Adjusted stats - how valuable?
View Single Post
11-05-2012, 01:20 PM
Czech Your Math
Join Date: Jan 2006
Originally Posted by
Overall you're prolly right but again, we deal mostly with outliers and the top 1% here so even a small number will exponentially change the overall picture for these players.
For example, based on the numbers Overpass presented here
Top tier players have a 13.4% increased share of total goals from 1983 compared to 2008.
You take Gretzky's 159 adjusted points from 1983 and that 13.4% increase changes his share so that he would have a projected 180 points in 2008 or a 9.3% increase projecting from 1983 to 2012 equaling 174 point projection.
Suddenly not so insignificant
Obviously this is just a rough example but even at about half that 13.4% at a 7% increase share, it changes that 159 to 170.
First, Overpass' numbers are using the top Y players on each team, not the top (N*Y) players in the league (where N = # of teams). That means as parity between teams increases, the numbers would likely show an increase (since there should be much less parity between top players than 4th liners).
Second, it again depends on the cause, because if the top lines became of higher quality in a manner disproportionate to lower lines, then the numbers would again show an increase, but it would not mean that it was actually any easier for top line players to score adjusted points (in absolute terms). Reflect on the curved-grade classroom with added students as an example.
Czech Your Math
View Public Profile
Czech Your Math's albums
Find More Posts by Czech Your Math