Handicap Stone vs Go Rank, Is it really linear?

From my personal experience of playing handicap games, I feel that x stone difference doesn’t translate well to x handicap stones especially for larger handicap and higher ranked players. I might be possible for 4k to give 6k 2 stones to even out the game. However, I don’t think 4k can give a 13k 9 stones with 50-50 chance of winning. Bu I feel that a 13k might be able to give 22k 9 stones with greater than 50% chance of winning.

What do you guys think? Is there a handicap-rank table anywhere?

1 Like

Hi there, i just looked on wikipedia. https://en.m.wikipedia.org/wiki/Handicap_(go)

What they have to say is interesting on this matter. However i do have to agree with you. For example… if a 3 dan player were to play a ddk player then not even 9 handicap stones will change the greater outcom of the game. (Speaking from a simmilar experience.)

However, there are sever factors to think about here i think.

  1. Just how well a player understands the game compared to how well the player plays the game.

  2. Areas of strength and weakness in play.

  3. Rank discrepensies and differences.

I think, a more accurate handy per rank difference would be closer to 1.5 stones per rank. But then as the wikia says, it matters how strong a player is and how well he will be able to utilose the power of a handy stone.

So your best bet would be to play a person in am even game, assess how well they are agaijst your own skills and then go from there and descide how many stones that person needs.

Or start with nine stones and work down. Both eays are equally as fun i think.

1 Like

I believe that a lot of it has to do with how well the handicap is utilized. A weaker player with nine stones will often obtain less advantage (in terms of %chance victory) than a stronger player with the same handicap (against an opponent equally stronger than him) due to his lack of skill.

It also has to do with rating systems. Most rating systems (especially OGS) have nothing to do with stone rankings, and the k/d rating divides are largely arbitrary at this point. We rate people using the Glicko Rating System and give each stone rank a corresponding rating conversion (I believe the scale is logarithmic to mimic how stronger players can use a stone more effective, increasing the difference of a stone rank), in a semi arbitrary manner (based more on statistical deviations than actual stones), thus not being much of an estimate.