[ Home ] [ Library ] [ Bookstore ] [ Contact ] [ Search ]
Simply, defensive average is the number of balls fielded by a player at a position divided by the number of balls hit to that fielder's zone of responsibility while he's playing that position. It's similar in concept to STATS' Zone Rating, though the zones for DA seem to be quite a bit larger than STATS. The zones cover the entire field, and overlap one another. DA is based on the Baseball Workshop data, which contains location data for batted balls; these locations are recorded to a finer degree of granularity than is used in determining DA. The location of a batted ball is determined by where it is caught or it lands if it's a fly ball, or where it is either fielded or goes through the infield for a ground ball. The Project Scoresheet scoring system, which is used by the Baseball Workshop, includes a diagram of the field, with the field divided into sections. Multiple sections are lumped together to form zones of responsibility. A fielder is charged with an opportunity if a ball is hit into his zone of responsibility, unless the ball is converted into an out by another player. A fielder is credited with an out if he converts the ball into an out. Unlike STATS' Zone Rating, a fielder is only credited with one out when he initiates a double play. Defensive Average does have some shortcomings. One is that there are a fair number of easy fly balls in the outfield that could be caught without difficulty by more than one outfielder. We don't know if there are biases in the outfield data concerning this; that is, we don't know if on some teams, the center fielder just takes all of those, inflating his DA, while on other teams, those are spread more evenly. There's also the issue of the relationship between the pitching staff and the defense. Will a bad pitching staff make a good fielder look bad? How do you tell if the defense is bad because the pitching staff is bad, or if the pitching staff is bad because the defense is good? Additional areas of future research include park effects and age effects.
Legend for DA reports:
Name obvious G games played at that position GOps ground-ball opportunities; the number of ground balls hit to the area of responsibility for that position when this player was playing that position. AGOps An adjustment to the GOps to account for batted balls for which I have no location data. GO ground balls turned into outs by this player. AGDA GO/AGOps FOps fly ball opportunities (anything that's not a ground ball) AFOps adjusted fly ball opportunities FO fly balls converted into outs AFDA FO/AFOps FE Fielding errors TE Throwing errors 1B number of singles through the area of responsibility. (on ground-balls for infielders, fly-balls for outfielders) 2B number of doubles through the area of responsibility. AGO/2B AGOps/2B AFO/2B AFOps/2b 3B number of triples through the area of responsibility. DPOps number of double play opportunities, defined as runner on at least first, less than two out, ground ball to area of responsibility. DPI number of double plays successfully initiated on these opportunities DP% DPI/DPOps
[ Home ] [ Library ] [ Bookstore ] [ Contact ] [ Search ]
Last Updated: Contact webmaster@stathead.com for corrections or problems
Copyright 1997-2001 by Keith Woolner. All included authors retain the copyrights to their original works.