TL;DR: OGS implements handicap 1 as no extra stone, with 0.5 komi because that is tradition. That gives black about 6 points extra, which is less than the 12/14 points actually needed to bridge the rank gap. Therefore it is not expected that handicap 1 to fully bridge one rank-gap. OGS-glicko2 appears to handle this correctly.
OGS implements handicap 1 as: no extra stone, with 0.5 komi. That is not arbitrary, it follows the traditional Go convention. The confusing part is that this is often still called “one stone handicap”, even though no extra stone is actually placed. Here is an external source, where you see that Handicap 1 is called “1 stone”. It doesn’t provide an extra stone, just no komi.
https://en.wikipedia.org/wiki/Handicapping_in_Go
So two things are true at the same time:
-
An actual extra handicap stone is understood/meant as compensating roughly one rank of strength.
-
The first “handicap stone” is not an actual extra stone. It is basically Black’s normal first move, combined with reduced komi, usually 0.5 to avoid draws.
As Shinuito, JonKo, I, and possibly others have already pointed out in this thread: if an actual extra stone is valued at roughly 12–14 points, then reducing komi by about 6 points cannot also be worth a full stone/rank. Six is simply less than twelve.
So yes, by this logic, the Handicap 1: 0.5 komi does not bridge the gap of one rank strength, it bridges roughly half of it.
If you have the expectation that it should, then that expectation is totally understandable, because it is often taught that in Go, ranks can be bridged by handicap. And the naming seems to confirm it. But it is not mathematically supported because black isn’t given the twelve to fourteen points necessary to make it true. It holds for the later actual handicap stones, but not for the first non-stone.
This is also why OGS-Glicko2 still expects white to win more than 50% in a handicap 1 game aginst black. It recognises that black is still weaker than white.
