers (second only to Kershaw among left-handed pitchers) and is tied for the 12th most glove-side break (1st among left-handers). All told, Hill's "new" fastball is getting over two feet of separation from his curveball (meaning that if he threw his fastball and his curveball at the same initial trajectory, their movement would have them crossing the plate at two separate points over 24 inches apart - second most in the game).

Changeups and other off-speed pitches are often evaluated by how distinct they are from the pitcher's primary fastball. Obviously, if the batter doesn't know what direction the pitch will move in, it becomes extremely difficult to hit. Given Hill's heavy curveball usage (over 50% of all his pitches), I would argue that his fastball should be evaluated in comparison to his curveball, since his curveball, not his fastball, is his primary pitch (or, at least since the beginning of his renaissance). Not only is Hill using his fastball more and more like a change of pace from his curveball, but it's becoming a better, more distinctive change of pace.

That's been a key for Hill. Check out this chart that plots his fastball/curveball separation against the swinging strike rate on his fastball:

While improving the separation from his curveball has been a big part of his improved fastball results, I believe the pitch has also gotten better in isolation. It's been shown that fastball spin rate correlates with swinging strikes. Look at how Hill has gradually increased the spin rate on his heater through the years:

That, too has had a correlation with his fastball whiff rate, though we sort of have a chicken-and-the-egg dilemma: did his spin rate cause the increase in whiffs, or was it the change in movement? Then again, didn't the spin rate change influence the change in movement?

While Hill's fastball has been the most enigmatic part of his mid-thirties breakthrough, I'd be remiss if I didn't devote more attention to his devastating hook. Observe:

As I mentioned above, he gets the most glove-side run on his curve of any left-hander, and the 12th most vertical drop of all pitchers (only Kershaw gets more drop among left-handers). It gets the 10th most whiffs per swing of any curveball, and the 5th most grounders. It also gets the 9th most pop-ups. It's been worth 4.8 runs above average, which is good for first in the American League (5th in the MLB).

Verdict: Hill features one of the best curveballs in the game, and has plus movement on his fastball. He's also seemingly figured out the perfect way to use the fastball to complement the curve. Moreover, his fastball is seemingly a different pitch from his pre-breakout fastball. As such, it's mostly safe to ignore the "old" Rich Hill and focus on the new. Obviously, some regression should be expected, but now's your time to buy while his owner might think he's still a small sample fluke. He's easily a top 30 SP going forward, with top 15 upside. Not bad for a 36 year old Quadruple-A player.

Don't forget to follow Pitching Breakdowns on Twitter (@PitchBreakdowns)!

No comments:

Post a Comment