overpass
Registered User
- Jun 7, 2007
- 5,274
- 2,819
I used the Hockey Summary Project records to track the average quality of opposition faced by NHL goaltenders from 1967-68 to 1986-87.
This is of interest for several reasons. For some, it may be worthwile in its own right to look at the way in which NHL teams deployed their goalies. For those who look at NHL goalie stats, it’s important to know how those stats were affected by the strength of opposition they faced. And for those who wish to compare the records of goalies on the same team, it’s very important to know which of them, if either, got the tougher starts.
Method: For each goaltender, I calculated the average winning percentage and the average goals per game of their opponents. I weighted each game equally, not by minutes played, which is not ideal but shouldn’t make much difference. I also made an additional adjustment for home/road. For each road game played by a goalie, I increased the opponent’s winning percentage by 40 points (0.040). I also multiplied their opponent’s goals per game by 1.07. For home games, I applied the adjustments in reverse, subtracting and dividing instead of multiplying and adding. I made no adjustments for back-to-backs or anything else that could affect the difficulty in a particular game.
Here are some results for several goaltenders. Just to be clear, the numbers I'm posting here are only the strength of opposition numbers. I'm not posting their actual records - hockey-reference.com has that easily available if you must see it.
Strength of Opposition – Ken Dryden vs Backups
Year | GP(KD) | GP(Oth) | OppW%(KD) | OppW%(Oth) | OppG/G(KD) | OppG/G(Oth)
1972 | 64 | 16 | 0.489 | 0.488 | 2.91 | 2.93
1973 | 54 | 27 | 0.487 | 0.474 | 3.14 | 3.17
1975 | 56 | 25 | 0.489 | 0.469 | 3.40 | 3.26
1976 | 62 | 22 | 0.490 | 0.433 | 3.40 | 3.22
1977 | 56 | 26 | 0.491 | 0.454 | 3.27 | 3.08
1978 | 52 | 30 | 0.502 | 0.443 | 3.30 | 3.07
1979 | 47 | 34 | 0.493 | 0.462 | 3.47 | 3.43
Not surprisingly, Ken Dryden tended to face stronger opponents than his backup goalies. This tendency was most pronounced during the dynasty years of 1975-76 to 1978-79.
Using a weighted average, here are my estimates for Dryden vs his backups in the regular season, and the amount of the difference that is explained by the different opposition they faced.
Win %: Dryden was 0.026 worse than his backups. He faced opposition that was, on average, stronger than his backups faced by by 0.034.
GAA: Dryden was 0.44 better than his backups. He faced opposition that was, on average, stronger than his backups faced by by 0.11.
Strength of Opposition – Tony Esposito vs Backups
Year | GP(TO) | GP(Oth) | OppW%(TO) | OppW%(Oth) | OppG/G(TO) | OppG/G(Oth)
1970 | 63 | 14 | 0.519 | 0.441 | 2.82 | 2.69
1971 | 57 | 25 | 0.483 | 0.504 | 3.01 | 3.00
1972 | 48 | 34 | 0.455 | 0.526 | 2.84 | 3.19
1973 | 56 | 32 | 0.470 | 0.560 | 3.07 | 3.41
1974 | 70 | 10 | 0.486 | 0.469 | 3.09 | 3.12
1975 | 71 | 12 | 0.507 | 0.449 | 3.43 | 3.25
1976 | 68 | 15 | 0.471 | 0.574 | 3.29 | 3.74
1977 | 69 | 14 | 0.495 | 0.569 | 3.27 | 3.70
1978 | 64 | 16 | 0.483 | 0.502 | 3.27 | 3.40
1979 | 63 | 17 | 0.462 | 0.552 | 3.39 | 3.63
1980 | 69 | 11 | 0.492 | 0.535 | 3.53 | 3.58
1981 | 66 | 15 | 0.493 | 0.529 | 3.81 | 3.96
1982 | 52 | 29 | 0.479 | 0.512 | 3.92 | 4.12
1983 | 39 | 39 | 0.493 | 0.470 | 3.80 | 3.81
Tony Esposito tended to face easier opposition than his backup goalies. Chicago preferred to play their backups on the road.
Weighted average estimates
Win %: Esposito was 0.059 better than his backups. He faced opposition that was, on average, weaker than his backups faced by by 0.034.
GAA: Esposito was 0.37 better than his backups. He faced opposition that was, on average, weaker than his backups faced by by 0.17.
These estimates are affected quite a bit by Esposito’s weaker last two years, in which he played fewer games, because of the weighting method. After removing those two seasons, from 1970-1981:
Win %: Esposito was 0.094 better than his backups. He faced opposition that was, on average, weaker than his backups faced by 0.054.
GAA: Esposito was 0.74 better than his backups. He faced opposition that was, on average, weaker than his backups faced by 0.23.
Strength of Opposition – Billy Smith vs Backups
Year | GP(BS) | GP(Oth) | OppW%(BS) | OppW%(Oth) | OppG/G(BS) | OppG/G(Oth)
1973 | 37 | 45 | 0.530 | 0.527 | 3.32 | 3.29
1974 | 46 | 36 | 0.530 | 0.503 | 3.23 | 3.17
1975 | 58 | 25 | 0.517 | 0.459 | 3.49 | 3.28
1976 | 39 | 44 | 0.452 | 0.524 | 3.29 | 3.48
1977 | 36 | 46 | 0.490 | 0.493 | 3.34 | 3.29
1978 | 38 | 46 | 0.517 | 0.466 | 3.28 | 3.24
1979 | 40 | 43 | 0.500 | 0.491 | 3.53 | 3.45
1980 | 38 | 47 | 0.497 | 0.501 | 3.50 | 3.53
1981 | 41 | 43 | 0.486 | 0.489 | 3.82 | 3.80
1982 | 46 | 36 | 0.514 | 0.459 | 4.04 | 3.90
1983 | 41 | 44 | 0.487 | 0.487 | 3.79 | 3.82
1984 | 42 | 49 | 0.495 | 0.485 | 3.84 | 3.88
1985 | 37 | 49 | 0.488 | 0.491 | 3.80 | 3.85
1986 | 41 | 45 | 0.512 | 0.485 | 3.85 | 3.84
1987 | 40 | 46 | 0.494 | 0.505 | 3.67 | 3.73
Billy Smith is an interesting case. He split regular season playing time equally with his fellow goaltenders for almost his whole career. And his regular season results were very similar to theirs. It appears that he faced slightly stronger opponents over most of his career, but not by much.
Weighted average estimates
Win %: Smith was 0.026 worse than his backups. He faced opposition that was, on average, stronger than his backups faced by 0.007.
GAA: Smith was 0.14 worse than his backups. He faced opposition that was, on average, identical in this respect to that which his backups faced.
Smith played opposite a number of different goaltenders, so we can break his results down by goaltender.
1973-74 (Gerry Desjardins): Smith’s winning percentage was better (0.034) against stronger opposition (0.015). His GAA was better (-0.29) against stronger opposition (0.05).
1975-81 (Chico Resch): Smith’s winning percentage was worse (-0.040) against stronger opposition (0.003). His GAA was worse (0.21) against stronger opposition (0.01).
1982-84 (Rollie Melanson): Smith’s winning percentage was worse (-0.038) against stronger opposition (0.017). His GAA was worse (0.09) against equal opposition in this respect.
1985-87 (Kelly Hrudey): Smith’s winning percentage was worse (-0.038) against stronger opposition (0.004). His GAA was worse (0.23) against weaker opposition (-0.03).
Strength of Opposition – Dan Bouchard vs Backups
Year | GP(DB) | GP(Oth) | OppW%(DB) | OppW%(Oth) | OppG/G(DB) | OppG/G(Oth)
1973 | 34 | 46 | 0.517 | 0.498 | 3.28 | 3.23
1974 | 46 | 36 | 0.500 | 0.506 | 3.15 | 3.15
1975 | 40 | 40 | 0.520 | 0.487 | 3.51 | 3.39
1976 | 47 | 37 | 0.529 | 0.483 | 3.49 | 3.40
1977 | 42 | 43 | 0.544 | 0.477 | 3.50 | 3.23
1978 | 58 | 26 | 0.490 | 0.483 | 3.28 | 3.23
1979 | 64 | 23 | 0.523 | 0.526 | 3.55 | 3.57
1980 | 53 | 32 | 0.520 | 0.426 | 3.56 | 3.38
1981 | 29 | 53 | 0.473 | 0.529 | 3.77 | 3.97
1982 | 60 | 22 | 0.511 | 0.537 | 3.98 | 4.06
1983 | 50 | 32 | 0.498 | 0.528 | 3.82 | 3.99
1984 | 57 | 29 | 0.504 | 0.515 | 3.86 | 3.99
1985 | 29 | 55 | 0.526 | 0.492 | 3.90 | 3.81
1986 | 32 | 61 | 0.474 | 0.517 | 3.85 | 4.05
Although he's not well remembered by history, Dan Bouchard outperformed the other goaltenders on his team by quite a bit over his career. (This was first brought to my attention on the Contrarian Goaltender's blog.) As a result, I decided to examine his strength of opposition here. It turns out that he faced clearly stronger opposition while with the Flames, and slightly weaker opposition with Quebec and Winnipeg. He was, at least, a much better NHL goaltender than Phil Myre.
Weighted average estimates
Win %: Bouchard was 0.056 better than his backups. He faced opposition that was, on average, stronger than his backups faced by 0.005.
GAA: Bouchard was 0.47 better than his backups. He faced opposition that was, on average, weaker than his backups faced by 0.01.
Breaking it down to his years with the Flames, and after leaving for Quebec:
1973-80 (Flames): Bouchard’s winning percentage was better (0.053) against stronger opposition (0.035). His GAA was better (-0.26) against stronger opposition (0.11).
1981-86 (post-Flames): Bouchard’s winning percentage was better (0.059) against weaker opposition (-0.022). His GAA was better (-0.66) against weaker opposition (-0.11).
This is of interest for several reasons. For some, it may be worthwile in its own right to look at the way in which NHL teams deployed their goalies. For those who look at NHL goalie stats, it’s important to know how those stats were affected by the strength of opposition they faced. And for those who wish to compare the records of goalies on the same team, it’s very important to know which of them, if either, got the tougher starts.
Method: For each goaltender, I calculated the average winning percentage and the average goals per game of their opponents. I weighted each game equally, not by minutes played, which is not ideal but shouldn’t make much difference. I also made an additional adjustment for home/road. For each road game played by a goalie, I increased the opponent’s winning percentage by 40 points (0.040). I also multiplied their opponent’s goals per game by 1.07. For home games, I applied the adjustments in reverse, subtracting and dividing instead of multiplying and adding. I made no adjustments for back-to-backs or anything else that could affect the difficulty in a particular game.
Here are some results for several goaltenders. Just to be clear, the numbers I'm posting here are only the strength of opposition numbers. I'm not posting their actual records - hockey-reference.com has that easily available if you must see it.
Strength of Opposition – Ken Dryden vs Backups
1972 | 64 | 16 | 0.489 | 0.488 | 2.91 | 2.93
1973 | 54 | 27 | 0.487 | 0.474 | 3.14 | 3.17
1975 | 56 | 25 | 0.489 | 0.469 | 3.40 | 3.26
1976 | 62 | 22 | 0.490 | 0.433 | 3.40 | 3.22
1977 | 56 | 26 | 0.491 | 0.454 | 3.27 | 3.08
1978 | 52 | 30 | 0.502 | 0.443 | 3.30 | 3.07
1979 | 47 | 34 | 0.493 | 0.462 | 3.47 | 3.43
Not surprisingly, Ken Dryden tended to face stronger opponents than his backup goalies. This tendency was most pronounced during the dynasty years of 1975-76 to 1978-79.
Using a weighted average, here are my estimates for Dryden vs his backups in the regular season, and the amount of the difference that is explained by the different opposition they faced.
Win %: Dryden was 0.026 worse than his backups. He faced opposition that was, on average, stronger than his backups faced by by 0.034.
GAA: Dryden was 0.44 better than his backups. He faced opposition that was, on average, stronger than his backups faced by by 0.11.
Strength of Opposition – Tony Esposito vs Backups
1970 | 63 | 14 | 0.519 | 0.441 | 2.82 | 2.69
1971 | 57 | 25 | 0.483 | 0.504 | 3.01 | 3.00
1972 | 48 | 34 | 0.455 | 0.526 | 2.84 | 3.19
1973 | 56 | 32 | 0.470 | 0.560 | 3.07 | 3.41
1974 | 70 | 10 | 0.486 | 0.469 | 3.09 | 3.12
1975 | 71 | 12 | 0.507 | 0.449 | 3.43 | 3.25
1976 | 68 | 15 | 0.471 | 0.574 | 3.29 | 3.74
1977 | 69 | 14 | 0.495 | 0.569 | 3.27 | 3.70
1978 | 64 | 16 | 0.483 | 0.502 | 3.27 | 3.40
1979 | 63 | 17 | 0.462 | 0.552 | 3.39 | 3.63
1980 | 69 | 11 | 0.492 | 0.535 | 3.53 | 3.58
1981 | 66 | 15 | 0.493 | 0.529 | 3.81 | 3.96
1982 | 52 | 29 | 0.479 | 0.512 | 3.92 | 4.12
1983 | 39 | 39 | 0.493 | 0.470 | 3.80 | 3.81
Tony Esposito tended to face easier opposition than his backup goalies. Chicago preferred to play their backups on the road.
Weighted average estimates
Win %: Esposito was 0.059 better than his backups. He faced opposition that was, on average, weaker than his backups faced by by 0.034.
GAA: Esposito was 0.37 better than his backups. He faced opposition that was, on average, weaker than his backups faced by by 0.17.
These estimates are affected quite a bit by Esposito’s weaker last two years, in which he played fewer games, because of the weighting method. After removing those two seasons, from 1970-1981:
Win %: Esposito was 0.094 better than his backups. He faced opposition that was, on average, weaker than his backups faced by 0.054.
GAA: Esposito was 0.74 better than his backups. He faced opposition that was, on average, weaker than his backups faced by 0.23.
Strength of Opposition – Billy Smith vs Backups
1973 | 37 | 45 | 0.530 | 0.527 | 3.32 | 3.29
1974 | 46 | 36 | 0.530 | 0.503 | 3.23 | 3.17
1975 | 58 | 25 | 0.517 | 0.459 | 3.49 | 3.28
1976 | 39 | 44 | 0.452 | 0.524 | 3.29 | 3.48
1977 | 36 | 46 | 0.490 | 0.493 | 3.34 | 3.29
1978 | 38 | 46 | 0.517 | 0.466 | 3.28 | 3.24
1979 | 40 | 43 | 0.500 | 0.491 | 3.53 | 3.45
1980 | 38 | 47 | 0.497 | 0.501 | 3.50 | 3.53
1981 | 41 | 43 | 0.486 | 0.489 | 3.82 | 3.80
1982 | 46 | 36 | 0.514 | 0.459 | 4.04 | 3.90
1983 | 41 | 44 | 0.487 | 0.487 | 3.79 | 3.82
1984 | 42 | 49 | 0.495 | 0.485 | 3.84 | 3.88
1985 | 37 | 49 | 0.488 | 0.491 | 3.80 | 3.85
1986 | 41 | 45 | 0.512 | 0.485 | 3.85 | 3.84
1987 | 40 | 46 | 0.494 | 0.505 | 3.67 | 3.73
Billy Smith is an interesting case. He split regular season playing time equally with his fellow goaltenders for almost his whole career. And his regular season results were very similar to theirs. It appears that he faced slightly stronger opponents over most of his career, but not by much.
Weighted average estimates
Win %: Smith was 0.026 worse than his backups. He faced opposition that was, on average, stronger than his backups faced by 0.007.
GAA: Smith was 0.14 worse than his backups. He faced opposition that was, on average, identical in this respect to that which his backups faced.
Smith played opposite a number of different goaltenders, so we can break his results down by goaltender.
1973-74 (Gerry Desjardins): Smith’s winning percentage was better (0.034) against stronger opposition (0.015). His GAA was better (-0.29) against stronger opposition (0.05).
1975-81 (Chico Resch): Smith’s winning percentage was worse (-0.040) against stronger opposition (0.003). His GAA was worse (0.21) against stronger opposition (0.01).
1982-84 (Rollie Melanson): Smith’s winning percentage was worse (-0.038) against stronger opposition (0.017). His GAA was worse (0.09) against equal opposition in this respect.
1985-87 (Kelly Hrudey): Smith’s winning percentage was worse (-0.038) against stronger opposition (0.004). His GAA was worse (0.23) against weaker opposition (-0.03).
Strength of Opposition – Dan Bouchard vs Backups
1973 | 34 | 46 | 0.517 | 0.498 | 3.28 | 3.23
1974 | 46 | 36 | 0.500 | 0.506 | 3.15 | 3.15
1975 | 40 | 40 | 0.520 | 0.487 | 3.51 | 3.39
1976 | 47 | 37 | 0.529 | 0.483 | 3.49 | 3.40
1977 | 42 | 43 | 0.544 | 0.477 | 3.50 | 3.23
1978 | 58 | 26 | 0.490 | 0.483 | 3.28 | 3.23
1979 | 64 | 23 | 0.523 | 0.526 | 3.55 | 3.57
1980 | 53 | 32 | 0.520 | 0.426 | 3.56 | 3.38
1981 | 29 | 53 | 0.473 | 0.529 | 3.77 | 3.97
1982 | 60 | 22 | 0.511 | 0.537 | 3.98 | 4.06
1983 | 50 | 32 | 0.498 | 0.528 | 3.82 | 3.99
1984 | 57 | 29 | 0.504 | 0.515 | 3.86 | 3.99
1985 | 29 | 55 | 0.526 | 0.492 | 3.90 | 3.81
1986 | 32 | 61 | 0.474 | 0.517 | 3.85 | 4.05
Although he's not well remembered by history, Dan Bouchard outperformed the other goaltenders on his team by quite a bit over his career. (This was first brought to my attention on the Contrarian Goaltender's blog.) As a result, I decided to examine his strength of opposition here. It turns out that he faced clearly stronger opposition while with the Flames, and slightly weaker opposition with Quebec and Winnipeg. He was, at least, a much better NHL goaltender than Phil Myre.
Weighted average estimates
Win %: Bouchard was 0.056 better than his backups. He faced opposition that was, on average, stronger than his backups faced by 0.005.
GAA: Bouchard was 0.47 better than his backups. He faced opposition that was, on average, weaker than his backups faced by 0.01.
Breaking it down to his years with the Flames, and after leaving for Quebec:
1973-80 (Flames): Bouchard’s winning percentage was better (0.053) against stronger opposition (0.035). His GAA was better (-0.26) against stronger opposition (0.11).
1981-86 (post-Flames): Bouchard’s winning percentage was better (0.059) against weaker opposition (-0.022). His GAA was better (-0.66) against weaker opposition (-0.11).