Saturday, November 19, 2005

Detroit Tigers Fielding Analysis for 2005

A few weeks ago, I used runs created to compare Tiger batters to other batters in the league and ERA+ to compare pitchers. Fielding is more difficult because there are still no fielding stats which measure fielding performance as accurately as hitting and pitching stats measure hitting and pitching performance.


The most commonly reported fielding measure is fielding percentage (FPCT) which is the infrequency with which fielders make errors on balls which they reach. It is calculated as (Total plays – errors)/total plays. Not making errors is a positive thing so this statistic has some value. It has also some important flaws though. First, errors are subjective and judgement varies from one official scorer to the next.


More importantly, FPCT says nothing about range. Some players get to a lot of balls which other players can not reach. Measurement of range is an area of sabermetrics which is still developing. The two most accessible range statistics, Range Factor (RF) and Zone Rating (ZR), are discussed below.



Range Factor was originated by Al Wright in the 1870s but it was virtually ignored for over 100 years before Bill James re-introduced it around 1980. Range Factor = (put outs + assists)/games. One of the limitations of RF is that it is a team dependent statistic. For example, an outfielder playing behind a predominantly groundball staff will have fewer opportunities than an outfielder on a staff with a lot of fly ball pitchers. Similarly, the range factors of all fielders on a team will be affected if their pitchers strike out a lot of batters.



Zone Rating was developed by STATS, Inc in the early 1990s. They divided the baseball field into zones and assigned the zones to different fielders. Zone rating is the percentage of balls hit into a player’s zone which are turned into outs. One of the problems with zone rating is that it depends on the reliability of those collecting the data. They need to judge whether balls were actually hit into the zone and distinguish between fly balls and line drives as there is a different zone for each.



Ultimate Zone Rating (UZR) developed by Mitchel Lichtman (who works part-time for the St. Louis Cardinals and often posts at Baseball Think Factory) and Probabilistic Model for Range (PMR) developed by David Pinto (Baseball Musings) both look at the ability to turn a ball into an out compared to the probability of turning a ball into an out. While ZR treats all balls hit into the zone the same way, UZR and PMR consider some balls within the zone to be more difficult to reach than others. Probabilities are assigned according to the direction of the ball, the batted type (fly ball, pop up, line drive, ground ball) and how hard the ball was hit. There are subtle differences between the two measures which I won’t discuss here.


UZR and PMR are probably the best measures of range of which I’m currently aware but they are not available for 2005 at this time. Other measures of fielding include Fielding Runs developed by John Thorn and Pete Palmer, Fielding Win Shares developed by Bill James and a statistic simply called “range” developed by David Gassko (Baseball Think Factory). None of these statistics use play by play data so they are probably not as accurate as UZR and PMR. Baseball Information Solutions promises an important new fielding statistic in the Fielding Bible, a soon to be published book. I’ll talk more about some of these statistics as they become available.


The chart below displays FPCT, RF and ZR for Detroit Tiger fielders.



Player

Innings

FPCT

RF

ZR

Shelton 1B

738

.993

-

.857

Polanco 2B

946

.995

5.39

.857

Guillen SS

625

.978

4.51

.833

Inge 3B

1400

.957

3.26

.801

Ordonez RF

672

.993

1.93

.869

Monroe RF

632

.971

1.94

.880

Logan CF

874

.979

2.93

.904



The following chart displays the rank for each player among all Major League regulars at that position. I used 600 innings as the cut off.


Player

FPCT Rank

RF
Rank

ZR Rank

Shelton 1B

23

-

23

Polanco 2B

1

5

4

Guillen SS

14

17

19

Inge 3B

14

1

7

Ordonez RF

4

22

17

Monroe RF

22

21

11

Logan CF

26

2

8



The discrepancies between range factor and zone rating, two statistics which are supposed to measure the same thing tell us they are not completely reliable. However, we can probably say that a player ranking high in both measures probably has good range. This means Placido Polanco, Brandon Inge and Nook Logan. FPCT is almost meaningless for outfielders because they make so few errors. The only Tiger regular infielder with good ranks across the board was Polanco.


Range does not really pertain to catchers so I’ll address catching statistics at another time.

No comments:

Post a Comment

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