| RC/25 Introduction (5/20) [message #1039161] |
Sat, 20 May 2006 20:05 |
|
The following posts list major league teams' regular position players' RC/25 by
position. Here are a bit of definitions:
RC/25:
A stat created by Bill James to measure how many runs a lineup of 9
of the same individual would score in a game. (The "real" version is RC/27,
since there are 27 outs in a game; however, since certain stats are not readily
available to us during the season, James has an abbreviated version known as
RC/25 to adjust for the absence of those stats. For the formula to RC/27,
please e-mail me. It is also available in Total Baseball and many other
reference manuals.)
A = H + BB - CS
B = 1.125 * 1B + 1.69 * 2B + 3.02 * 3B + 3.73 * HR + .29 * BB
+ .492 * SB - .04 * SO
(Formerly B = TB + .64 * SB + .24 * BB - .03 * K)
C = AB + BB
RC = ((A + (2.4 * C)) * (B + (3 * C)) / (9 * C)) - (.9 * C)
O = AB - H + CS
RC/25 = RC / O * 25
There are three additional Bill James adjustments that I do not use:
1. He adjusts for BA with runners on base. Originally, the adjustment was
not published; but it has been, in the most recent STATS Major League
Handbook and other sources. I am, however, not incorporating this
change since I have found this change to be arbitrary and also
increasing team-dependence.
2. He also adds additional points for a non-bases empty HR. Again, this
is highly arbitrary -- why not add points for a non-bases loaded
1B, 2B, or 3B? Further, this increases team-dependence as well, which
reduces the usefulness of RC.
3. He then adds up all players' RC on a team and adjusts it in a factor
that is based on whether the team scored more than it is expected or
less. I find that this defeats the point of RC, so I do not use this
adjustment.
PRC/25:
Park-adjusted RC/25. Park adjustment is a topic that I don't know how
to explain well in a paragraph, so I will not attempt to try. However, many
works that explain it are available on the net, and any search engine should
yield a number of useful articles. For these numbers, only this year's park
factors are used.
Value:
An estimate of how many runs a player is above the ever so elusive
"replacement level" (that is, a level of performance at which a player can
easily be found at little or no cost to a team). For these posts, replacement
level is estimated at 50 points of OBP and 50 points of SLG below the average
for each position.
Notes:
1. For these posts, the players who were traded during the middle of the
season have their park factors prorated by AB + BB with the teams.
(This is not mathematically correct, but is the best I can do for now.)
2. Please send to me corrections if I misplaced someone's position.
"SABREmetrics isn't about statistics, it is about the search for new
evidence." - Bill James (thanks, Bill Reich)
============================================================ ===================
GO (LOS ANGELES) ANGELS OF ANAHEIM!
============================================================ ===================
Nelson Lu (nlu [at] cs.stanford.edu)
|
|
|
