What's wrong with Oakland A's reliever Liam Hendriks?

There's a lot to like about relief pitcher Liam Hendriks. He throws hard, often operating in the high-90s last year for Toronto. He has always kept the walks low, but after moving from the rotation to short relief in 2015 he began striking out more than a batter per inning to help push his ERA below 3.00. He's Australian, and the last time the Oakland A's had an Aussie reliever it was Grant Balfour closing games for back-to-back division winners. Sure, his acquisition came at the cost of a popular player in Jesse Chavez, but the bullpen needed serious help and this seemed like one guy who could provide it.

Unfortunately, the Liam Hendriks Experience has not included the electric stuff that used to put opponents into a haze. Instead it's felt more like a Nickelback album -- an import from Canada featuring lots of hits, but I'm yet to meet anyone who actually enjoys them.

How bad has it been for Liam this season? Here's a side-by-side comparison with last year (stats are shown as rates per nine innings):

The biggest problem is that he's allowing nearly twice as many hits as he did last year, with his opponents' average increasing from .238 to .397. His strikeout and walk rates are still fantastic, and in fact he's only allowed one walk in his first nine outings. He's allowed two homers already after serving up just three all of last season, but that's not enough to sink him on its own. Clearly it's the overwhelming parade of simple base hits that have turned him into a gascan so far.

It's tempting to look at these numbers and just brush them off as a fluke. Hit rates fluctuate for most pitchers, and even more so for relievers, so this is probably just an oddity that will work itself out. Heck, his FIP is a sparkling 3.23, a full five runs lower than his ERA, and often that's a sign that you should have patience and let the universe work itself out. But baseball isn't always that simple, so let's poke around and see if we can learn anything about why Hendriks might be struggling.

The best way to look for red flags in a player's numbers is to use the past as a baseline and then see what has changed. However, Hendriks really only has one year of relevant history to look back on. Prior to 2015 he was an ineffective starter, and it was only when he moved to shorter stints in the bullpen that he revamped his entire approach and found success. Therefore, all we can do is compare last year to this year. Here are a few differences, with guesses for whether there is anything to them.

Velocity/Movement

The big story of Hendriks' 2015 success was that he started throwing harder. Rather than pacing himself at 90-92 mph for a starter's workload, he pumped his fastball up to 95 mph on average, and according to Brooks Baseball he averaged 97 in a half dozen outings. So far this year he has dropped down about 1 mph -- Brooks and FanGraphs disagree slightly on where he's at now, but they agree that he's gone down by about one tick.

That's not something I'm worried about at the moment. It's common for pitchers to ease up their velocity as the summer heats up, and in fact Hendriks threw harder this April than he did last April. And even if he never recovers that lost tick, I don't think one mile of velocity accounts for five extra runs on your ERA.

As for the movement on his pitches, this isn't an area of analytic expertise for me. But when I look at this graph, it seems to me that his pitches are moving in roughly the same ways as last year. Maybe a little bit less horizontal on his four-seam fastball, but nothing is jumping out to my layman's eye.

Pitch selection

For his entire career, both starting and relieving, Hendriks has been largely a three-pitch guy: four-seam, sinker, slider, with none of those pitches making up a majority. (There's also a change and a curve that get tossed in now and then.) This season, he's thrown his four-seamer a full two-thirds of the time, with his slider experiencing a heavy reduction and his sinker all but disappearing. I don't have an explanation for this, but it remains relevant in the next section ...

Pitch values

His four-seam fastball was his bread-and-butter last year. It hadn't been in the past, but once he started throwing it harder it became his best pitch. This year, it's getting absolutely smashed and it's been the worst of his three primary offerings. And remember, he's throwing it way more often as well. So, he took his best pitch and began throwing it more, and now it's both his worst and most-used pitch. I don't know why that is, especially since the velocity and movement don't seem that much different. Meanwhile, the slider that he stopped throwing as often appears to be his most effective pitch so far.

Batted balls

Last year Hendriks was able to keep the ball on the ground, which makes sense for a sinker/slider guy. His groundball rate 46.3%, not an extreme rate but still in the top half. This year, that rate has gone down to 37.2%. That's not a problem on its own, but we're looking for anything that has changed. Batters are elevating the ball more against Hendriks than they did before, and that doesn't seem like a coincidence considering he's throwing his breaking pitches less often.

Here's another thing I don't have an answer for: batters are going to the opposite field a whopping 46.5% of the time against Liam (career rate: 26.9%), which ranks second in all of baseball behind only Jim Henderson of the Mets. I imagine that this is a metric which will normalize over time, as no pitcher has exceeded 39% in that category since 2012 (Ryan Cook, as chance would have it).

Plate discipline

Over at FanGraphs, you can look up "heat maps" of the strike zone based on various measures -- how often does the pitcher throw to certain locations, how often do opposing batters swing at those spots and make contact, etc. I played around with those for a while, and rather than embed a bunch of them here I'll just sum up my findings.

Against righties last year, Hendriks generally kept the ball down and away. However, batters swung more often at the up-and-in stuff and didn't do a lot with it; the only thing they could get a hit off was down-and-in. This year, he's still mostly keeping it down and away, but when he doesn't the batters are taking advantage of every other location. In particular, they are slamming anything up in the zone, whether on the inside or outside half of the plate.

As for lefties, last year he kept the ball on the outside half, which worked well because batters couldn't do much with it and made their most contact up-and-in. This year, he's all over the place against lefties and they're feasting on the inside half of the plate.

Overall, FanGraphs' plate discipline stats say that batters are chasing out of the zone less often, but when the ball is in the zone they swing and make contact more often than last year. He's seen a huge drop in swinging strikes and first-pitch strikes.

Conclusions?

This post was primarily supposed to be an exploratory mission and my goal was just to present the above information, but I guess I owe it to you to take a crack at drawing a conclusion.

Let's put it all together. Hendriks is still throwing hard and hitting the zone, but it's questionable whether he's commanding the ball and getting the same quality of strikes. He's favoring his four-seam fastball more than ever, but it has stopped being an effective pitch. Batters are making more contact and whiffing less, partly because they are chasing less often, and when they do make contact they're putting the ball in the air and to the opposite field.

There are a few possibilities. He could be hurt, given a slight drop in velo and an avoidance toward his breaking ball, but I don't think that's the case and I'm not going to speculate on it. He could have been a fluke in 2015, but his success was built upon sustainable FIP-based metrics rather than simple BABIP luck; even some regression should leave him a quality reliever. Or, perhaps the league is starting to figure him out a bit, and he needs to adjust.

Personally, I keep coming back to that slider. That's where a lot of his whiffs came from last year, and so I don't think it's a coincidence that both (slider and whiffs) decreased at the same time. Why is he shying away from it, and could bringing it back help him? Does he need it in order to keep hitters on their toes, so that they can't sit dead-red on his heater and settle for line drive singles the other way? Or is all of this one big fluke, with the FIP gods laughing hysterically as they spell "80085" on their calculators?

We're only through a month of the season, and just about every baseball stat is still subject to small-sample hijinks at the beginning of May. Jason Hammel leads the NL in ERA, Neil Walker is tied for third in MLB in homers, and Justin Upton has a .606 OPS while leading the league in strikeouts. Many things will change between now and September, for better and for worse. Hopefully Hendriks' results will be one of those things.