When the Astros claimed Lucas Harrell off waivers in July 2011, our initial question was "Which OKC starter does he replace or which Houston reliever does he replace?" He spent six weeks at Triple-A, going 5-2 with a 1.71 ERA/1.25 WHIP and a .275 BABIP.
He got the call-up to Houston in September, and in his Astros debut threw 5.1IP against Milwaukee, allowing 3H/0ER, 4K:2BB. He came out of the bullpen a few times over the month, made an inefficient - but not terrible - start in a 19-3 loss to Colorado, and closed out his season with a perfect inning against the Cardinals.
Harrell was in the mix for a rotation spot in 2012, mainly because he was out of options. He was initially ruled out of the rotation in the middle of Spring Training but pitched better as March 2012 went along, which was basically the opposite of rotation favorite Jordan Lyles - who was 0-3 with an 8.41 ERA. Lyles was optioned to OKC and Harrell got the rotation spot.
And, for a season, was perhaps the Astros most reliable starter. In 2012, Harrell made 32 starts. He went at least seven innings ten times; 6+ IP 16 times. In 19 starts, he posted a Game Score of 50 or better, 11 were a 60 or better. The team actually went 14-18 in his starts, meaning they went 41-89 in games Harrell did not start.
Fast forward a year (don't worry, we'll come back to 2012), and Harrell has been a disaster: He lost his rotation spot and has been ineffective in the bullpen. Among all MLB pitchers who have thrown at least 110IP (Harrell is at 122.1IP), only the Padres' Jason Marquis has a lower WAR as assigned by FanGraphs. Harrell is one of three pitchers with 110+ IP to post a negative WAR: Jerome Williams (-0.3), Harrell (-0.5), and Marquis.
It's not bad luck
In 2012, Harrell had a .289 BABIP - a little lucky, but not a Happian BABIP. So far this season, Harrell has a .301 BABIP, which is perfectly normal. His ERA is 5.37, and while that's almost exactly in line with his 5.41 FIP, it's decently higher than his 4.92 xFIP. But that still isn't very good.
Harrell has allowed 5.15 BB/9, an increase of more than 1.5 BB/9 from 2012. 12.6% of the batters he has faced in 2013 have walked. He's just throwing fewer strikes. 1328 of his 2293 pitches have been strikes, or a tick under 58%. In 2012, 62.4% of his pitches were thrown for strikes.
Fewer groundballs, more HRs
In 2012, Harrell induced groundballs at a 57.2% clip. This year that's down to 52%. What's more, 15% of the flyballs Harrell is allowing are leaving the yard - 4th-highest in baseball (min. 110IP) - and up from 9.7% in 2012. Okay, so he's getting more flyballs, and a higher percentage of those flyballs are going over the fence.
His fastball has been terrible
While Harrell's velocity hasn't seen a significant dropoff (92.1mph in 2013; 92.5mph in 2012), It's effectiveness is seriously in question. FanGraphs put his wFB (Fastball runs above average) in 2012 at 13.4. This year? Wait for it...-14.3. Pitchf/x doesn't show such a marked difference (-1.3 in 2012 to -8.1 in 2013), but the damage is being done in that his fastball just isn't a good pitch this season.
Nobody is fooled
Hitters are swinging at 39% of Harrell's total pitches - down 3.4% in 2012. Yet they're making more contact (+1.2% from 2012) overall. Harrell is also throwing far fewer of his pitches in the zone - 41%, down from 49.2% in 2012. So if you know that Harrell isn't going to throw strikes, why swing?
Here's how it breaks down:
Harrell held righties to a .663 OPS in 2012 (.883 OPS in 2013); .713 OPS for lefties (.783 OPS in 2013). He also had a .647 OPS at home (.850 OPS in 2013), .712 on the road (.800 OPS in 2013), with an identical .293 BABIP.
Here's the cycle: Harrell isn't throwing as many strikes, so hitters aren't swinging as much. Harrell gets too much of the plate with his largely ineffective fastball, and it's leaving the park at a markedly higher rate. If he tries to nibble, he misses, and walks the batter. He's leaving the ball up, and hitters are driving it out of the park. That, my friends, is what has happened. And it's starting with his fastball.