Alburquerque Nearly Un-hittable

It is well documented that Tigers reliever Al Alburquerque is striking out batters at an incredible clip.  With 47 strikeouts in 29 2/3 innings, he leads the American League with 14.3 strikeouts per nine innings.  If he keeps up that pace, he will have one of the top ten strikeout rates in the history of baseball.

His 2.74 ERA is a little less impressive, but we know that ERA is not a good statistic for relief pitchers.  One reason is because relievers often enter games with runners on base and give up another pitcher's runs.  Conversely, they may leave games with runners on base and get charged for runs that scored due to hits allowed by other relievers.

One alternative to ERA is the Fielding Independent Pitching (FIP) statistic.  FIP  estimates what a pitcher's ERA should be based on fielding independent statistics - walks, hit batsmen, strikeouts and home runs.  FIP is independent of fielders and also the performance of other relievers.  Alburquerque currently has a very low 2.22 FIP because of his astronomical strikeout rate and because he has yet to allow a homer.

Weighted Component ERA (WERC) is another ERA estimator which adds hits to everything that FIP includes.  Thus, it estimates what a pitcher's ERA should be based on walks, hit batsmen, strikeouts, home runs and hits.  Why might WERC be useful for relief pitchers?  There is growing evidence that relievers (especially very good ones) have more control over the results of batted balls than starters.  The phenomenon has been studied by Lewie Pollis of Wahoo Blues and Dave Studenmund at The Hardball Times among others.  It has been found that at least some relievers tend to allow fewer hits on batted balls throughout their careers than most starters.

One theory is that relievers only have to pitch to a few batters per game and can throw their best stuff on every pitch, whereas starters need to pace themselves throughout the game.  Therefore, relievers may be able to induce weaker contact than starters.  With this in mind, it might make sense to include hits in the evaluation of relievers even if it has been shown that most starters have limited control over batted balls.  Thus, WERC might be a good alternative to FIP in evaluating relievers.


Because he has allowed just 4.6 hits per nine innings to go with his large number of strikeouts and zero homers, Alburquerque's WERC is even lower than his FIP.  In fact, he leads the American League with a 1.66 WERC.  The AL league leaders with 25 or more appearances in 2011 are listed in the table below. 

Table 1: AL Reliever WERC Leaders as of July 17, 2011

Player	Team	G	IP	Base Runs/9 IP	WERC
Al Alburquerque	DET	28	29.2	1.79	1.66
Scott Downs*	LAA	35	31.0	1.81	1.69
David Pauley	SEA	35	48.1	1.94	1.81
Daniel Bard	BOS	44	46.1	2.02	1.88
Koji Uehara	BAL	40	44.0	2.19	2.04
Jesse Crain	CHW	41	41.2	2.32	2.16
Glen Perkins*	MIN	37	35.2	2.38	2.22
Mariano Rivera	NYY	37	35.0	2.43	2.26
Kyle Farnsworth	TBR	41	38.0	2.44	2.26
Joe Smith	CLE	37	34.2	2.47	2.30
Grant Balfour	OAK	37	36.2	2.51	2.34
David Robertson	NYY	39	36.1	2.55	2.37
Vinnie Pestano	CLE	39	35.1	2.70	2.51
Marc Rzepczynski*	TOR	41	37.0	2.72	2.53
Chris Perez	CLE	37	34.1	2.74	2.55
Juan Cruz	TBR	38	34.0	2.74	2.55
Brandon League	SEA	40	37.2	2.75	2.56
Matt Albers	BOS	31	38.0	2.77	2.58
Sergio Santos	CHW	36	42.0	2.82	2.63
Jordan Walden	LAA	40	39.0	2.88	2.68

Other Tiger relievers are listed in Table 2 below.  Set-up man Joaquin Benoit (3.48) and closer Jose Valverde (3.51) trail the Amazing Al by a wide margin.  Now, one thing that WERC does not do is take into account the game situation in which relievers pitch, but that's a topic for another post.  For now, I'll just say that Alburquerque has been close to un-hittable when he has pitched.

Table 2: Tigers Reliever WERC Leaders as of July 17, 2011

Player	G	IP	Base Runs/9 IP	WERC
Al Alburquerque	28	29.2	1.79	1.66
Joaquin Benoit	40	35.0	3.74	3.48
Jose Valverde	44	41.2	3.77	3.51
Daniel Schlereth*	29	25.2	5.57	5.18
Ryan Perry	20	22.1	5.64	5.25

The raw data used to create the statistics in this post was extracted from Baseball-Reference.

Tigers Starters Getting Hit Hard

In previous posts, we saw how well Justin Verlander ranks among American League starters on various statistics - Weighted On Base Average Against (wOBAA), Base Runs Above Average (RAA) and Weighted Component ERA (WERC).  Unfortunately, the rest of the staff has not fared well at all.

Table 1 shows that all Tigers starters not named Verlander have wOBAA's above .330.  How bad is that? Consider that the average wOBAA for AL starters with 50 or more innings is .316. 

Table 1: wOBAA for Tigers Starters (as of July 10)

Player	Team	IP	wOBAA
Justin Verlander	DET	151.0	.242
Phil Coke*	DET	79.1	.331
Rick Porcello	DET	96.0	.342
Brad Penny	DET	110.0	.345
Max Scherzer	DET	111.1	.354

Table 2 illuatrates that all Tigers starters other than Verlander are below average in Base Runs Above Average (RAA).  Max Scherzer has been particularly bad based on this measure with 12 runs below average.  This means that he has cost the Tigers an estimated 12 runs compared to the average starter with 111 innings.  He ranks 67th among 73 starters with 50 or more innings pitched. Rick Porcello and Brad Penny rank almost as low as they are tied for 58th in the league at -6 RAA

Table 2: Base Runs Above Average for Tigers Starters(as of July 10)

Player	Team	IP	Base Runs	RAA
Justin Verlander	DET	151.0	39	33
Phil Coke*	DET	79.1	39	-1
Brad Penny	DET	110.0	58	-6
Rick Porcello	DET	96.0	52	-6
Max Scherzer	DET	111.1	64	-12

Finally, Table 3 tells us that the Tigers #2 through #5 starters are all above 4.00 on WERC. The league average for starters with 50 or more innings pitched is 3.73.  For what it's worth, Charlie Furbush has a 5.05 WERC in 29 innings. 

Table 3: WERC for Tigers Starters (as of July 10)

Player	Team	IP	Base Runs/9 IP	WERC
Justin Verlander	DET	151.0	2.32	2.16
Phil Coke*	DET	79.1	4.41	4.11
Brad Penny	DET	110.0	4.76	4.42
Rick Porcello	DET	96.0	4.85	4.51
Max Scherzer	DET	111.1	5.21	4.85

Are the Tigers starters really this bad?  Scherzer and Porcello are probably not as awful as their aggregate statistics suggest.  A look at Scherzer's game log shows that he has allowed two runs or fewer in 10 of his 19 starts.  The problem is that when he's bad, he's really bad.  He has allowed five or more runs seven times. Porcello's season has been similar - 9 games of 2 or fewer runs and  six of five or more runs.  So, they have at least given the Tigers a good chance to win in more than half their starts.  However, they have been hit hard too often and they need to reduce the beatings in the second half.

The raw data used to create the statistics in this post was extracted from Baseball-Reference.