View Single Post
Old
11-03-2012, 11:02 PM
  #182
Czech Your Math
Registered User
 
Czech Your Math's Avatar
 
Join Date: Jan 2006
Location: bohemia
Country: Czech_ Republic
Posts: 3,717
vCash: 500
Quote:
Originally Posted by Rhiessan71 View Post
It's not about the math at this point, it's about the reality the math is trying to tell us and the difference between 4th line goals in 1985 and today is a hell of a lot more than just 1-2 goals more than what Adjusted Stats is telling us.
The math tells us the relative value of the goals/points in each season. This isn't exactly the same as the difficulty of attaining a certain level of production. There at least three reasons why scoring could comparatively change between tiers:

1. The relative quality of one or more tiers changed in comparison to one or more other tiers.

2. The relative opportunity of one or more tiers changed in comparison to one or more other tiers.

3. The league environment changed in a way that helped one or more tiers proportionately more/less than one or more other tiers.

I have given examples of # 1 & 2:

#1- The much larger influx of overseas players was composed more of forwards in the top tiers than bottom tiers. This may have been a stylistic difference in overseas players, or due to more incentive for the better players to leave for the NHL than the more marginal players, or the preference of NHL teams for certain types of players (e.g. they may have been more confident due to more/better scouting of better players or believed the fulfilled certain roles better than others).

#2- The increase in PP opportunities helps 1st/2nd line players almost exclusively, since they are the players who get these additional opportunities. Of course, it helps players who score a higher % of their points on the PP significantly more than those who score a higher % of their points at ES.

You seem to be focused on #3, which is probably the most difficult type of effect to more directly measure.

Quote:
Originally Posted by Rhiessan71 View Post
Only the top Russians and most of the top Czechs were missing in the 80's and there are 43% more 1rst, 2nd, 3rd and 4th line spots today than there were then.
Just adding the Russians and the Czechs do NOT account for a 43% increase in top line players and not every single Russian or Czech that has come over since is a top tier player.
There was a time in the very late 80's and early 90's where these top Russians and Czechs did affect top Tier scoring like you say. That was only for a short time though, before the League expanded to 43% more teams. Once again spreading the talent out and the data supports this.
I don't have exact numbers for all forwards, but from what I remember the number of total forwards from overseas has been ~25% since the mid-90s, maybe slightly less. The US has been a more consistent and stronger presence since the mid-80s as well. The current NHL has 9 more teams out of 30, so overseas players would need to be ~30% for there to be the same number of Canadian/US forwards in the NHL. Of course there have been changes in population as well, so it doesn't appear that there has been a substantial decrease in the minimum quality of NHL forwards since the 80s. OTOH, the quality of forwards in the top tier has substantially improved, as can be seen by the table below of top 2N scorers (where N = number of teams):

Year Cze Fin Rus Swe USA Other Total
1977 0% 0% 0% 5% 0% 0% 5%
1978 0% 0% 0% 3% 0% 0% 3%
1979 3% 0% 0% 6% 0% 0% 9%
1980 2% 0% 0% 5% 2% 0% 10%
1981 5% 2% 0% 5% 2% 0% 14%
1982 5% 2% 0% 2% 2% 0% 12%
1983 5% 2% 0% 5% 2% 0% 14%
1984 2% 2% 0% 5% 12% 0% 21%
1985 5% 2% 0% 7% 5% 0% 19%
1986 2% 5% 0% 2% 14% 0% 23%
1987 2% 7% 0% 2% 7% 0% 19%
1988 2% 2% 0% 5% 12% 0% 21%
1989 2% 5% 0% 7% 10% 0% 24%
1990 0% 2% 2% 2% 11% 0% 18%
1991 2% 0% 5% 0% 20% 0% 27%
1992 4% 0% 4% 2% 18% 0% 29%
1993 4% 4% 8% 2% 14% 0% 33%
1994 4% 2% 11% 4% 11% 0% 32%
1995 2% 2% 11% 6% 13% 0% 33%
1996 12% 2% 12% 4% 17% 0% 46%
1997 12% 2% 13% 6% 13% 0% 46%
1998 10% 5% 14% 5% 14% 0% 48%
1999 13% 4% 15% 9% 15% 0% 55%
2000 19% 2% 11% 7% 12% 2% 53%
2001 18% 2% 12% 10% 18% 0% 60%
2002 13% 2% 13% 8% 15% 0% 51%
2003 18% 5% 12% 10% 10% 0% 55%
2004 20% 3% 6% 11% 14% 0% 55%
2006 14% 3% 11% 14% 10% 0% 52%
2007 13% 5% 13% 11% 5% 2% 48%
2008 11% 2% 11% 11% 5% 3% 44%
2009 10% 2% 15% 8% 11% 3% 48%
2010 3% 3% 11% 10% 11% 2% 40%
2011 9% 5% 6% 8% 14% 5% 47%
2012 11% 5% 6% 10% 11% 3% 47%

Quote:
Originally Posted by Rhiessan71 View Post
Besides, if what you propose were the reason, there should have been an increase in top tier scoring while other tiers remained the same.
That wasn't the case.
All tiers scored less at variable rates.
The factors I've mentioned don't happen in a vacuum. I'm talking about comparative scoring between tiers, not raw data by tier. Yes, scoring decreased across all tiers due to other factors: more emphasis on defensive systems, the lack of rules enforcement (clutch & grab, etc.), changes in equipment (goalies in particular), etc. You contend that these factors hurt lower scoring players proportionately more than higher scoring players, but that's just a supposition with no real proof. The "proof" you offer is the change in comparative scoring by tiers. Yet I've provided valid alternative explanations, based on substantial changes in the comparative quality of the tiers (overseas players) and comparative opportunity of tiers (PP opportunities). Additionally, I've presented ITT that 6/7 of the top peak/prime adjusted scorers played primarily in higher scoring seasons, which does not seem to support your assertion at all, and in fact supports the opposite. So I would say the burden of proof is on you to prove that the factors which decreased overall scoring also disproportionately decreased the scoring of lower scoring, bottom tier players.

Quote:
Originally Posted by Rhiessan71 View Post
And nothing you have said counters the much more supportable (and quite frankly common sense) explanation that scoring has decreased through a filtering process and not a flat % across the board.
I'm not saying that it decreased equally across the board. In fact the numbers suggest it hasn't. What I'm saying is that unless you determine most/all of the important factors which caused comparative scoring between different types/tiers of players to change, then you cannot confidently decide whether/how to further adjust the "simple adjusted" stats to better reflect the change in difficulty for those players to attain various levels of production (and so more fairly compare their production to that of players from other seasons).

Czech Your Math is offline   Reply With Quote