Defensive Runs Allowed

[jon cavin]

I may have been foolish today but it had to be done. Can someone give me
a formula, if their is one, for defensive runs allowed. the player in
question was Joey Cora of the Seattle Mariners. I said he would lose the
Mariners more games this year with his defense then he would win with his
offense. My friend begged to differ and now we are trying to figure out
a way to prove whether either one of us would be right. Although it's a
meaningless stat, we were going to use GWRBI's for games he won for the
M's. But what should we use for games he loses? In my mind it's no
contest but my friend disagrees. Thank you for your time and help.

jon cavin
--
Dorian Kim | Co-moderator rec.sport.baseball.analysis. Questions, comments,
flames should be sent to base...@cic.net and/or baseball...@cic.net
Archive site: ftp://ftp.etext.org/pub/Sports/Baseball/Analysis/

[Dale J Stephenson]

n914...@cc.wwu.edu (jon cavin) writes:

>I may have been foolish today but it had to be done. Can someone give me
>a formula, if their is one, for defensive runs allowed. the player in
>question was Joey Cora of the Seattle Mariners. I said he would lose the
>Mariners more games this year with his defense then he would win with his
>offense. My friend begged to differ and now we are trying to figure out
>a way to prove whether either one of us would be right. Although it's a
>meaningless stat, we were going to use GWRBI's for games he won for the
>M's. But what should we use for games he loses? In my mind it's no
>contest but my friend disagrees. Thank you for your time and help.

I use this formula:

The terms:
DA -- Defensive Average, or % of balls hit to fielder turned into outs.
NHS -- Hits prevented beyond a league average fielder [based on DA & AOps]
NEB -- Extra Bases prevented beyond a league average fielder
[based on AOps, 2b/AOps, 3b/AOps]
NDP -- Double plays turned beyond a league average fielder [based on DPI%,
DPOps]
FR -- My Fielding Runs. NHS * .72 + NEB * .31 + NDP * .50
(.72 = TBIII's run value for an out (-.30) with the value for a single (.42))
(.31 = TBIII's difference between double (.73) and single (.42), or triple
(1.04) and double (.73))
(.50 = TBIII's value for an out on base (-.50))

By this measure, Cora cost the Mariners about -15 runs last year -- or
roughly a win and a half. Cora would need only 2 GWRBI to win the bet
by your measure -- by my measure, Cora's bat cost the Mariners another
half a game last year (-6 BR/A -- though positional adjustments would
get some of that back.)

--
Dale J. Stephenson |*| (st...@cs.uiuc.edu) |*| Baseball fanatic

Matthew Stephenson for the HOF! b. Aug 20, 1995, 10 lbs 12 oz
Future centerfielder for the World Champion Atlanta Braves

[Vinay Kumar]

jon cavin (n914...@cc.wwu.edu) wrote:

: I may have been foolish today but it had to be done. Can someone give me

: a formula, if their is one, for defensive runs allowed.

There is one, but it's complex. My advice is to use one of the
numbers found on the net (Dale Stephenson's are most commonly used).
You can find them at
ftp://ftp.baseball.org/pub/baseball/stats/1995/index.html).

: Although it's a

: meaningless stat, we were going to use GWRBI's for games he won for the
: M's.

If it's meaningless, why are you assigning meaning to it?

I'd suggest using some more exact measure for offense, such as batting
runs (also made public by Dale Stephenson), Equivalent Runs (Clay
Davenport) or Runs Created/Value (Nelson Lu). All these stats are
available from the same URL above.

: But what should we use for games he loses?

You're implying that he only loses games on defense. A terrible
offensive player loses games on offense, and a great defensive player
wins games in the field, too.

--
Vinay Kumar
vi...@baseball.berkeley.edu http://ocf.berkeley.edu/~vinayk
vin...@ocf.berkeley.edu
cs61...@fox.cs.berkeley.edu