Saturday, January 12, 2008

Ranking the Shortstops - 2007

This is the tenth part of my series on fielding in 2007. The table of contents for the entire series is listed below:

Basic fielding stats
Converting Zone Rating to something useful
Revised Zone Rating
Probabilistic Model of Range
Fielding Bible
Ultimate Zone Rating
Fan Fielding Survey versus range measures
Outfield arms
Ranking the second basemen
Ranking the shortstops
Ranking the third basemen
Ranking the first baseman
Ranking the center fielders
Ranking the right fielders
Ranking the left fielders
What about catchers?

The table below shows the runs saved above average per 150 games for all the major league shortstops with 600 or more innings in 2007. Tigers new shortstop Edgar Renteria's numbers ranged from -9 on Zone Rating to +5 on revised Zone Rating. His Average Range Score (ARS) was -3 which means he saved 3 fewer runs per 150 games than the average shortstop. This is better than old shortstop Carlos Guillen who was 17 runs saved below average. Each system ranked Guillen significantly below average and he finished in the bottom 5 overall. This comes to no surprise to Tigers fans who also gave him very low scores on the Fan Fielding Survey (FFS) .

Any time the subject of objective evaluation of shortstops is brought up, the Yankees Derek Jeter immediately comes to mind. Long heralded as an excellent fielder by some television broadcasters, Jeter generally ranks poorly on modern defensive statistics. This year was no exception as all 5 play by play systems ranked him as the worst or next to worst shortstop in baseball. His average score (a whopping 27 runs saved below average) placed him dead last. Brendan Harris was second to the bottom at -19.

The top shortstop in 2007 was old friend John McDonald who played in 31 games for the 2005 Tigers. Now a Blue Jay, McDonald was outstanding according to all 5 systems and finished with an ARS of +26. Second in the majors and first in the National League was Rockies rookie sensation Troy Tulowitzki at +25.


Table 1: Runs Saved Above Average (per 150 games) for shortstops in 2007

player

team

inn

ZR

RZR

PMR

+/-

UZR

FFS

ARS

McDonald

Tor

799

20

29

21

33

N/A

22

26

Tulowitzki

Col

1,375

15

28

37

26

20

18

25

Vizquel

SF

1,219

27

13

5

16

16

22

16

Pena Jr

KC

1,274

3

22

24

14

10

3

14

Furcal

LA

1,210

10

-1

23

N/A

N/A

15

11

Reyes

NYM

1,431

14

15

-8

9

23

19

11

Bartlett

Min

1,194

7

6

19

15

4

-4

10

Crosby

Oak

814

17

9

-1

N/A

N/A

-0

8

Wilson

Pit

1,142

-5

15

11

9

N/A

14

7

Greene

SD

1,396

11

9

-3

5

N/A

18

6

Theriot

ChC

859

14

3

-2

N/A

N/A

6

5

Gonzalez

Cin

873

5

8

-0

N/A

N/A

1

4

Rollins

Phi

1,441

-0

-3

11

5

N/A

19

3

Izturis

ChC/Pit

656

-3

11

-1

N/A

N/A

-11

3

Cabrera

LAA

1,331

3

-6

4

N/A

6

15

2

Lugo

Bos

1,228

-1

2

4

N/A

N/A

-11

2

Tejada

Bal

1,069

7

5

-10

N/A

N/A

-12

1

Hardy

Mil

1,272

-10

6

-0

6

N/A

1

0

Renteria

Atl

1,019

-9

5

-4

N/A

N/A

-1

-3

Drew

Ari

1,283

-8

-11

-1

N/A

N/A

7

-7

Peralta

Cle

1,348

-12

-15

7

N/A

N/A

-24

-7

Uribe

CWS

1,305

1

-16

-9

N/A

N/A

-6

-8

Betancourt

Sea

1,302

-8

-14

-3

N/A

N/A

5

-9

Eckstein

StL

944

-5

-8

-9

-15

-10

-15

-9

Young

Tex

1,291

-10

-15

-22

-12

-15

-7

-15

Guillen

Det

1,074

-7

-18

-18

N/A

-24

-22

-17

Ramirez

Fla

1,302

-14

-22

-2

-29

-20

-17

-18

Lopez

Was

927

-14

-25

-20

N/A

-14

-20

-18

Harris

TB

752

-25

-1

-26

-25

N/A

-25

-19

Jeter

NYY

1,318

-22

-28

-31

-26

-27

-9

-27

Friday, January 11, 2008

Ranking the second basemen - 2007

This is the ninth part of my series on fielding in 2007. The table of contents for the entire series is listed below:

Basic fielding stats
Converting Zone Rating to something useful
Revised Zone Rating
Probabilistic Model of Range
Fielding Bible
Ultimate Zone Rating
Fan Fielding Survey versus range measures
Outfield arms
Ranking the second basemen
Ranking the shortstops
Ranking the third basemen
Ranking the first baseman
Ranking the center fielders
Ranking the right fielders
Ranking the left fielders
What about catchers?

Billfer and Matt are probably the only ones still reading this series so I'll get to the positional rankings before I lose them as well. As I said in one of my comments, range metrics are more repeatable than some people realize. That is, if a player does well on one of these metrics one year, he is likely to do well again in the following year. For those who like correlation coefficients, the correlation between consecutive years for PMR, ZR, and several common hitting and pitching stats are shown below:

ZR .75
SLG .67
OBP .67
PMR .55
FIP .55
BA .43
ERA .33

So, for the last 2 years anyway, ZR is more repeatable than such stable stats as slugging and OBP. And PMR is more stable than ERA and batting average. Repeatability is an important measure of reliability which indicates these metrics are measuring a real skill.

One problem we have encountered throughout this exercise though is that the metrics do not agree for some players. Thus, I am going to take the average runs Saved Above Average per 15 games (RSAA/150) across the five metrics (ZR, RZR, UZR, FB+/-, UZR). This should help alleviate the effect of one really high or low score on one of the metrics for a particular player. Each player will have from 3 to 5 scores included in their average depending on the availability of data on the Fielding Bible +/- system and UZR.

I am not going to include the Fan Fielding Result in the average because it is measuring something different from the other metrics - skill rather than performance. It also has the potential problem of being too subjective. I will, however include it in the tables for comparison.

The results for second basemen are shown in Table 1 below. You can see that Placido Polanco was 4 runs above average on ZR, +12 on RZR, 9 on PMR, 8 and FB +/- . there were no available UZR data for Polanco. Thus, he has 4 scores and the Average Range Score (ARS) is 8 runs saved above average. his Fan Fielding score of +17 is not included in the average. Polanco ranked 8th in the majors on ARS. So, even though he had a perfect 1.000 fielding average, there were some second baseman who saved more runs for their teams.

The major league leaders were Chase Utley of the Phillies (+20) and Mark Ellis of the Athletics.
The trailers were Craig Biggio of the Astros (-19) and Rickie Weeks of the Brewers (-18).


Table 1: Runs Saved Above Average (per 150 games) for second basemen in 2007

player

team

inn

ZR

RZR

PMR

+/-

UZR

FFS

ARS

Utley

Phi

1,167

20

23

20

19

16

10

20

Ellis

Oak

1,322

25

14

15

14

25

24

19

Phillips

Cin

1,371

-2

14

34

8

N/A

19

13

Hill

Tor

1,410

11

12

13

16

14

13

13

Hudson

Ari

1,183

-1

7

19

17

15

22

12

Cano

NYY

1,409

10

4

16

12

15

-6

11

Matsui

Col

863

18

1

13

14

N/A

7

11

Polanco

Det

1,209

4

12

9

8

N/A

17

8

Kendrick

LAA

751

5

9

7

9

N/A

-6

7

Kinsler

Tex

1,137

1

14

18

6

-9

-14

6

Lopez

Sea

1,231

7

2

9

N/A

N/A

-1

6

Grudzielanek

KC

947

-2

7

4

7

N/A

6

4

Iguchi

CWS/Phi

1,043

4

0

4

N/A

N/A

-4

3

Belliard

Was

1,005

-5

-4

15

N/A

N/A

-9

2

Roberts

Bal

1,330

1

4

-0

N/A

-4

18

0

Pedroia

Bos

1,141

6

-6

-0

N/A

N/A

4

-0

Kennedy

StL

630

4

4

-10

N/A

N/A

-10

-0

DeRosa

ChC

709

12

3

-16

N/A

N/A

-0

-1

Castillo

Min/NYM

1,158

-7

-3

4

N/A

N/A

11

-2

Barfield

Cle

1,032

-4

-11

14

N/A

-7

5

-2

Giles

SD

948

4

-2

-20

N/A

6

2

-3

Sanchez

Pit

1,273

-9

-10

-8

N/A

N/A

-3

-9

Johnson

Atl

1,153

-12

-13

-10

N/A

-13

-2

-12

Kent

LA

1,084

-8

-6

-16

-12

-23

-20

-13

Durham

SF

1,028

-14

-13

-23

-10

N/A

-23

-15

Uggla

Fla

1,384

-22

-3

-21

-14

-17

-15

-15

Weeks

Mil

984

-28

-5

-20

-17

N/A

-19

-18

Biggio

Hou

937

-18

-12

-27

-18

N/A

-27

-19

Wednesday, January 09, 2008

Outfield Arms - 2007

This is the eighth part of my series on fielding in 2007. The table of contents for the entire series is listed below:

Basic fielding stats
Converting Zone Rating to something useful
Revised Zone Rating
Probabilistic Model of Range
Fielding Bible
Ultimate Zone Rating
Fan Fielding Survey versus range measures
Outfield arms
Ranking the second basemen
Ranking the shortstops
Ranking the third basemen
Ranking the first baseman
Ranking the center fielders
Ranking the right fielders
Ranking the left fielders
What about catchers?

Earlier this week, John Walsh - a writer for The Hardball Times - wrote his annual report on outfielder arms. With the exception of the Fan Fielding Survey, all the measures that I have discussed in this series have focused on an outfielder's ability to catch fly balls. That is the most important part of outfield defense but the ability to stop the running game by throwing runners out or preventing them to advance is also important.

Using the the retrosheet database, Walsh considered any situations where a runner had a chance to take an extra base on a ball hit to the outfield (e.g. single with runner on first and second base unoccupied). For each outfielder, he counted the number of advancement opportunities, the number of times he threw a runner out (kill) and the number of times he prevented a runner from advancing (hold). From this, he determined how the outfielder compared to league average in each category. Then, using the run expectancy matrix (e.g how likely is it for a run to score with a man on first and third and no outs compared to a runner on first and one out), he then calculated runs saved above/below average for each outfielder.

The results for the Tigers in 2007 are on Table 1 below. Using Curtis Granderson as an example, the table reads as follows: There were 222 opportunities for runners to take an extra base on Granderson's arm in 2007. His kill+ ratio (99) says that he was 1% below the average center fielder in throwing out base runners. His hold+ ratio (100) says he was exactly average. Walsh estimated that Granderson prevented 1.1 runs above average per 200 opportunities (or approximately a full season) with his arm.

I then came up with the Range Runs column by taking the average of the five RSAA/150 estimates discussed earlier in the series. In Granderson's case, he prevented 19.8 runs above average with his range. The final column is the sum of the Arm Runs/200 opps (which is roughly equivalent to 150 games) and Range Runs/150 game columns. We can see that all three Tigers outfielders finished above average in 2007.

New Tigers outfielder Jacque Jones finished off the charts at 35.9 runs above average thanks to outstanding numbers on both arm and range. He had a good year defensively with well above average range scores on all range metrics and 8 assists in 76 opportunities. However, I think magnitude of the final result is partially the result of a relatively small sample size (645 innings).

Table 1: Runs Saved Above Average (per 150 games) for Tiger Outfielders in 2007.

Pos

player

Opps

Kill+

Hold+

Arm Runs/150 games

Range Runs/200 opp

Total Runs/150 games

LF

Monroe

103

154

106

4.3

1.3

5.6

CF

Jones

76

321

99

13.7

22.2

35.9

CF

Granderson

222

99

100

1.1

20.0

21.1

RF

Ordonez

142

58

108

-1.8

3.9

2.1

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