A CITIZEN'S BLOG

"Why is a double denied on defense so much less admirable than a double delivered on offense?"

-George F. Will, Men At Work, Page 276.

Good question. The subject of defense is one that I've talked about a lot on this blog because it is considered to be the "Next Big Thing", in the words of ESPN's Peter Gammons, in baseball. It is not a topic that is given its due, as you might imagine from Will's quote above, in part because it is difficult to quantify and comprehend.

There's a lot of talk about defense these days, from people like Baseballmusing's Dave Pinto (who developed Probalistic Model of Range) and Mike Humphries (who wrote about his Defense Regression Analysis for The Hardball Times), who have devised new and exciting theories about defense. Teams like the Oakland A's have chosen to emphasize fielding as a low-cost means of remaining competitive against free-spending adversaries like the Mariners and Angels. The Red Sox's decision to deal Nomar Garciaparra and others for Orlando Cabrera and Doug Mientkiewicz, decisions made to upgrade their notoriously porous defensive alignment, may have been the shot in the arm the team needed to get back into the AL playoff race and win the World Series.

I’ve spoken often in the past about defense. It’s an interesting topic, sort of the “final frontier” of sabremetrics. A lot of the recent work in the field has concentrated there because so much has been done on topics like pitching and batting. Unlike batting and pitching it is difficult to put fielding into numerical categories, so there is a lot of room for debate and the theories and methods used are fairly controversial. E.g., I rely a lot on “Zone Rating” to discuss fielding during the season because it is a better measure of a player’s abilities than Fielding Percentage* (which only measure how a player does with balls he gets to) or Range Factor^ (which revolves around how often your pitchers put the ball into play at you).

*Fielding Percentage: (Putouts + Assists) / (Putouts + Assists + Errors). How often the player successfully handled the ball.
^Range Factor: (Putouts + Assists) * 9 / IP. Essentially measures how much a player is involved in defensive plays.

“Zone Rating” has its flaws – it measures balls a player “should” have gotten to. Subjective to a certain extent, though I like it because it measures a players defensive range more than FP or RF.

This will be a sizable post because there is a lot to talk about when it comes to the Phillies defensive progress. Today I’ll discuss the team generally, and then the individual players tomorrow.

The Team … The Phillies were pretty good defensively in 2004. They were third in the NL in DER*, behind just St. Louis and Los Angeles. Something people don’t understand about the Phillies is that the team has been pretty good defensively over the years. Consistently the team has been one of the top teams in Zone Rating and Fielding Percentage.

(You’ll have to take my word on that: SI.com and ESPN.com have both dropped team defense from their stats pages, which makes rating team defense much, much harder. Fortunately I did print out the 2002-2004 stats last fall, so I have them as a reference.)

* Defense Efficiency Ratio (DER): (Batters Faced – (Hits + Walks + Hit By Pitch + Strikeouts)) / (Batters Faced – (Home Runs, Walks + Hit By Pitch + Strikeouts)) How often fielders convert balls put into play into outs.

I use DER to rate team defense because it is pretty objective and, frankly, it’s all I got.

According to Baseball Reference, the Phillies were fourth in DER in 2001, 2002 & 2003, and they were third in ’04. Here is how the Phillies have been doing this season:

DER:
Milwaukee: .718
Pittsburgh: .717
Chicago: .707
Washington: .705
New York: .700
Philadelphia: .700
St. Louis: .694
San Diego: .694
Houston: .693
Loa Angeles: .692
Atlanta: .692
San Francisco: .690
Florida: .685
Arizona: .670
Cincinnati: .669
Colorado: .665
Nl average: .693

Not so great compared with years past, but not bad. Bottom-line: it is above the NL average with some sub-par players on the field. e.g, big bats like Jim Thome, Bobby Abreu and, to a certain extent, Pat Burrell, are defensive liabilities for the team, yet the team has persevered with good play.

Still, the team doesn’t seem to be playing defense with the vigor it has in seasons past, one of the reasons (along with what I’d refer to as the “power outage”, i.e., Jim Thome’s struggles at the plate) the team got off to a slow start. An indication that the Phillies aren’t playing good defense is that the Phillies’ FIP ERA is basically the same as their actual ERA. What do I mean by that? The Phillies actual ERA (4.64) is just 0.04 higher than their FIP ERA*. If the Phillies were playing good defense the team’s FIP ERA would be much higher than its actual ERA: right now it is actually lower. Last year the Phillies FIP ERA was 0.07 higher than its actual ERA, meaning the team’s fielders covered up for some pitching mistakes.

* FIP – Fielding Independent Pitching: (13*HR+3*BB-2*K / IP) + League Factor Evaluates a pitching by how he would have done with an average defense behind him by keeping track of things that a pitcher can control (walks, strikeouts, home runs allowed) as opposed to things he cannot (hits allowed, runs allowed).

Here is what the Phillies have done in past seasons:

Zone Rating: (Rank, behind leader)
2004: .851 (6th, -.014)
2003: .859 (2nd, -.001)
2002: .863 (4th, -.005)

Not too shabby.

I’d be remiss if I didn’t discuss PMR: PMR took a team’s actual DER and compared it with its projected DER (based on a fairly complex formula predicting outs that I won’t get into). The Phillies did very well, very nearly outperforming the team DER. I admit some skepticism as to PMR's conclusion that the Red Sox and Cubs were outstanding defensive teams (huh?), but it is a good system and I use it, along with Defense Regressional Analysis, to evaluate individual players.

Tomorrow: individuals.