Saturday, December 15, 2007

FIP Analysis for Tiger starters in 2007

In an earlier post, I discussed team run prevention using FIP ERA to measure pitching performance and DER to measure fielding performance. Now, I will evaluate the performance of individual starting pitchers using FIP ERA. There were 64 American League pitchers with 17 starts (approximately a half season) or more in 2007. Table 1 below lists Detroit Tiger starting pitchers (plus Dontrelle Willis) in 2007. Table 2 lists all 64 qualifiers in the league (plus Willis).

In both tables, ERA represents the pitcher’s actual ERA. FIP ERA represents the pitcher’s equivalent ERA based on fielding independent statistics (K,BB,HR,HBP) only. The average ERA and FIP ERA in the American League was 4.50. The DER is the Defensive Efficiency Ratio of the team when the given pitcher was on the mound. A higher DER indicates that he received more fielding support. The league average DER was .70 which means that 70% of balls in play not including homers were turned into outs.

FIP-Actual is the difference between the pitcher’s FIP ERA and actual ERA. It tells how much the actual ERA was helped or hurt by non FIP factors. A negative number indicates that the pitcher probably pitched better than his actual ERA. A positive number says that he probably pitched worse than his actual ERA. LOB% is the percentage of runners put on base which were left stranded. The league average LOB% was .70.

The tables show that Jeremy Bonderman had a FIP-Real of -0.79 indicating that he pitched better than his actual ERA. His DER (.681) was lower than any starting pitcher on the team which means that he may have received less fielding support than his pitching mates.

Since FIP ERA is a better predictor of future performance than actual ERA, Bonderman’s 4.20 FIP might normally bode well for the upcoming year. The bad news is that his FIP ERA has been far better than his actual ERA for 5 straight years indicating that it might be something more than just bad luck. I believe it is caused by a propensity to give up runs in bunches. Last year, most of those bunches seemed to occur in the first inning where he allowed 37% of his runs. This apparent tendency also shows up in his low LOB% (.66).

The most encouraging thing about Bonderman's 2007 season is that he had a 3.44 FIP ERA and 3.48 actual ERA in the first 16 games. At some point in the second half, he got hurt and his ERA ballooned to 7.38 (with a 5.00 FIP ERA) in his final 12 games. The biggest question is how healthy he will be in 2008.

Like last year, Justin Verlander's FIP ERA (4.09) was higher than his actual ERA (3.66). Verlander had strong fielding support (.721) and a high LOB% (.75). LOB% does not always indicate a skill but he has had a high one for two straight years so it will be interesting to see if he keeps it up. In limited action (63 innings), kenny Rogers also had a higher FIP ERA(4.43) than ERA (5.13). The same can be said for three pitchers no longer with the organization: Chad Durbin, Jair Jurrjens, and Mike Maroth. All of these pitchers had ERAs over 5.00 and better than average DERs indicating that they may have somewhat lucky.

Nate Robertson (4.73 FIP ERA, 4.76 ERA) and Dontrelle Willis (5.10, 5.17) had FIP ERAs in line with their real ERAs. This suggests that their real ERA were indicative of how they pitched in 2007. Like Bonderman, Willis may have pitched hurt last year. For more on Willis' 2007 season, check out Bilfers Detroit Tiger Weblog

What can we expect for next year? I'm confident that the still developing Verlander will be one of the top 10 pitchers in the league. Robertson will probably continue to be close to league average or just slightly worse which should be good enough with the Tigers high powered offense. If they are going to get into post-season, I don't think he can't be their #2 pitcher though. I believe the keys to the success of this team will be Bonderman, Willis and Rogers getting healthy and pitching better than league average. Maybe not all three can be expected to bounce back but they will need at least two. At age 43, I'm skeptical about Rogers pitching a full season after his health problems of 2007. I'm more hopeful for Bonderman and Willis but they are defintely question marks.

The raw data for this report were abstracted from The Hardball Times database.

In a later post, I will look at relief pitchers


Table 1: FIP ERAs for Tiger Starters (plus Dontrelle Willis) in 2007


FIP ERA Rank

Name

IP

ERA

FIP ERA

DER

FIP-Actual

LOB %

19

Verlander

201.7

3.66

4.09

.721

0.43

.75

23

Bonderman

174.3

5.01

4.22

.681

-0.79

.66

40

Robertson

177.7

4.76

4.73

.692

-0.03

.71

55

Willis

205.3

5.17

5.10

.682

-0.07

.70

63

Durbin

127.7

4.72

5.73

.731

1.01

.74

N/A

Jurrjens

30.7

4.70

5.40

.785

0.70

.66

N/A

Maroth

78.3

5.06

6.53

.693

1.47

.77

N/A

Miller

64.0

5.63

5.41

.673

-0.22

.71

N/A

Rogers

63.0

4.43

5.13

.722

0.70

.69



Table 2: FIP ERAs for AL Starters (plus Dontrelle Willis) in 2007

FIP ERA Rank

Name

Team

IP

ERA

FIP ERA

DER

FIP-Actual

LOB %

1

Beckett

BOS

200.7

3.27

3.22

.696

-0.05

.75

2

Sabathia

CLE

241.0

3.21

3.27

.689

0.06

.74

3

Bedard

BAL

182.0

3.16

3.33

.717

0.17

.78

4

Escobar

LAA

195.7

3.40

3.50

.701

0.10

.73

5

Kazmir

TB

206.7

3.48

3.58

.667

0.10

.75

6

Blanton

OAK

230.0

3.95

3.59

.701

-0.36

.68

7

Halladay

TOR

225.3

3.71

3.65

.699

-0.06

.70

8

Lackey

LAA

224.0

3.01

3.66

.699

0.65

.76

9

Haren

OAK

222.7

3.07

3.82

.713

0.75

.76

10

McGowan

TOR

169.7

4.08

3.82

.727

-0.26

.68

11

Hernandez

SEA

190.3

3.92

3.83

.667

-0.09

.75

12

Vazquez

CHA

216.7

3.74

3.92

.712

0.18

.75

13

Wang

NYA

199.3

3.70

3.92

.705

0.22

.72

14

Baker

MIN

143.7

4.26

3.94

.677

-0.32

.72

15

Santana

MIN

219.0

3.33

3.96

.729

0.63

.78

16

Pettitte

NYA

215.3

4.05

4.00

.678

-0.05

.71

17

Shields

TB

215.0

3.85

4.00

.718

0.15

.72

18

Carmona

CLE

215.0

3.06

4.05

.720

0.99

.78

19

Verlander

DET

201.7

3.66

4.09

.721

0.43

.75

20

Mussina

NYA

152.0

5.15

4.11

.660

-1.04

.66

21

Meche

KC

216.0

3.67

4.14

.704

0.47

.73

22

Weaver

LAA

161.0

3.91

4.14

.688

0.23

.74

23

Bonderman

DET

174.3

5.01

4.22

.681

-0.79

.66

24

Clemens

NYA

99.0

4.18

4.28

.707

0.10

.68

25

Buehrle

CHA

201.0

3.63

4.33

.713

0.70

.76

26

Schilling

BOS

151.0

3.87

4.33

.704

0.46

.76

27

Matsuzaka

BOS

204.7

4.40

4.35

.701

-0.05

.74

28

Silva

MIN

202.0

4.19

4.35

.701

0.16

.71

29

Sonnanstine

TB

130.7

5.85

4.35

.674

-1.50

.61

30

Westbrook

CLE

152.0

4.32

4.37

.697

0.05

.70

31

Saunders

LAA

107.3

4.44

4.38

.670

-0.06

.73

32

Burnett

TOR

165.7

3.75

4.44

.739

0.69

.76

33

Garland

CHA

208.3

4.23

4.46

.716

0.23

.66

34

Bannister

KC

165.0

3.87

4.52

.739

0.65

.70

35

Guthrie

BAL

175.3

3.70

4.52

.730

0.82

.75

36

Batista

SEA

193.0

4.29

4.61

.690

0.32

.73

37

Millwood

TEX

172.7

5.16

4.65

.660

-0.51

.68

38

Loe

TEX

136.0

5.36

4.67

.679

-0.69

.62

39

Gaudin

OAK

199.3

4.42

4.71

.695

0.29

.72

40

Robertson

DET

177.7

4.76

4.73

.692

-0.03

.71

41

Byrd

CLE

192.3

4.59

4.77

.691

0.18

.71

42

Tavarez

BOS

134.7

5.15

4.79

.699

-0.36

.63

43

Wakefield

BOS

189.0

4.76

4.79

.718

0.03

.68

44

Colon

LAA

99.3

6.34

4.81

.643

-1.53

.63

45

Washburn

SEA

193.7

4.32

4.83

.716

0.51

.71

46

Contreras

CHA

189.0

5.57

4.84

.674

-0.73

.63

47

Perez

KC

137.3

5.57

4.84

.668

-0.73

.67

48

McCarthy

TEX

101.7

4.87

4.88

.700

0.01

.67

49

DiNardo

OAK

131.3

4.11

4.93

.714

0.82

.67

50

Bonser

MIN

173.0

5.10

4.98

.681

-0.12

.70

51

Jackson

TB

161.0

5.76

4.98

.659

-0.78

.66

52

Burres

BAL

121.0

5.95

5.03

.667

-0.92

.68

53

Marcum

TOR

159.0

4.13

5.05

.733

0.92

.77

54

Cabrera

BAL

204.3

5.55

5.06

.701

-0.49

.67

55

Willis

FLA

205.3

5.17

5.10

.682

-0.07

.70

56

Weaver

SEA

146.7

6.20

5.11

.673

-1.09

.64

57

Santana

LAA

150.0

5.76

5.21

.676

-0.55

.67

58

Litsch

TOR

111.0

3.81

5.23

.725

1.42

.74

59

de la Rosa

KC

130.0

5.82

5.27

.675

-0.55

.68

60

Padilla

TEX

120.3

5.76

5.37

.681

-0.39

.64

61

Danks

CHA

139.0

5.50

5.60

.691

0.10

.71

62

Ramirez

SEA

98.0

7.16

5.60

.652

-1.56

.59

63

Durbin

DET

127.7

4.72

5.73

.731

1.01

.74

64

Trachsel

BAL

140.7

4.48

5.73

.725

1.25

.75

65

Tejeda

TEX

95.3

6.61

6.26

.692

-0.35

.64




3 comments:

  1. Well done. I find this stuff very interesting. We all know that "wins" are a misleading statistic, but I think more and more people are starting to come around and can see that ERA can be quite misleading as well.

    ReplyDelete
  2. Foul balls seem like they may play into things. Some people can instigate more foul balls. A non-caught foul would be considered out of play whereas a caught would, I'm guessing, be considered in play. Does this explain some of the reason for "increased defensive support" for some pitchers. I'd imagine Oakland would benefit the most from this.

    Perhaps not though looking at the oakland pitcher numbers.

    Also, I doubt the real formula comes up with such nice coefficients. HR*13? Are these rounded off and if so does that effect the numbers much?

    Also it would be interesting to see how well FIP correlates with slugging percentage + OBP.

    Solid analysis Lee. Once again, enjoy reading your posts.

    ReplyDelete
  3. Adam, ballparks definitely do affect DER. Some of the other more sophisticated defensive stats I'll get to later are less affected by ballparks.

    Last year the correlation coefficient for FIP and OPS was .87 which is pretty high.

    I got the FIPs from the Hardball times site. I also calculated them using the integer coefficients. They came out different so THBT probably went to one or two decimal places. The differences were really small though. The biggest I saw just eyeballing it was about 0.10.

    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