@S_Alexander I’m curious where this formula comes from.
I did an analysis of the historical data from EGF tournament games (about 1 million games since 1996) and I got a curve like this (converted to your units, although an arbitrary constant may be added to the rating axis)
Overall, it is somewhat similar, but the curvature of my graph is stronger and the rating scale is wider.
I think that formula is not from Prof.Glickman. It’s a formula to convert from Glicko-2 ratings to go ranks and I don’t think that go ranks was a concern of Prof. Glickman.
The Glicko-2 ratings are as described by Prof. Glickman, the rank conversion formula was created by anoek by mapping it to certain accounts that were considered “known and stable” though which accounts he chose and how are unknown to me.