Friday, October 21, 2005

Tiger ERA Analysis

Earlier in the week, I spent some time evaluating Tiger batters using the statistic “runs created”. Now I’ll take a look at the pitchers. One of the better statistics for evaluating overall pitching performance is ERA+. ERA+ is more useful than ERA for comparing pitchers pitching in different environments whether it be leagues, ballparks or years ERA+ is a pitcher’s ERA relative to the league average ERA. It is also adjusted for the impact of the pitcher’s home park. A league average ERA+ is 100. An ERA+ of 105 would indicate that the pitcher had an ERA 5% above the league average. Similarly, an ERA+ of 94 means that the pitcher was 6% below the league average.

A pitcher pitching his home games in a pitcher friendly park will have his ERA+ adjusted downward and a pitcher pitching in a hitter friendly park will have his ERA+ adjusted upward. So if two pitchers have the same ERA but one of them pitches for the Dodgers and the other pitchers for the Rockies, the Rockie pitcher will have a higher ERA+. Ballpark factors are very complex and I’ll probably discuss them further in a future post but for now I’ll just point you toward US Patriot if you want a detailed explanation.

It’s important to note that the league average of 100 includes a whole bunch of pitchers who pitched poorly in a few games. So, I threw out all of the pitchers with fewer than 25 innings pitched. Also keep in mind that reliever ERA+’s are inflated because relievers often give up other people’s runs.

Because of the issues discussed above, I took all of the pitchers in the American League in 2005 with 25 or more innings pitched and divided them into primary starters and primary relievers. The median ERA+ for the starters was 106 and the median ERA+ for the relievers was 120.

The tables below give the ERA+ figures for Detroit Tiger pitchers in 2005. The raw data were extracted from statistical tables at The Hardball Times.


Starters

Pitcher

IP

ERA

ERA+

AL Median

***

***

106

Johnson

210

4.54

102

Maroth

209

4.74

98

Robertson

196.7

4.48

103

Bonderman

189

4.57

102

Douglass

87.3

5.56

83

Ledezma

49.7

7.07

66


Relievers

Pitcher

IP

ERA

ERA+

AL Median

***

***

120

Spurling

70.7

3.44

135

German

59

3.66

127

Walker

48.7

3.7

125

Rodney

44

2.86

162

Farnsworth

42.7

2.32

200

Ginter

35

6.17

75

Dingman

32

3.66

127

Urbina

27.3

2.63

176

Percival

25

5.76

81

Colon

25

6.12

76




The four most frequently used starting pitchers – Jason Johnson (102), Mike Maroth (98), Nate Robertson (103) and Jeremy Bonderman (102) were all slightly below the league median for primary starters. The fifth Starters – Sean Douglass (83) and Wil Ledezma (66) were very far below the league median. Douglass is no longer with the organization and Johnson will be a free agent this off-season. It’s no secret that they will be looking to bolster their pitching staff during the winter.

Most of their most frequently used relievers – Chris Spurling (135), Franklin German (127), Jamie Walker (125), Fernando Rodney (162), Kyle Farnsworth (200) and Craig Dingman (127) were above the league median. Barring trades, all of them except Farnsworth (who was traded during the season) should be returning next year. Of course, ERA+,s, just like ERAs, are kind of tricky for relievers but these figures are still encouraging.

6 comments:

  1. This is the second time this week I was surprised by Chris Spurling. I thought that he was acceptable this season, but I didn't view him as above average. When I was looking at Win Shares a couple weeks ago, he also performed well.

    ReplyDelete
  2. I agree. I didn't get the feeling he was that good either until I started looking at his stats. He didn't allow many base runners though. He actually did have a pretty good season.

    ReplyDelete
  3. is there such a stat as WHIP+? personally, if i'm going to use one stat to evaluate a pitcher, it would be WHIP. having the additional adjustments would be icing on the cake.

    ReplyDelete
  4. Zimm, there is a stat called component's ERA (CERA) which takes all of a pitcher's walks, hits, homeruns,etc and determines what a pitcher's ERA "should be" based on that. It's usually a better predictor of future performance than ERA. When I find the figures for 2005, I'll post them.

    Lee

    ReplyDelete
  5. Lee, CERA sounds intersting. looking forward to seeing it.

    ReplyDelete
  6. ESPN has the component ERA. It's in the ERC column at the address below.

    http://sports.espn.go.com/mlb/stats/pitching?seasonType=2&type=pitch5&sort=ERA&split=0&season=2005&pos=all&hand=a&league=al&ageMin=17&ageMax=51

    The formula is at Cyril Morong's site:

    http://www.uri.edu/personal/jkuh9192/conc.html

    ERC%=ERC/ERA.

    It's not adjusted for ballpark though. I'd still like to find that.

    ReplyDelete

Twitter

Blog Archive

Subscribe

My Sabermetrics Book

My Sabermetrics Book
One of Baseball America's top ten books of 2010

Other Sabermetrics Books

Stat Counter