There are various problems to that. a) For example, one stone might be worth X points between dan players, but then be worth X+Y stones between SDK players and then a lot more between DDK players because they are not as good at invading. b) Also, as @Uberdude pointed out, not all stones are equal and that also varies per level. The tengen stone or a side stone as the fifth handicap stone might be worth more or less, again depending on the level. c) Also, as you add stones and they start to interact with each other, their values change as well. e.g. four stones is a whole board ni-ren-sei, but six stones are two san-ren-sei. Again those might be worth X stones between dan players, but good luck being DDK near SDK and facing that and expecting to have a fair chance of winning.
Considering that people have been debating for many many decades about the worth of komi (just one stone), imagine the arguments that would be generated by multiple stones and their interconnections and the differences of worth between ranks.
Personally I have a small algorithm for the whole thing - obviously totally arbitrary and set for my level - that goes:
If someone is 2 ranks below, no handicap. Not only the rankings are fluid, but on a good day anyone could play much better or on a bad day play much worse. 2 ranks = acceptable fluctuation.
3-5 ranks below I give out a -16.5 komi, we try a few games and if the person is much stronger than its actual rank we can adjust.
5-8 ranks is -25.5 komi.
9+ ranks I go with -40.5 komi
The numbers are, as I said, arbitrary, but they are based on the logic of that at 3-4kyu I should have a 50/50 chance of winning with 17+ points against someone that is 7-9k (better direction of play), or by 26+ points against DDK players (maybe win a couple of fights) or 41 points against high DDK/TDK (probably by capturing some group).
I honestly canāt see how even the AI could come up with definitive negative komi numbers that would make everyone happy to apply them to their games.
One stone difference corresponds to 2 * komi. If fair komi is 7 points, then two players with 5 ranks difference should play with -14*5+7 = -63 points komi to get a balanced match. Of course to achieve this, White needs to play aggressively and try to kill a group.
How does this work? Isnāt komi, by definition, supposed to compensate for a one stone difference?
If I remember correctly, the traditional idea of someone playing against a person 1 rank below them was to give them Black and first move and remove the komi (that is a one stone handicap, if you will), so that one stone is worth the komi, not 2*komi.
So, I am a bit confused by the double komi value.
I most certainly cannot defeat a 9k with -63 points, not half the time. Not even once.
For a visual aid, 64 points means that you need to take almost 1/4 of the board from the opponent and then tie or defeat them on the rest of it as well:
No, when Black moves first we give Black one stone and then White gets a 7 point komi to create a different configuration where noone is expected to win (yes, AI says that even with 7 komi there is a very very slight advantage to White, but it like half a point iirc? Letās not get into that)
It is like having two children, an icecream and 2 euros.
Configuration one is to give the icecream to child A.
Configuration two is to give the icecream to child B.
Their difference is the icecream (one stone)
Komi now appears in the form of the 2 euros.
So, if I give one child the icecream and the other the euros to buy an icecream, then they are mostly equal, we are not back in a situation where one child got the icecream and the other got nothing.
So, yes, the whole difference is 14 points, but you split those points to make it as fair as possible. One player gets the stone, one player gets the 7 komi. Ergo the first stone = 7 komi
It is a common misunderstanding to think that komi is worth one stone, rather than half.
Hereās an explanation I wrote on reddit 5 years ago, maybe that helps you understand:
Itās a common but wrong idea that komi is worth the first move advantage: itās only half of it. To see this consider a normal game with komi and black moves first in which we agree komi makes the game fair (nevermind if the exact ācorrectā komi is 6.5 or 7.5 etc). Now imagine black passes for his first move, what change do we have to make to make the game fair again? If we do nothing white moves first and has komi, what should happen is white moves first and the other player (black now) gets komi. So we need to take komi away from white and give it to black, i.e. a swing of twice komi. So blackās pass was worth (negative) twice komi. So blackās first move is worth twice komi.
Thinking about the empty board like this is the correct way to approach the problem. Thinking about handicaps is not, because traditional handicaps are all a half stone aka komi advantage to white:
1 rank difference playing as no komi is komi aka half stone difference so half stone advantage to white given the players are actually 1 stone/rank apart
2 ranks difference playing with 2 stones and no komi is a 1.5 stones difference so half stone advantage to white
3 ranks difference playing with 3 stones and no komi is a 2.5 stones difference so half stone advantage to white
4 ranks difference playing with 4 stones and no komi is a 3.5 stones difference so half stone advantage to white
Let me give an analogue example. Coins of 10, 9,ā¦,1 euros are on a table. Two players pick coins alternatively.
The first player will get 10+8+6+4+2 euros, the second one 9+7+5+3+1 euros, so the first player wins by 5. But you will agree that the first move is worth 10 euros.
I like handicap in that it forces me to explore new styles. As a DDK given influence, I had to learn to use it. The few times I have been White in a handicapped game, I have had to learn the proper level of agression (as opposed tho my chess career, where aggression was always rewarded).
Yes, for many reasons: a) For that game weād have to adjust with -77 points, so I won with less than ten points even by killing off all those groups. b) That was a correspondence game with no handicap, so the opponent was, at some point, forced to do something about all those points. Had there been a handicap, a lighter reduction would have led to an inevitable victory for the opponent, because noone could invade solid territory second-third line corners and live, so I had no other points to gain. c) In a live game some of those groups would have probably lived.
So, in a fair game, the opponent MIGHT panic, but a large moyo, make a deep invasion and lose the fight, but give them 77 points komi from the start and theyād be defending and reducing all day long. Other people might be able to break through even that playstyle and find so many points, but I cannot.
Even worse, Iād be probably the one playing the reckless moves. Iād have 77 points to cover, you do not do that by playing tame joseki and balanced exchanges.
With 77 pts reverse komi the weaker player would play safe, but the stronger player would also complicate the game and hope that the weaker player makes more mistakes.
You might be pleasantly surprised how often people 5 stones weaker than you find creative ways to mess up you never imagined and make the reckless moves work.
