Please (re)instate ranks beyond 25k and 9d

I doubt that would be technically possible for reasons I’ve outlined in the last two paragraphs of this post (although I’m not a developer and don’t really have a clue beyond speculation): Please (re)instate ranks beyond 25k and 9d - #46 by Maharani

Just for clarification, the automatch and ranking system works even for strong 9d players because the ELO rating is used rather than the displayed rank (which is simply converted from the ELO by a simple formula). The current system already makes strong 9d give handicap stones to weaker 9d. It just doesn’t display the stronger ones as 10d. My suggestion to display ratings as 10d etc. is purely cosmetic.

1 Like

Interesting, I didn’t know that.

1 Like

That is not currently the case. I suggested that we might make a change like that, if this proposal did not get up.

It’s not the case because the glicko/ELO rating extends beyond the “max”. So both players likely have ratings where if one beats the other, the weaker still remains 9d.

GaJ

1 Like

I’m against rank caps too.
It leads only to increasing of stone difference within 9d and within 25k.
The purpose of kyu and dan is to be more helpful than glicko numbers.

3 Likes

I like your username ;D

1 Like

The way I understand it, a 9th dan amateur is about the strength of a top professional, or a 7-9p. Any legitimate (say) 11 dan would be stronger than most, if not all, pros. I, for one, favor a ceiling, but with no floor, or at least a lower floor - say, 30, 35, or even 40 kyu. (Maybe not 40 kyu, but the first Go book I ever read - Go: More than a Game, by Peter Shotwell - says a beginner starts at 35 kyu, a practice I never heard of elsewhere, but one that might be, for all I know.)

Amateur ranks in offline federations only extend up to 7d anyway, right? Regardless of rank, there are very few amateurs of even low-tier professional strength (though there are some) and none of top professional strength. Typically these super-strong amateurs eventually pass the professional qualification exam / tournament and become pros anyway.

3 Likes

That’s how I understand it. In offline federations, there is no 8d (or 9d). Once you’re 7d, you can take the exam to become 1p.

So since we here at OGS DO have 8d and 9d ranks, I don’t see a reason to cap it there :slight_smile:

1 Like

There’s maybe a handful of players who are actually 5d+, and most of them don’t even play. ~.~ Talking about 9d+ ranks is totally irrelevant imho.

On the other hand, talking about 26-30k might not be.

2 Likes

I think 9d+ ranks would mainly be interesting for comparing bots such as Leela and AQ.

With all due respect, I disagree. A rank exists only with respect to standing among a population. Since the bots will have very little in the way of contact (games) with said population (in this case OGS players), these kinds of ranks would be misleading. If you just want to compare bots, we could have a separate bot ladder to satisfy your comparison needs. (Ideally then, any comparison between bots should also happen on equal hardware.)

2 Likes

I am convinced by this thread that there is a good case for having ranks down to 35k. When it comes to the dan ranks, I wouldn’t make any changes unless a significant number of dan players were requesting it. It should be up to them.

5 Likes

The real question is, would @anoek and @matburt even consider introducing ranks weaker than 25 kyu? I would imagine the time to do it would have been during the introduction of the new glicko rating system. Should we take the fact that they didn’t as a telling sign of their intent not to?

1 Like

This topic was automatically closed 91 days after the last reply. New replies are no longer allowed.