View Single Post
Old
10-17-2012, 02:25 PM
  #105
Czech Your Math
Registered User
 
Czech Your Math's Avatar
 
Join Date: Jan 2006
Location: bohemia
Country: Czech_ Republic
Posts: 4,846
vCash: 500
Quote:
Originally Posted by Sturminator View Post
I'm just going to have to say no to this. Measuring a defenseman's value by the number of points he scores misses the point, as scoring is not a defenseman's primary role on a hockey team, rare cases notwithstanding. Yes, I think the composite ratings of the hockey writers over a twenty year period are considerably more valuable in measuring the contributions of European defensemen than are their raw points totals. I completely missed the fact that you were looking at point totals when you started your analysis because I just assume that no one would consider points the best measure of the value of defensemen. We obviously look at this problem from very different perspectives. I think yours is badly misguided, at least in the context of this statistical discussion, but I'll just leave it at that.
We'll have to agree to disagree then. I don't pretend points are the best, and certainly not the only way to evaluate defensemen, but they are the best and only way that I consider objective. The biased opinions of a bunch of sportswriters are not what I would term objective, statistical evidence. The "true" number of non-Canadian top tier D-men is very likely between the floor (% non-Canadian d-men in NHL... ~25%?) and the ceiling (% non-Canadian d-men in top X in points amongst d-men... 40-60%?). Unfortunately, this is a very large range. However, using Norris/AS voting doesn't really solve anything IMO.

If the question is "how much did non-Canadian talent affect Norris/AS voting over the past 15-25 years?" then Norris/AS voting would be useful. However, to measure the impact of non-Canadian d-men on the "top tier" (say top 20-60 d-men in the NHL), then such voting is not objective and does not have the depth of information necessary to estimate such impact.

This thread relates to Ovechkin and Lindros, and in the case of forwards, points is a much better indicator of value. It's clear that since Lindros entered the league, non-Canadian talent has made a huge impact amongst top tier forwards. Therefore, since that time, a forward ranking top Y in a category is not directly comparable to a forward ranking top Y before that time. It's more accurate that a forward ranking in the top "Y/2" (or 50% of Y) before the early-mid 90s is roughly comparable to a forward ranking Y since. For instance, an 8th place finish in points in recent years is more comparable to a 4th place finish before the '90s than it would be to another 8th place finish IMO.

Czech Your Math is offline   Reply With Quote