I appreciate the detailed responses overpass, and while I don't think I'm going to make paragraph responses in this post to those points explicitly, I'm going to try to compose some in the near future. Let me first try a different tack in illustrating my initial point, breaking the sample down to individual seasons. There are 12 seasons for which we have save percentage data pre-expansion, 55-56 through 66-67. We have more than 12 seasons post-lockout, but let's choose the last 12 years, 12-13 through 23-24. Now let's take the AS-1, AS-2, AS-3, and AS-4 finishers each year, and average out their GAA and save percentages, as well as games played. Here's an initial table:
| LA GF | LA GF/G | LA GF/82 | | LA GF | LA GF/G | LA GF/82 |
55-56 | 177 | 2.53 | 207.3 | 12-13 | 127 | 2.65 | 217.0 |
56-57 | 188 | 2.69 | 220.2 | 13-14 | 219.0 | 2.67 | 219.0 |
57-58 | 196 | 2.80 | 229.6 | 14-15 | 218.0 | 2.66 | 218.0 |
58-59 | 203 | 2.90 | 237.8 | 15-16 | 219.0 | 2.67 | 219.0 |
59-60 | 206 | 2.94 | 241.3 | 16-17 | 223.0 | 2.72 | 223.0 |
60-61 | 210 | 3.00 | 246.0 | 17-18 | 240.0 | 2.93 | 240.0 |
61-62 | 211 | 3.01 | 247.2 | 18-19 | 244.0 | 2.98 | 244.0 |
62-63 | 208 | 2.97 | 243.7 | 19-20 | 208 | 2.97 | 243.7 |
63-64 | 194 | 2.77 | 227.3 | 20-21 | 162 | 2.89 | 237.2 |
64-65 | 201 | 2.87 | 235.5 | 21-22 | 255.0 | 3.11 | 255.0 |
65-66 | 213 | 3.04 | 249.5 | 22-23 | 258.0 | 3.15 | 258.0 |
66-67 | 209 | 2.99 | 244.8 | 23-24 | 253.0 | 3.09 | 253.0 |
| | 2.88 | 235.8 | | | 2.87 | 235.6 |
Basically, almost exactly the same scoring level in each 12 year period. Also, because there were 3 shortened seasons in the second sample, teams only played 912 games during those years, compared to the 910 games played pre-expansion. [I could have opted for 05-06 through 16-17 as an alternative sample, though that would've been lower scoring, at 225.5.]
Now, here's the aggregated average of the postseason all star place each year. There were a few years with missing votes pre-expansion, so for those years I chose the best season by GA%- not already picked - ends up being 3 Worsley seasons, 2 Bower seasons, and 1 Simmons season to cover 57-58 3rd/4th, 58-59 3rd/4th, 59-60 4th, 61-62 4th.
| | | AS-1 | | | | | | | AS-2 | | | |
| GP | Shots | Sh/G | GA | GAA | SV% | | GP | Shots | Sh/G | GA | GAA | SV% |
66-67 | 66.83 | 2015.17 | 30.170 | 160.17 | 2.413 | 0.921 | | 59.17 | 1677.33 | 28.132 | 141.92 | 2.428 | 0.915 |
23-24 | 55.42 | 1649.75 | 29.781 | 118.42 | 2.175 | 0.928 | | 54.50 | 1568.58 | 28.702 | 116.50 | 2.176 | 0.926 |
| | | AS-3 | | | | | | | AS-4 | | | |
| GP | Shots | Sh/G | GA | GAA | SV% | | GP | Shots | Sh/G | GA | GAA | SV% |
66-67 | 50.42 | 1586.67 | 31.322 | 130.67 | 2.582 | 0.918 | | 54.67 | 1758.00 | 31.624 | 153.08 | 2.788 | 0.913 |
23-24 | 59.00 | 1693.83 | 28.582 | 135.58 | 2.319 | 0.920 | | 54.08 | 1606.83 | 29.610 | 124.33 | 2.352 | 0.923 |
You have a couple outliers in games played, but 4 of the 8 samples are right at 55 games. When you break down the games played into 10s, 11 pre-expansion goalies played 70, while only 2 modern goalies did, then 16 pre-expansion goalies played between 60-70, while 19 modern goalies did (27-21 total), then 9 pre-expansion goalies played between 50-60, while 13 modern goalies did (36-34 total), then 3 pre-expansion goalies played between 40-50, while 10 modern goalies did (39-44 total), and finally 9 pre-expansion goalies played between 30-40, while 4 modern goalies did (48-48 end total).
Professor What did an aggregate of the 6th best goaltender from the lockout just above, and came up with a GAA of 2.36 and a save percentage of 0.919, virtually indistinguishable from the AS-4 line. There's basically a packet of 4th-10th goalie-seasons that you'd take ahead of the AS-4 pre-expansion because all of those seasons are below-average. They don't magically become above-average stats just because they're the 4th best season of the 6-10 total seasons. I mean, look at the GA numbers - in virtually the exact same number of games (54), modern goalies are giving up 29 less goals in a year.
I had a different point continuing on from here, but it expanded enough that I'm gonna put it in a separate post. Let's just conclude by doing a quick comparison - the Serry Tawchuk character from my last post, and Semyon Varlamov's career from 11-12 to 23-24. Before I post the stats, let me state that I'm not trying to downgrade Sawchuk or upgrade Varlamov - Sawchuk's probably still in my top 10, Varlamov is not making my top 60, if he makes it at all. The seasons I'm excluding from Sawchuk are 5 peak years as good as any goaltender in history. Anyways, here's the stats:
Name | GP | Shots | Saves | Sh/G | GA | GAA | SV% | Min |
Sawchuk | 569 | 16979 | 15408 | 29.84 | 1571 | 2.81 | 0.907 | 33526 |
Varlamov | 552 | 16727 | 15320 | 30.30 | 1407 | 2.67 | 0.916 | 31611 |
And here's the league around them -
Sawchuk and
Varlamov - and the most relevant issue is what to set the minimum games played to get your rankings. I have it set at 100 games initially, but you could also use 200 or 400.
| Sawchuk | | Varlamov |
100 | 5/16 GAA; 7/16 SV% | | t43/108 GAA; t16/108 SV% |
200 | 5/9 GAA; 6/9 SV% | | 32/54 GAA; t14/54 SV% |
400 | 4/5 GAA; 5/5 SV% | | 19/26 GAA; t8/26 SV% |
All of the lower games played goalies (apart from Hodge) are below Sawchuk in their statistics, so he's anywhere between the 3rd-5th best goaltender of his era, depending on how much you value the 9000 more minutes he played than Johnny Bower, and what you think of Worsley, no matter which games played metric you use. On the other hand, you're basically dividing Varlamov's sample in half every jump, 108 to 54 to 26. Based on a minimum of 400 games played, Varlamov is somewhere between the 8th and 19th best goaltender of his era. Feel free to go through all 26 to rank him more precisely, my exact ranking isn't really the purpose of this comparison.
In a scoring environment that's virtually identical, their shots faced and saves are nearly identical, and I'm not even trying to claim anything about which stats are better - they're just broadly similar. I was casting about for another similar goaltender, and this one turned up. Scoring's significantly higher in this person's era between 67-68 and 77-78 (I cut out their first year and their last 4), but here's the career:
Name | GP | Shots | Saves | Sh/G | GA | GAA | SV% | Min |
Vachon | 576 | 16464 | 14899 | 28.58 | 1565 | 2.80 | 0.905 | 33530 |
Vachon made it to the Hall of Fame, but it took him 35 years to do so.
Here's his contemporaries and their numbers. It's weird that there's 6 guys that played 500 games, and nobody that played 400 in the sample, but whatever. Vachon's probably the 5th or 6th best goaltender of his era.
I don't know whether or not the aggregate comparisons or the 3-way comparison make my point in any sensible manner, but I've spent way too much time on this post to delete it now. I will be posting that separate post shortly after this, which also may or may not illustrate my points.