Stats with Steven: Factoring in Batted Balls

Sept. 23, 2012; Anaheim, CA, USA; Los Angeles Angels starting pitcher Jered Weaver (36) pitches against the Chicago White Sox at Angel Stadium. Mandatory Credit: Jayne Kamin-Oncea-USA TODAY Sports

Around this time last season I wrote up a primer at Runs Batted Out for a new version of Fielding Independent Pitching that I had been tracking. It was called batted-ball fielding-independent-pitching, or bbFIP, and it attempted to judge how a pitcher performed, regardless of defense like the pitching metric fielding-independent pitching (FIP) already does. The large difference is that bbFIP attempts to adjust for the type of contact the pitcher induced, rather than assuming that the pitcher had no control over the batted ball.

For the explanation of how to caluculate bbFIP, I’ll lift straight from my primer:

"The simple explanation is that sabermetric guru Tom Tango figured out unintentional walks, hit by pitches, line drives, strikeouts, and pop-ups had roughly similar run expectancies on either the positive or negative side. This first part of the formula incorporates these outcomes and calls them the “BIGS”, which is [(unintentional walks + hit by pitch + line drives) – (Strikeouts + Pop-ups)]. Tango then takes that total and divides it by the number of batters faced. Let’s use the shorthand formula of:[(UBB + HBP + LD) – (K + PU)] / PA.The next part of the equation is what Tango called the “SMALLS”. Outfield fly balls and ground balls had similar run expectancy so the second part of the equation is (outfield fly balls – ground balls). Of course we divide this by batters faced as well. The final equation for “SMALLS” is:( FB – GB ) / PANow we put it all together as an equation. Doing some fancy math that I won’t bother to get into here, we multiply our “BIGS” equation by 11 and our “SMALLS” equation by 3, which giving us the difference in run expectancy. At the very end of the equation we add a simple constant (C) to get bbFIP onto a scale similar to ERA (which the normal equation for FIP also does. In the end our final calculation is:bbFIP = {11*[(UBB + HBP + LD) – (K + PU) / PA] + 3*[(FB – GB) / PA]} + CbbFIP really allows us to weed out those outliers from FIP that out- or under-perform their peripherals on a consistent basis. The original FIP equation is highly dependent on home runs, strikeouts and walks, and basically assumes that each pitcher should regress towards the mean in terms of BABIP, but BABIP itself still depends on batted-ball type. Ground balls and pop-ups will turn into runs much less often then fly balls or line drives."

So that gives us the formula. It is a fair amount of math, but I feel it helps add another dimension to evaluating pitcher performance. There are some issues with bbFIP, like with any metric–since all of these classifications are done by humans there is something called “stringer bias”, which just states that the people recording these statistics aren’t perfect at classifying batted balls. Another fault is that weak ground balls and weak flyballs aren’t separated from overall groundballs and flyballs. This hurts pitchers such as Jered Weaver and Matt Cain who have an ability to produce lazy flyballs at a higher rate than most pitchers.

Most Mondays, I will update this statistic and post the Angels’ pithers results here. For now, I will leave you with how all of those currently on the 40-man roster. These were their bbFIP numbers (along with the components to give you an illustration of how they were arrived at) from last season.

Player

BB%

GB%

FB%

LD%

PU%

ERA

FIP

bbFIP

Joe Blanton

191

20.2

4.2

44.6

29.7

23.4

2.4

4.71

3.91

3.84

Sean Burnett

56.2

23.8

5.0

57.4

21.6

19.8

1.2

2.38

2.79

2.86

David Carpenter

39.2

18.0

9.3

41.8

31.8

22.7

3.6

4.76

4.93

4.38

Scott Downs

45.2

16.5

8.8

60.4

21.5

18.1

0.0

3.15

3.66

3.89

Barry Enright

3.2

0.0

5.0

26.3

57.9

15.8

0.0

14.73

7.46

7.35

Ernesto Frieri

36.4

11.2

26.3

43.6

21.1

9.0

2.32

3.58

2.80

Steve Geltz

9.1

27.3

28.6

57.1

14.3

0.0

4.50

6.59

7.80

Tommy Hanson

174.2

21.2

9.3

39.8

36.3

20.7

3.2

4.48

4.57

4.23

Kevin Jepsen

44.2

21.3

6.7

35.2

36.1

5.7

3.02

3.21

4.14

Michael Kohn*

12.1

24.4

53.7

7.3

14.6

7.30

10.32

4.34

Ryan Madson*

60.2

25.2

6.5

48.8

29.4

17.5

4.4

2.37

2.25

2.30

Nick Maronde

25.9

11.1

41.2

47.1

11.8

0.0

1.50

2.26

3.55

Brad Mills

33.3

0.0

33.3

41.7

25.0

0.0

0.00

0.69

2.58

Garrett Richards

14.8

10.7

45.5

30.7

21.6

2.2

4.69

4.62

5.15

Andrew Taylor

2.1

0.0

28.6

30.0

10.0

30.0

11.57

8.24

6.46

Jason Vargas

217.1

15.9

6.2

40.2

35.9

19.4

4.6

3.85

4.69

4.46

Jered Weaver

188.2

19.2

6.1

36.0

38.8

21.1

4.0

2.81

3.75

4.40

Jerome Williams

137.2

17.1

6.1

53.6

26.3

18.2

1.8

4.58

4.15

3.90

C.J. Wilson

202.1

20.0

10.5

50.3

27.7

19.9

2.2

3.83

4.04

4.16

*Refers to 2011 statistics for players who missed entire 2012 season

Home/LA Angels News

Stats with Steven: Factoring in Batted Balls — bbFIP and The Angels

By Steven McEwen | Mar 10, 2013