Beltre, Reese key to better M's defense
Defense, the Mariners' strong suit in their glory days, was middle of the American League pack last year, and the pitching staff's effectiveness...
Special to The Seattle Times
Defense, the Mariners' strong suit in their glory days, was middle of the American League pack last year, and the pitching staff's effectiveness suffered.
This year's squad, bolstered by the acquisitions of Adrian Beltre and Pokey Reese and the shift of Randy Winn back to left field, might return the Mariners to defensive leadership.
By the new Sabermetric measures of defense, Beltre last year was the most effective third baseman in the major leagues. By one of the new analysis systems, Reese got to more balls a shortstop might handle than anyone else in the majors. A new measure of defense indicates that the infielders' historical skills would have saved Mariners pitchers 115 hits last season. Using an estimated-runs method, this translates to saving about 55 runs, which projects to a difference of five or six more wins last season.
There are factors that could undermine this improvement, or not. Bret Boone's defense as measured by the numbers has deteriorated every year since 2001, and significantly. But numbers can't tell you if Boone's extraordinary 2004 nosedive was a loss of physical skills or just poor concentration. If it was focus, not physical decay, that cut into his effectiveness, he might partially spring back.
If Reese's injury keeps him out a long time, the team will be counting on Wilson Valdez, whose record is unclear. His numbers are very good, but over so few innings it's impossible to know just how effective he will be.
How do we judge ability based on defensive numbers?
The official defense stat, fielding percentage, made great sense in 1873. Fielders wore no gloves, and one out of every six balls put into play was muffed. Today, however, only one in 51 infield plays results in an error, and the difference in the number of errors per month between the most and least error-prone at a position is tiny.
The idea of measuring fielding prowess by the exception (the 2 percent of balls that turn into errors) while completely ignoring what defenders do on the other 98 percent is absurd.
Sabermetrics (the study of baseball numbers) has used several measures to improve fielding analysis.
Range Factor (RF) measures putouts plus assists per game.
The actual difference between a second baseman who's a little below average and one who's the best is not the difference in fielding percentage — a half-dozen errors a season — it's about RF, which differs by a play per game or about 155 plays a season for a regular.
A newer measure is Zone Factor (ZF). It's not perfect, but it's a useful indicator of prowess.
Invented in 1986 by researcher John Dewan, ZF's foundation is a chart that divides the field into zones for each fielder except the pitcher and catcher. A scorer charts into which zone each hit ball goes and whether the fielder turned it into an out.
Zone Factor adjusts for opportunity by delivering the percentage of balls in the fielder's zone he got to and dealt with properly (putouts plus assists, divided by balls hit into a player's zone). The stat is designed to equalize for ballpark and pitching staffs that deliver varying numbers of grounders or fly balls or strikeouts, all of which affect each position's number of chances.
But ZF has limitations. The results are based on human judgment (the zone lines don't appear on the field, so some balls that land near the seams can be judged differently by different chart-fillers). And how does the tracker chart one of those plays where both Ichiro and Jeremy Reed get to the ball in Reed's zone and Reed backs off to let Ichiro take it?
Baseball researchers are developing more sophisticated systems now based on more elaborate math that remove more "noise" from the numbers and deliver more accurate results. Two of these systems are Bill James' Defensive Efficiency Record (for teams) and Mitchel Lichtman's Universal Zone Ratings (for individuals). The logic and math for them is ornate, but some major-league teams' front offices now apply them to refine their own analysis.
Of all the areas of baseball statistical research, the one with the most undiscovered truths is defense. That makes it a most fascinating pastime — second only to watching outstanding defenders take the wind out of opponents' hitting.
|Out of the zone|
|Bret Boone, a four-time Gold Glove winner, had the worst Zone Factor rating among regular American League second basemen last season. His ZF rating has gone down each of the past three seasons.|
|Mariners' fielding factors|
|Range factor is putouts plus assists per nine innings — basically, how many outs per game a player makes. Zone factor adds putouts and assists, then divides by the total number of balls hit into a player's zone. Below is a look at how present and former Mariners have fared.|
|Pos.||Year||Player||League avg. RF||Player
|League avg. ZF||Player ZF|
|*Small, therefore questionable, sample. **Best rating in league.|
Jeff Angus will write a weekly feature on the new baseball statistics, dissecting how they work, describing what they reveal about the game and how that affects the Mariners. He is a management consultant and author of the Management by Baseball Web log at cmdr-scott.blogspot.com and the book of the same name. He can be reached at firstname.lastname@example.org.