It just doesn't seem that way at all...even if we just keep it in the field of goalies...how much more impressive is Hasek's 1994 to 2001 (6x First-Team, 2 MVPs, 0 Cups) or Roy's 1986 to 1993 (3x First-Team, 2x Second-Team, 2 Smythes, 2 Cups) as compared to:
Plante's 1956 to 1962 (3x First-Team, 3x Second-Team, 1 MVP, 5 Cups)
Sawchuk's 1951 to 1955 (3x First-Team, 2x Second-Team, 3 Cups)
Brodeur's 2003 to 2008 (3x First-Team, 2x Second-Team, should have had a Smythe, 1 Cup)
Dryden's 1976 to 1979 (4x First-Team, 4 Cups)
Hall's well, he's basically wall to wall, let's go 1957 to 1966 (6x First-Team, 2x Second-Team, 1 Cup)
I want to make some general comments about the changes in the competitive environment and the value of awards for post-expansion goalies, because I think it is important both for explaining why Hasek and Roy are in fact dominant relative to earlier goalies and the relative value of All-Star finishes for goalies compared to skaters.
Here are the 10 goalies with the most top-5 save percentage finishes ever (using the
Hockey-Reference definition):
1. Glenn Hall, 12
2. Jacques Plante, 11
3. Gump Worsley, 10
3. Johnny Bower, 10
5. Patrick Roy, 9
5. Dominik Hasek, 9
7. Tony Esposito, 8
7. Billy Smith, 8
9. Ken Dryden, 6
9. Bernie Parent, 6
The top four guys all played in the Original Six, the bottom four dominated the '70s/early '80s, and in between are the only two that played a game after 1990 (also the only two eligible for voting this round). It's pretty clear that there is a strong era bias here that favours guys that played in earlier eras compared to everybody who played in the 1990s or later.
So why was it easier to rack up more elite save percentage finishes in some eras than others? League size had something to do with it, for sure, given that it was clearly easier for a top goalie to finish in the top 5 at a time when there were only six starting goalies all playing the vast majority of their team's games (although obviously you had to be very, very good to even play in the first place). Yet league size alone doesn't explain all the 1970s guys on the list, or why the likes of Chico Resch and Pete Peeters have as many Top-5 save percentage finishes as Martin Brodeur or Roberto Luongo.
1. Goalie Competition
In 1993-94, the league expanded to 26 teams, and ex-Communist bloc goalie talent really started to make an impact in the NHL (Irbe and Hasek had their first seasons as starters, Trefilov and Shtalenkov were rookies, Khabibulin made his IHL debut, etc.). Also, by the early '90s the U.S. was regularly developing top-level goalies, something it had pretty much failed to do entirely between Frank Brimsek and Tom Barrasso.
In the '60s and '70s, every top-5 save percentage finisher was Canadian. In the '80s, I count only four non-Canadians, all making the list once each (Barrasso, Lindbergh, Casey, Takko). Then from 1991-2000, 21 of 50 top-5 save percentage seasons were recorded by a goalie who represented a country other than Canada internationally. Since then it has only continued to grow (in the last 10 years, 34 of 50 spots have gone to non-Canadian goalies). That strongly implies that the goalie talent pool has grown much deeper, even taking into account expansion to 30 (now 31) teams.
This is also supported by the fact that the league average save percentage has increased from .881 in 1990 to the .912-.914 range it has been in for most of the past decade, despite restrictions on goalie equipment, despite improved stick technology, despite rule changes to increase offence, despite goons getting driven out by depth skill players and all the other things that should have made it easier, not more difficult to score. Most of that save percentage improvement took place between 1991 and 1997, the same period of significant goalie talent influx from outside of Canada (and that's not even taking into account the increased depth of Canadian goaltending, led by the new breed from Quebec).
2. Team Effects
The other thing that made it easier for some goalies to get recognition was that goalie success and awards voting used to be more strongly linked to team success.
I have a stat called win threshold that is an estimate of team strength based on a team's average goals scored and shots against per game (Win Threshold = (SA - GF) / SA)). The lower the win threshold, the more likely a team is to win the game, since they don't need as strong of a performance from their goalie. Here is the correlation coefficient between a team's win threshold and their team save percentage in recent eras:
The Original Six (1955-1967): -0.545
The Expansion Era (1968-1979): -0.394
The Open Eighties (1980-1990): -0.306
The Talent Influx (1991-1997): -0.133
The Dead Puck Era (1998-2004): -0.081
Post-Lockout Era (2006-2018): -0.047
A negative number means that the lower the team's win threshold (i.e. the better the team), the higher the save percentage. If you were on a great team, it was easier to play goal, and if you were on a bad team, it was harder. There was a decently strong relationship between goalie and team that persisted up until the early '90s, but the effect became much less pronounced throughout the 1990s, 2000s and 2010s. This means that the last 25 years or so have been the most team independent years in NHL history for goalie stats, and the reason that save percentage has become an increasingly relied upon tool for goalie evaluation reflects that fact.
(
Important to note: General trends don't mean that there aren't outliers, like the Lemaire Devils or Julien Bruins or whoever. Those still exist, but the general pattern remains clear. Also important to note: If you aren't believing these numbers right now because you are thinking that this doesn't match what you've seen in the NHL very recently, your eyes actually aren't wrong. This trend has reversed since 2013-14, to the point where the correlation for 2016-17 and 2017-18 combined is -0.357, which surprisingly enough puts us right back in 1980s territory. That is a big reason why last year's Vezina nominees were the starting goalies on the top three teams in the standings, for example.)
If the era makes it easier for a goalie to look good on a good team, then goalies on good teams should get disproportionately more awards votes. I'm looking primarily at post-expansion right now, so I divided this period up into 11 year segments to compare the team results of First or Second All-Star goalies. The following chart gives the average points of the goalie's team, the average points of an average team (based on schedule length and OT/SO rules), and the adjusted strength of the team in an 80 game season with no loser points. The last two columns are how many of the 22 All-Star goalies played on teams that finished in the top 3 in the league, and how many played on teams that finished outside of the top-third of the league:
Team Strength of All-Star Goalies:
| Period | Pts | Avg | Adj | Top 3 Tms | Not Top 1/3 |
| 1972-82 | 109 | 79 | 110 | 14 | 0 |
| 1983-93 | 101 | 80 | 101 | 12 | 2 |
| 1994-04 | 97 | 84 | 92 | 5 | 7 |
| 2006-16 | 104 | 91 | 91 | 7 | 7 |
[TBODY]
[/TBODY]
Those numbers basically speak for themselves. When it was easier to put numbers on good teams, and when you basically got excluded you from AST contention because you didn't play on a top team, the competition for awards between the good goalies on the top teams was actually not that much tougher than during the Original Six era. And obviously, the link between AST goalies and Stanley Cups is going to be far stronger in eras where nearly all of the AST goalies are on Stanley Cup contenders.
Speaking of All-Star finishes, here are the top goalies since 1993-94:
| Goalie | 1st | 2nd |
| Brodeur | 3 | 4 |
| Hasek | 6 | 0 |
| Bobrovsky | 2 | 0 |
| Thomas | 2 | 0 |
| Holtby | 1 | 1 |
| Lundqvist | 1 | 1 |
| Rinne | 1 | 1 |
| Luongo | 0 | 2 |
| 8 tied with | 1 | 0 |
| 15 tied with | 0 | 1 |
[TBODY]
[/TBODY]
That's a grand total of two goalies who have more than two AST finishes in the last 25 years, both of whom are inner-circle Hall of Famers and almost universally rated within the top 6 or 7 all-time. From 1931-32 to 1992-93, 14 different goalies made it three or more times, which is triple the rate on a per-season basis. Also, 23 goalies have only one All-Star selection over those 62 seasons. That's actually the exact same number as goalies who have 1 AST from 1993-94 to present.
In other words, you simply can't rate goalies based only on their AST seasons and Stanley Cups anymore, as of course they won't look dominant relative to goalies from earlier eras.
Goalies Compared to Skaters
What about goalie All-Stars relative to other positions?
Post-1994 (24 seasons):
LW: 23 different players (48 spots)
C: 24 different players (48 spots)
RW: 24 different players (48 spots)
D: 36 different players (96 spots)
G: 31 different players (48 spots)
Pre-1994 (62 seasons):
LW: 48 different players (124 spots)
C: 43 different players (124 spots)
RW: 41 different players (124 spots)
D: 77 different players (248 spots)
G: 50 different players (124 spots)
There are about 30% more different goalies than centers or wingers in the post-1994 group, and that's despite 29% of the goalie ASTs in that period being hoarded by two guys. And surprisingly, there are only 5 more unique All-Star defenceman than goalies over the last 24 seasons.
I think All-Star spots for defencemen are clearly less valuable because they more frequently get awarded to the same guys. Someone like Shea Weber has more ASTs than every goalie since 1994 not named Hasek or Brodeur. If you are a voter who likes to use AST voting to compare across positions, I think a rough translation would be something like 2 for a D-man = 1 for a goalie for the '90s-'10s, and something like 3-to-2 for forwards compared to goalies during the same period.
The positional splits are closer to expected in previous eras, when there were relatively fewer individual goalies that got selected and a lot more repeat winners. It's probably still not exactly 1-to-1, and defencemen probably get a few more votes based on reputation (and also probably because defencemen have fewer obvious statistical categories to be judged in than forwards and goalies).
So yes, once you account for the eras they played in and their goalie competition Hasek and Roy were absolutely dominant, both relative to other goalies and relative to other elite skaters.
