THE BOOK--Playing The Percentages In Baseball

Friday, March 06, 2009

How much is a great fielder worth?  25 runs.

Here’s one way to tell.

Last year in 2008, with Mark Ellis on the field — perhaps the best-fielding 2B of our generation — the A’s allowed 457 runs while recording 3035 outs.  This means they allowed 4.07 runs per game.  When Ellis was not on the field, they allowed 233 runs on 1270 outs, or 4.95 runs per game.  That difference is a whopping 0.89 fewer runs per game, when Ellis is on the field.  In 2008 anyway.

Now, this by itself doesn’t prove anything.  There are many reasons why the A’s could have given up fewer runs when Ellis was on the field, by coincidence.  It depends who is pitching, who the other fielders are, which team is batting, and who his backups were.  And, just some lucky bounces.  This is the reason we would like a large sample size.  The games Ellis played and games he missed just won’t cut it.  By increasing the number of players, we hope that all these other externalities work themselves out in the wash, leaving us with one major input (the great fielder) and the one output (runs allowed).

So, using WOWY (With Or Without You) based on balls in play, I selected the best twenty or so infielders (2B, SS, 3B) since 1993.  It’s mostly the names you know: Everett, Sanchez, Bartlett, Rolen, Reese, Hudson, Inge, etc.  By looking at a large enough number of great infielders, the idea is that all the noise around them will cancel out.  My only additional constraint was that he must have been on the field for at least 1000 outs for a given team-season, and must have been off the field for at least 1000 outs for that same team-season.

I came up with 68 such seasons since 1993.  The total number of games played was 5386 games on the field and 5400 games off the field.  You have to admit that that’s alot of games.  When the star fielders were on the field, their team allowed 4.60 runs per game, and when they weren’t on the field, they allowed 4.83 runs per game.  Per 162 games, this difference comes out to 37 runs.

I repeated this exercise with the outfielders: Erstad, Beltran, Cameron, Endy Chavez, etc.  Ichiro was part of my initial group, but since he barely misses any games, he’s not included.  I have 60 outfielder seasons.  Their teams allowed 4.89 runs with them, and 5.00 without them, for a difference of 19 runs per 162 game season.

And I did the same for firstbasemen: Minky, Derrick Lee, Tex, etc.  I have 96 such seasons.  They allowed 4.89 runs with these great fielding 1B, and 4.99 without, for a difference of 15 runs.

Roughly speaking, that gives us 26 runs (3 parts 37, 3 parts 19, 1 part 15) of fewer runs scored when you have one great fielder on the field, than when you don’t.

Here’s another way to tell.

UZR is shown on Fangraphs.com for seasons 2002-2008.  The best fielder by far is Andruw Jones.  Including his abysmal 2008 season, he was +139.4 runs in 883 games, or an average of… 26 runs per 162 games.

Sweet, I know. 

If I take the top 20 fielders in UZR, minimum 162 games, their average is +18 runs per 162G.  So, we should feel quite confident that a great fielder adds some 20-ish or so more runs than an average fielder, per 162 games. 

I know most readers here see the range for fielding as +/- 20 runs.  That’s pretty much spot on.