[URL='https://hfboards.mandatory.com/members/quoipourquoi.80345/']quoipourquoi[/URL] said:
I don’t think reporting an 8th place ranking from 2 votes is necessary, but I also don’t think reporting as few as 2 players in some seasons and as many as 12 players in others is the answer either.
Sure, Patrick Roy had a pretty solid 1996-97 (Vezina nominee on the President’s Trophy winner), but is it more notable than his 1988-89 and 1989-90? Of course not. And the actual Hart ranking doesn’t say that it is either - just the subsequent disqualification of the two seasons because of the acknowledged limitations of the ballot used in those years.
I mean, if we’re adding a condition that would disqualify a 3rd place Hart finish in 1980-81 where a player scored 135 points just because there have been random votes that resulted in peculiar placements 8, 9, or 10 names down the ranking in other years, then the altered publication of Hart rankings is probably doing more damage than alleviating the theorized problem.
At the very least, I’d say report a minimum number of names each year (I think top-5 was the standard used throughout the positional projects), and then apply the minimum - because you’re probably correct in picking up relevant allocation of Hart shares beyond those top rankings.
And then if someone wants to argue that Patrick Roy wasn’t a top-5 player in 1988-89 because he only stole so many 3rd place votes from Gretzky and Lemieux and Yzerman, we can cross that bridge. Otherwise I think reporting the Hart placement of five goaltenders in 2003-04 alone (but not also 1990-91 Ray Bourque) might be a stretch.