from Amazon
Wednesday, August 05, 2009
The hockey replacement level
Tom Awad lays out all the bits and pieces that goes into his version of replacement level (GVT). Tom has been pretty generous with the community, with sharing all his historical data, and now with letting people to recreate his metrics.
Alot of what he does is somewhat similar to what I do. We disagree fundamentally regarding how to adjust plus/minus. Neither one of us is right or wrong, because we rely on shaky assumptions. Basically, as the sample size gets bigger, my process is better. As the SS is smaller, his process is better. And, FWIW, I use a similar process to his when the SS is small. That is, I use multiple methods based on the size of the sample, because the assumptions have to change. The most notable one is that the larger the sample, then the more players you play with. It’s basically a “strength of schedule” type of limitation. We discussed this a few weeks ago.
Recent comments
Older comments
Page 1 of 344 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date