Can someone explain to me how to adjust numbers by year played?
Michael Howes
What stats I know I taught myself.....so I don't know if the following is
legit.
If I want to compare a player from the 70s to a player from the 80s will the
following work.
Calculate league average RC/27 for the year(s) in question
Calculate each players RC/27
Calculate the % difference in the league average RC/27 between the years
Apply that % to the players RC/27
is this accurate?
Or do you have to do the league average on a lower level, for example
calculate the % difference in H, BB, 2B, 3B, HR...etc...then apply that to
the players raw numbers and then do the RC/27 calculation.
next...if I have old data from a year....all the player data. Can someone
explain to me how to calculate stadium adjustments? And can they be applied
like the year adjustments above?
thanks for the stats lesson
mike
Clifford Blau
"Michael Howes" <mho...@xbaptismofsolitude.org> wrote:
>If I want to compare a player from the 70s to a player from the 80s will the
>following work.
>
>Calculate league average RC/27 for the year(s) in question
>Calculate each players RC/27
>Calculate the % difference in the league average RC/27 between the years
>Apply that % to the players RC/27
>
> is this accurate?
Seems okay to me. The usual method is to compare the player to the
league average, but your method should give the same results.
>Or do you have to do the league average on a lower level, for example
>calculate the % difference in H, BB, 2B, 3B, HR...etc...then apply that to
>the players raw numbers and then do the RC/27 calculation.
Not necessary. Doing the adjustments on the RC/27 data is perfectly
fine.
>next...if I have old data from a year....all the player data. Can someone
>explain to me how to calculate stadium adjustments? And can they be applied
>like the year adjustments above?
Roughly, calculate the difference between the runs per inning for both
teams in each team's home and road game; use that ratio to adjust the
player stats. Using Total Baseball's figures is easier than
calcuating it yourself.
------------
Clifford Blau
http://users.erols.com/brak
David J. Grabiner
"Michael Howes" <mho...@xbaptismofsolitude.org> writes:
> What stats I know I taught myself.....so I don't know if the following is
> legit.
>
> If I want to compare a player from the 70s to a player from the 80s will the
> following work.
>
> Calculate league average RC/27 for the year(s) in question
> Calculate each players RC/27
> Calculate the % difference in the league average RC/27 between the years
> Apply that % to the players RC/27
>
> is this accurate?
This is the correct method. RC/27 is a measure of value, and you want
to estimate a player's value in a different context. An average player
in 1975 and an average player in 1985 did just as much to help their
teams win.
> Or do you have to do the league average on a lower level, for example
> calculate the % difference in H, BB, 2B, 3B, HR...etc...then apply that to
> the players raw numbers and then do the RC/27 calculation.
You might want to do this if you wanted to estimate how a player woul
ddo under different condition.
> next...if I have old data from a year....all the player data. Can someone
> explain to me how to calculate stadium adjustments? And can they be applied
> like the year adjustments above?
To compute a park effect, let P be the ratio between runs scored in the
park and runs scored in the team's road games (thus by the same
players), and N the number of teams in the league. The park factor is
N(1+P)/(2N-2+2P); it is slightly less than half as strong as the park
effect because you are comparing a team playing half its games in the
park to a league playing 1/N of its games in the park. For example, if
there are 640 runs scored in Dodger Stadium and 800 in Dodgers road
games in a 16-team league, the park factor is 16(1.8)/(32-2+1.6)=0.911.
To use a park factor, divide any run-dominated figure by the park
factor. Thus, in the above example, a 3.64 ERA in Dodger Stadium
corresponds to 4.0 in a neutral park, and 91 RC in Dodger Stadium
corresponder to 100 in a neutral park. To adjust OBP or SLG, use the
square root of the park factor; if a team's OBP and SLG are both
mutliplied by X, then its OBP*SLG, which is proportional to runs, will
be mutliplied by X^2.