most articles say that it is 'believed' to be no more than 2-3 stones without putting any other reference.
according to goratings.org (active) top pro and bottom pro has 1044 different in Elo, in KGS 230 Elo account for 1 stone difference (I got this from deepmind alphago paper), so this would be 4.5 stones, I'll smooth the number down to 4 stones (this is my own bias hehe), which sound reasonable to me, but what do you think of this?

Unfortunately I doubt elo difference maps to handicap stones in a uniform way. This is pure speculation, but I would guess that pros are better at using the handicap advantage and carrying an advantage from opening to endgame. So I would imagine four stones is "bigger" for a pro than for a typical KGS player. The only way to know for sure would be to have pro handicap games, which aren't too common these days. Someone who knows the history better might be able to say more from the time period when pros did play with handicap.

...undefined.

Because nowadays there are stronger players with lower ranks.

Can we observe indirectly using high rank player in KGS? If we set 7D KGS player as a bottom pro, the top KGS player right now is around 10D, and of course the real world top pro is obviously stronger than top KGS player, that would put the real world top pro around 11D (or more). 11-7 = 4 can this work?


What do you mean? I don't talk about professional ranking system at all.

Why would a bottom pro be 7d kgs?

I try to connect those account to real world pro, for example, account user HYang is Huiren Yang, Chinese 1p rank no. 771 out of 836 player in goratings, is around 7D in KGS.

There was also that female pro, I forget her account, who was 6d on KGS.

The publication WeiQiTianDi would publish a table of the active pros in China with their ratings based on actual tourney results; I don't know if they still do this (annually ? bi-annually ? ). Maybe someone with access to a recent issue can confirm this.

( I also don't know if Japan and Korea have similar, publicly available rating systems based on current tourney results. )

Baduk.or.kr seem to have a monthly ranking table of the top 100 Korean pro's, probably based on Dr Bea Teal's ranking system.
It is in Korean, however.

I believe Gogameguru had a link to an explanation of the rating system.

What's interesting is that mamumamu0143's site gives monthly ranking of all Japanese pro's using a different rating system, monthly world rankings using the same ranking system, and then somehow, from somewhere, we can compare it to Dr Bae Teal's quarterly table of the top 100 pro's (all the names are in kanji)

_________________
On Go proverbs:
"A fine Gotation is a diamond in the hand of a dan of wit and a pebble in the hand of a kyu" —Joseph Raux misquoted.

What do you mean by professional "levels" in this thead's title then? I think I had the same impression as RJ.

My understanding of the existing ratings systems is that they attempt to express the likely winning percentage between different players. This may or may not map to handicap stones. On mamumamu's site in his analysis of komi he writes that we lack sufficient handicap game results between pros to extend the results from komi to handicap stones.

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

I'm trying to find the strength difference between active top pro and active bottom pro that is ranked by Elo ranking system (to be specific, Whole History Rating system used by goratings.org), not professional dan system. By professional levels I mean any active pro, not level like 1p - 9p. Sorry if you misunderstood because of my poor English skill. I just copied that sentence from Wikipedia article 'Go Professional' under the topic 'Discrepancies among professionals'

Hi pookpooi, I understood your question; no problem with your English.

A related side question: what are you going to do with this information ?
For example, I don't follow F-1 racing;
but I heard Michael Schumacher was #1 at a time.
Maybe his average time was some small fraction of one second better than his closest competitor. That's an interesting trivia -- not much I can do with it.
Or: why do you seek this information ? (I'm curious.)

You don't have sufficient data (a sample of handicap games between "top" and "bottom" active pros) to test your hypothesis, and you are being vague on some terms. For example, what exactly makes an "active" pro? 1 game a year? 1 game a week? Someone who still plays Go? I also think it's unreasonable to base it off KGS ratings. You don't know what conditions Huiren Yang is playing under or how seriously he plays online. And would he even be considered active? Unless you set up an experiment where you can convince professionals that meet your criteria to play handicap games seriously (good luck convincing Rin Kaiho to play a 3 stone handicap game against Ke Jie based on their ELO scores), all of this is just wild guessing.

You mean: "rated by an Elo style rating system". Ranks might then be derived from ratings.

It started from when I'd enough money and I want to invest in deep learning start-up, but it's long story you don't wanna hear the whole of it.
The shortcut, every time I read article about Go professional I've to read 'the difference is believed to be no more than 2-3 stones' and I've begin to wonder, how do they get that number? Is that reliable? So with my free time right now that I don't wanna spend it on practicing go, I'm trying to find that answer by myself (with a lot of internet help)

Goratings.org only rank pro that had play at least one game in the last 1,000 days, that's what I use to define active pro.

We don't know how many stone is inside the black box, we can't open it. We've been told (from anonymous) that inside the box is 'believed' to be 2-3 stones. But I don't believe him, and I'm not randomly guess the answer either. Instead, I'm shaking the box to hear what it sounds like, and compare it to another box I made that I put 4 stones in it. When the box is finally open I may be completely wrong, may be there's no stone in it at all, or may be, there are tens of them!

Thank you, that's exactly what I mean.

Hi pookpooi, Thanks -- You want a trustworthy little bird.

400 years ago an 8 dan, if one existed, would give a ("pro") shodan 4 stones, and a 9 dan would give a shodan 4-5 stones. Over time that handicap diminished, probably because shodans became stronger on average. 100 years ago a 9 dan would give a shodan 3 stones. This difference between ranks lasted until WWII, when 9 dans began to proliferate. The average strength of 9 dans probably diminished by as much as 1/2 stone. It is also possible that the average strength of shodans increased. Around 1980 Ing devised pro ranks such that a 7 dan would give a shodan 2 stones and a 9 dan would give a shodan the equivalent of 2.5 stones. These days pros no longer give handicaps to other pros. Maybe the 2-3 stones you have heard comes from Ing's ranking system.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Now what I need more from these information is that, after given such a handicap, does winrate balanced to around 50:50? (I don't think so)
But I agree that pros shouldn't give handicaps to other pros, and I also strongly agree with the current professional dan system (that the rank never decrease and you can promote by having a good tournament result, or etc.)

Lee Sedol beat a group of Western pros down to a three stone handicap, and the Japanese professional vs. amateur Meijin contest often sees the professionals winning with 2 stones.

These are both very weak signals, and your question is still not perfectly defined. Still, I think that >= 2 stones is a safe bet.

_________________
Occupy Babel!

Yeah, that's the kind of information I want to hear more. I hope those matches are not play on the internet cause some people might say that they're not playing seriously because the match happen online.