Phil Humber makes his season debut Monday night, and since it’s the 9th game of the season already, that doesn’t speak volumes about organizational confidence in him. It mirrors his treatment from last season, when he was a completely unknown quantity and hidden from the rotation until the 8th game, and even stuffed in the bullpen a couple of times. It was an ignoble beginning, but he was in no place to complain.
From there he improbably cruised through his first 15 starts, clocking 101.2 IP with a 2.57 ERA. All-Star numbers, from a guy with no business going to the All-Star game. Opposing batters only had 22% of their balls in play fall for hits, which ain’t no kind of normal (It’s around 7-8% shy of normal). There was plenty of talk that it was a fluke, but since the expectation for him was nothing, a flukey 100-inning stretch was more than enough to be grateful for.
After that, Humber got what he deserved, his opposing batting average with balls in play for the rest of the season was .364, his ERA was 5.61, and he barely averaged over 5 innings a start.
Humber wasn’t any kind of impressive upon arrival to the Sox. His fastball velocity and movement remotely special, his change-up was worse, and he had an injury history that was such that if he read it aloud, Rich Harden would offer to buy him a drink.
He had a good curveball, but if he couldn’t throw it for a strike early in the count, he was forced to throw strikes with below-average offerings, which is typically how long home runs are born. How the hell he got through some of his earliest starts in this manner is a mystery. How the Sox managed to stagger to such an awful start while Humber was spinning straw into gold is also a mystery, but the kind no one wants to really think about.
Well, it’s not entirely a mystery how Humber did it. He’s a journeyman pitcher working under Don Cooper, so of course he learned a new pitch. Cooper installed a slider, and to fantastic results considering Philip’s only beginning to get familiar with it. FanGraphs rated the pitch as 2.7 runs above average on the year.
Yet Humber was only really tapping into its potential of the pitch when he was making his miracle run. According to TexasLeaguers, he was only throwing his slider 10.8% of the time during that stretch, and getting swings and misses 17.1% of the time. For his less charmed latter half of the season, he ramped up his usage of the slider to 18.3% of the time, and it became a devastating offering, generating swinging strikes 23.4% of the time. In turn, he de-emphasized his ineffective change into more of a show-me offering, which it sorta always was.
Ostensibly, this approach was disastrous because Humber’s results became worse. However he struck out 8.03 batters per 9 IP while only walking 2.28, which is a superb ratio. It was also accompanied by a spike in home runs allowed, and a lot more line drives being sprayed around the field. That can’t be ignored, as some of that was the league adjusting to him, but some of that could also be Humber tiring under the biggest inning load of his career, and his luck evening out in a rather painful way.
Humber finished with a 3.75 ERA on the season, with no glaring statistical anomalies to suggest the 163 innings he threw weren’t of that quality. His xFIP was 3.86, his SIERA was 3.89, and his True ERA was, wait for it, 3.75. For what it’s worth (little), he backed that up with a solid Spring where he had a 3.18 ERA and struck out 13 batters in 17 IP. If the ability to be a major league pitcher has suddenly left him again, it didn’t show up in March
He’s not the above-average strikeout pitcher he was in the second half, but he’s also not operating on hopes and prayers like the 1st half. He’s a control pitcher, with two different breaking balls he can turn to and a legitimate major league out-pitch. That’s not a star, but it’s a hell of a lot more than what the typical 5th starter brings to the table.
It should certainly be enough to give the Orioles trouble.