With all due respect to #J_the_Man for his recent well written piece and #Phil Coke's Brain for his continued insanity, I wanted to go more in depth regarding this topic. More specifically, I don't think there is any "trying" to fix Phil Coke. The only way to "fix" him, is to cut him. I have written a few articles in the past that have seemed to gain positive reviews. That said, this is the article that I've always wanted to write for this site.
Let's start at the beginning. Phil Coke was a lowly touted prospect out of junior college drafted in the 26th round by the New York Yankees. He had a surprisingly efficacious cup of coffee in 2008 as a rookie posting a 0.61 ERA, a 1.63 FIP, and 0.5 WAR over ONLY 12 relief appearances in pinstripes. He had a less than stellar 2009, sporting a 4.50 ERA, 4.68 FIP and only 0.3 WAR in over four times the innings. This was also the first year that we saw his extreme righty/lefty splits as he posted a 3.90 FIP against lefties, but a 5.89 FIP against righties (ranking #374/399 in all pitchers who threw 20 innings vs. RHP that year).
Phil Coke became a Tiger in 2010 as a throw in to the throw in of the Max Scherzer/Austin Jackson deal. He had a modest 2010 by a 3.76 ERA, 3.23 FIP, and 1.1 WAR over 64 innings. That year, he was the Tigers' second most effective reliever (He was actually tied with Joel Zumaya with 1.1 WAR, but Zumaya only pitched 38 innings that year). Still, he was more than doubly effective than the then closer Jose Valverde who posted 0.5 WAR in 63 innings. However, despite his successful year, his righty/lefty splits were still evident with a 2.56 FIP against righties, and a 3.77 FIP against righties. This can mostly be explained by his K/BB ratio. While he posted a 10.05 K/9 vs. 2.51 K/9 against lefties (ratio of 4), his numbers severely regressed on the other side of the plate with a 5.25 K/9 and 4.50 BB/9 (ratio of 1.17).
That year turned out to be his audition for a starting role. Over the next year, despite poor "classic" statistics (wins, ERA, etc.), he turned out to be very effective, churning out a 3.57 FIP over 108 innings which equated to 1.8 WAR. However, this is the first season we really witnessed his weaknesses. Despite being effective as a starter (3.67 FIP over 74 innings), his peripherals were much worse. Although his FIP was relatively similar (3.35 that season in 34 innings of relief), his peripherals were much worse. As a starter, he posted 4.46 K/9 rate vs. 3.62 BB/9 rate (ratio of 1.23). BUT, as a reliever, he posted an 8.47 K/9 rate vs. a 2.65 BB/9 rate (ratio of 3.20). Thus, Phil Coke was always best suited in the reliever role.
The splits became even more evident in 2012 (2.87 FIP vs. lefties, 4.29 FIP vs. righties), but his success was mainly dictated by his role in the 2012 playoffs. As Jose Valverde imploded (Which was neither the first or last time. Please pray for all Mets fans), Phil Coke was phenomenal in the playoffs. Over 10 and 2/3 innings, he struck out 13, walked only 2, and only surrendered 1 run over that postseason (although that one run ended the World Series in game 4). However, over 3 playoff series, Coke faced 39 batters. Of those, 20 were left handed, ONLY 8 were right handed, and 11 were switch hitters. Given the Lefty dominant lineups of the Giants and Yankees, as well as some strategy from Jim Leyland, we can see why Phil Coke had so much success in the 2012 playoffs.
Last year was extremely frustrating for him. Although he was just as effective against lefties (2.76 FIP in 20+ IP), he was even worse against righties (5.76 FIP, ranking #467/503 with at least 10 IP vs. righties). This was again demonstrated by his K/BB rate which was amazingly reversed with only 5.09 K/9 vs. 7.13 BB/9 vs. righties in 2013. As we all expected by the aforementioned trends, Phil Coke is a LOOGY (As discussed by Jonah Keri at Grantland:
"LOOGY (exp.) — acronym for Left-handed One-Out GuY, meaning a lefty reliever who comes into the game usually for the express purpose of facing one left-handed batter. While that kind of matchup can help teams, it comes at a cost, with benches being shrunken to comically small sizes, thus limiting a manager’s options should he want to pinch hit, pinch run, or substitute defensively late in games.")
Ironically, the one situation Phil Coke should have pitched, was the infamous Ortiz HR during last year's ALCS. Ortiz has a career .111/.150/.167 line against Coke in 18 ABs while posting a .262/.332/.456 against all lefties since 2008.
This year, Phil Coke has again been atrocious. We haven't even mentioned his shutdown to meltdown ratio. A "shutdown" essentially means adding 6% to your team's expected win percentage in a single game, while a meltdown means subtracting 6% from your team's expected percentage (briefly described by fangraphs here and further extrapolated by Jonah Keri here). Since 2012 to present, there are only two pitchers with more "meltdowns" than Phil Coke. And those pitchers (Nate Jones and Matt Belisle) both have more than double the "shutdowns" that Phil Coke has (41, 59, and 19, respectively) over that same period.
To me, the answer was to cut Phil Coke during the offseason. We had a perfrectly reasonable LOOGY option in Darin Downs who posted better overall numbers (3.05 FIP vs. lefties, 3.98 FIP vs. righties) than Phil Coke. Downs' LOOGY success was even more evident in 2012 when he posted a 2.03 FIP vs. lefties. To me, it seems that the more effective and cheaper option (Downs), was probably the better option.
BUT, despite Coke nearly blowing one game (4/5 vs. BAL) and completely blowing another (4/8 vs. LAD) in his two sole appearances this year, I doubt he will go anywhere anytime soon. Ian Krol, the throw in to the (awful) Fister deal, had a similar split rate with a 3.14 FIP against lefties and a 5.71 FIP against righties last year. The only other lefty in the bullpen is actually a starter in Drew Smyly. He still has a split (2.52 FIP vs. lefties, 3.50 FIP vs. righties over his career), but his overall numbers are so good that it doesn't matter who he's facing. So unless Casey Crosby overcomes his control issues (unlikely), Robbie Ray progresses more quickly than we expect, or we trade for something better, Phil Coke may be the best LHP in the Tigers bullpen. Let's just hope Brad Ausmus recognizes that he shouldn't pitch to right handed batters, especially in high leverage situations.
Because, unfortunately, pointing at pop-ups and slamming your glove on the ground doesn't make any more outs or win any more games.