View Single Post
Old
04-06-2013, 02:14 PM
  #93
Hawkey Town 18
Moderator
 
Hawkey Town 18's Avatar
 
Join Date: Jun 2009
Location: Chicago, IL
Posts: 4,058
vCash: 500
Quote:
Originally Posted by Sturminator View Post
Coffey was 6th in NHL scoring in the 1994-95 season, playing in Detroit. Did he have the talent to finish second in league scoring on his own merits? Certainly.



Jarri Kurri was 8th in league scoring in 1988-89 with Gretzky playing in LA. He was a very talented scorer, with or without the Great One, whose ass he spent a lot of time covering.



Obviously we have to treat Cashman and Hodge's production here carefully, and everybody knows that. Those two were never outlier scorers based on talent. But the simple fact of the matter is that those points were scored, and we cannot arbitrarily act as though they do not exist.



Because he scored those points. The whole purpose of the option #3, or averaging system, is to find some sort of middle point between the outliers and the pack and use that as a benchmark on the assumption that this will create the least distortions (not on the assumption that it is perfect).

Also, we cannot simply remove Bobby Orr because he is Bobby Orr. The VsX system treats all players and all seasons equally, and identifies and removes outliers based on a clear, mathematical system. Arbitrarily removing players because of their names or because they are "70's Bruins" is no system, at all, and leads to nothing but endless, and fruitless, disagreement.
I appreciate the response, but I don't understand why you are using things players did in other seasons to make some kind of prediction on whether they are worthy enough to be considered the #2 scorer in a season where we already know how they performed?

I would like answers to these questions...

- Do you think Coffey still outscores Goulet by 5 points in 1984 with a high-end #1 center instead of an offensive freak like Gretzky as a teammate?
- If the answer is yes, then we just disagree, and can move on. If the answer is no, why wouldn't you use Goulet's point total as the benchmark?
- Exact same two questions in regards to Jari Kurri and Dale Hawerchuk in 1985.
- Does Bobby Clarke's offensive performance in 1974 equate to a vs.X score of 82 in most other seasons?
- If the answer is yes, then again, we disagree, and can move on. If the answer is no, what do you think is the best way to treat this season when doing player comparisons?


Here are a few more comments that are not so easily addressed with simple questions like the above...

In regards to the bolded, I agree that we cannot arbitrarily remove players just because of their team. They should be decided on a case-by-case basis. I actually think that the closer an outlier teammate is to the next guy, the easier it is to dismiss him, like in the 1984 and 1985 examples I gave. The more difficult cases are when they have a big lead, like Hodge's 18pt lead over Clarke in 1974. How do we know if all 18 of Hodge's points can be attributed to Orr and Espo? If you want a structured system, perhaps a good compromise would be to average the outlier teammate and best non-outlier teammate scores, and use that as the benchmark? For example, the 1984 score would be123.5, and 132.5 for 1985, and 95.5 for 1974 (this excludes Orr).

As far as Bobby Orr goes, he is such an anomaly that he really has to be treated differently than everyone else. Never has any other defenseman led the NHL is scoring. Never has a defenseman finished 2nd in scoring with such large gaps to 3rd place. A defenseman outscoring all or almost all of the league by that much is an outlier of some kind, and I think it's something that should be accounted for when doing player comparisons. The reason for this can be seen by looking at Bobby Clarke's score of 82 in 1974, which to me clearly undervalues his season when comparing him to scores of 82 in most other seasons.


Last edited by Hawkey Town 18: 04-06-2013 at 02:21 PM.
Hawkey Town 18 is offline   Reply With Quote