View Single Post
01-19-2013, 08:22 AM
Registered User
Canadiens1958's Avatar
Join Date: Nov 2007
Posts: 14,096
vCash: 500
John Ross Roach II

Originally Posted by TheDevilMadeMe View Post
The main reason Roach's GAA is higher is because as an AVERAGE, it averages all years together, and Roach played in the early 1920s, when scoring was much higher than in the late 1920s or even the early 1930s. Roach played from 1920-21 to 1934-35. Connell played from 1924-25 to 1936-37. Chabot played from 1927-28 to 1936-37.

When they were all in the league together (1927-28 to 1934-35), these were their GAAs:

Hainsworth 1.95
Gardiner 2.02
Connell 2.03
Thompson 2.04
Chabot 2.05
Roach 2.14
Worters 2.23

Those 7 are the only goalies who played 280+ games during the time frame, no other goalie played more than 179 games.

That time frame, however, includes Roach's declining phase but not the declining phases of Connell and Chabot. If you look at 1927-28 to 1932-33 (Roach's last season as a full starter), you get:

Hainsworth 1.82
Thompson 1.87
Connell 2.06
Roach 2.08
Chabot 2.10
Gardiner 2.10
Worters 2.14

Same 7 goalies, these 7 all broke 220 games, no other goalie had more than 127 games played.

You can't just compare raw GAA from different time frames when league scoring changes wildly (as it did from the high scoring early 20s to the deadest of dead puck late 20s). Comparing apples to apples, Roach's GAA fits right in with Connell and Chabot. And Roy Worters, widely considered by NHL GMs the best goalie in the late 1920s, who still got significant recognition from writers in the early 30s, has the worst GAA of anyone who was able to keep his starting job during the time frame. Shows the impact of team effects on GAA, even in that era.

Let's look at the performance of NHL goalies from 1920-21 thru 1926-27:

Who is the only regular with a GAA above 3.00? John Ross Roach.
So all the others - Clint Benedict, Georges Vezina, Jake Forbes were at least 0.23 better per game. Effectively Roach was fourth in what was mainly a four team league.

Throw in your data and he was 4th - 6th in a larger league that ranged up to 10 teams. Basically a pick-em between him and Lorne Chabot.

As for declining phase arguments. Roach, Chabot and Worters all had 1900 birthdays, Roach the oldest by 3.5 months over Chabot 4 months over Worters. Hainsworth was a 1895 birthday. So Roach declined faster than the other three.

Either cumulative or split the GAA defines John Ross Roach very accurately.

John Ross Roach was a pedestrian goalie with good longevity.

Canadiens1958 is offline   Reply With Quote