Hahn pointing system

snorri · #1

The recent Wins to Go Innovation Tournament used something called the "Hahn pointing system."

As reported in the AGA newsletter:

Quote:

There were eight rounds and the Hahn Pointing system, developed by Professor Sang-Dae Hahn of Myongji University, was used to determine tournament winners. Instead of assigning 1 or 0 for a win or loss , the Hahn system awards points from 0 to 100 based on the difference in game counts. Close games with a difference between .5 to 10 points in game counts results in the winner receiving 60 points and the loser 40 points, but a decisive win with a difference of 40 points or more results in the winner receiving all 100 points and the loser receiving nothing. This encourages fighting, aggressive games.

Thoughts?

DrStraw · #2

DGS had a tournament like this. It is basically bangneki. I joined it to see what it was like but did not enjoy it. It warps the game too much.

RobertJasiek · #3

A great advantage of go is the linearity of scores: every adjacent two scores are 1 point apart. Hahn abandons this by making 10 points steps. Another great advantage of go is the open end of maximal scores. Hahn abandons this by setting a maximum; my ability to kill tremendous groups in tremendous moyos is punished. However, the main characteristic of go is the same result regardless of the size of the score. Hahn abandons this in order to invide taking increasingly great risks; however, one is not rewarded for playing for the greatest possible score but one is punished for the risk driven to its unreasonable limit. I do not know if this year had it, but the tournament had had rewards for winning streaks in successive rounds, meaning that intentional losses could be an advantage in order to then have a higher chance of a longer winning streak. I do not know if it still allowed to miss rounds intentionally in order to win the tournament.

All aspects change normal go strategy and create a go variant game. That's why I have never participated in this tournament. I want to play go - not the worst form of bangneki.

xed_over · #4

LittleGolem also has this...
http://www.littlegolem.net/jsp/games/ga ... ?gtid=go19

MinjaeKim · #5

RobertJasiek wrote:

A great advantage of go is the linearity of scores: every adjacent two scores are 1 point apart. Hahn abandons this by making 10 points steps. Another great advantage of go is the open end of maximal scores. Hahn abandons this by setting a maximum; my ability to kill tremendous groups in tremendous moyos is punished. However, the main characteristic of go is the same result regardless of the size of the score. Hahn abandons this in order to invide taking increasingly great risks; however, one is not rewarded for playing for the greatest possible score but one is punished for the risk driven to its unreasonable limit. I do not know if this year had it, but the tournament had had rewards for winning streaks in successive rounds, meaning that intentional losses could be an advantage in order to then have a higher chance of a longer winning streak. I do not know if it still allowed to miss rounds intentionally in order to win the tournament.

All aspects change normal go strategy and create a go variant game. That's why I have never participated in this tournament. I want to play go - not the worst form of bangneki.

Professor Hahn mentioned you in his writing interestingly enough ;-)

. He did misspell your name as "Yasiek", but here it goes.

Hahn Sangdae wrote:

바둑 룰만 공부하는 약간은 이상한 친구 야시크(Robert Yasiek)도 베를린 사람이다. 내가 며칠 전 베를린에 있을 때 그가 나를 향해 걸어 온다. 나는 당연히 나에게 인사하러 오는 줄 알고 미소로 쳐다보고 있는데 내 앞에 오더니 첫 마디가 “난 당신 제도를 싫어한다(I don’t like your system)” 한다. 내가 “왜 싫어하냐?” 했더니 “바둑 원칙에 어긋난다” 한다. 내가 “바둑원칙에 대해 뭘 안다고 그러느냐? 내 대회에 출전해 봤느냐?” 하니까 “내가 그런 대회에 참가할 것 같으냐?” 하더니 다른 데로 가 버린다. 그는 나에게 “Hello!” 한 마디도 안 했다.

나는 2000년부터 그와는 세미나에서 두 번을 만난 걸 비롯 거의 매년 만나고 있고 나와는 대화도 많이 한 친구다. 내 옆에 앉아 있던 독일 바둑인들에 의해 이 얘기는 전 유럽에 퍼졌다. 야시크를 잘 아는 사람들이 이 얘기를 듣는 순간 배를 잡고 웃는다. 룽어 회장이 나에게 “그게 전형적 야시크다(That’s typical Yasiek!)” 한다.

Robert Jasiek who is a bit weirdo studying only the go rules is also a Berliner. When I was in Berlin a few days ago, he walked towards me. I supposedly thought that he's coming to greet me so I smiled, but his first word to me was "I don't like your system". I asked, "Why don't you?", and he replied, "It is contrary to the fundamentals of baduk." I said, "What do you know about the fundamentals of baduk? Have you ever participated in my tournament?". He says, "Do you think I'd participate in such tournament?", and went away. He didn't even say a single "Hello!" to me.

Since 2000 when we were in the same seminar, I met him almost every year, and I had a lot of talk with him. By the German go players sitting next to me, this story spread over Europe. People who know Jasiek well laughs out loud when they hear this. Bernhard Runge said "That's typical Jasiek!".

http://www.tygem.com/news/news/viewpage.asp?pagec=&seq=11235

Bill Spight · #6

The Hahn point system sounds something like International Match Points (IMPs) in contract bridge team events. Team events can also use total point scoring or what is called board-a-match scoring, where the team that wins a hand gets one point and the opponents get zero. In bridge the form of scoring does not particularly affect aggressiveness. For instance, at board-a-match one may take chances because at worst you only lose one point.

Each scoring method has its pluses and minuses, but the World Championships use IMPs.

Uberdude · #7

RobertJasiek wrote:

A great advantage of go is the linearity of scores: every adjacent two scores are 1 point apart. Hahn abandons this by making 10 points steps.

In normal Go the fact scores are 1 point apart all turns into the same binary win/loss as far as tournament results are concerned. So if you win by 13 points in normal Go and feel happy because 13 is your favourite number then your tournament result is a win and loses the information you won by 13. In Hahn Go if you win by 13 points you can also feel happy and then that information is also lost in the tournament scoring result of +70, but some information of the margin of victory is retained. So I didn't really understand the criticism here. Of course this splitting or results into buckets of 10 is actually an unnecessary one: you could say win by 0.5 is 60 points, 1.5 is 61, 2.5 is 62 etc all the way up to 40.5 or more is 100 points; let us call this Continuous-Hahn. I suspect they don't do this because it's a little more complex for the tournament organisers to record the results. Would RJ prefer Continuous-Hahn?

RobertJasiek wrote:

Another great advantage of go is the open end of maximal scores. Hahn abandons this by setting a maximum; my ability to kill tremendous groups in tremendous moyos is punished.

Again, I don't see how Hahn is any worse than normal Go, in fact it's better. In normal Go if you kill a big group and win by 100 points then you can feel smug, and you just get a win on the tournament result sheet, the information of your large margin of victory is lost. In Hahn Go, you can also feel smug about your 100 point win, and you get a bigger win for the tournament score. In both cases your game record can say you won by 100.

RobertJasiek wrote:

However, the main characteristic of go is the same result regardless of the size of the score.

Agreed.

RobertJasiek wrote:

Hahn abandons this in order to invide[invite] taking increasingly great risks; however, one is not rewarded for playing for the greatest possible score but one is punished for the risk driven to its unreasonable limit.

In normal Go there is no reward (other than smugness etc) for going for a big win, so it seems a bit odd to complain about Hahn rewarding you for a 40 point win but not more for a 100 point.
Yes the discontinuity and cap at 40 is inelegant and we could debate other graph shapes for point margin vs scoring result instead of stepped linear that could make it a bit nicer and taper off for large wins, but that won't change the fact you are rewarded for winning with a bigger margin so it's a different game.

RobertJasiek · #8

MinjaeKim wrote:

Professor Hahn mentioned you in his writing

[i]Robert Jasiek who is a bit weirdo studying only the go rules is also a Berliner. When I was in Berlin a few days ago, he walked towards me. I supposedly thought that he's coming to greet me so I smiled, but his first word to me was "I don't like your system". I asked, "Why don't you?", and he replied, "It is contrary to the fundamentals of baduk." I said, "What do you know about the fundamentals of baduk? Have you ever participated in my tournament?". He says, "Do you think I'd participate in such tournament?", and went away. He didn't even say a single "Hello!" to me.

If this is "typical" for anything, then for its misrepresentation of my person.

I met Sangdae Hahn at a few occasions. For the first time, in Incheon at the 1. International Conference on Baduk, where we were treated like VIPs including almost a week of sightseeing in Seoul. Concerning only the Berlin-related occasions, it was as follows:

1) A group of Koreans incl. Hahn travelled through Europe by bus. Before Prague, they noticed that they missed contact information for go players, called me (a Berliner) because they could easily find my phone number and asked me to give them contact information of go players in Prague. I do not recall if Hahn, his wife or another person called me. Anyway, I looked in the internet for half an hour. When they called me again, I gave them some suitable contact information.

2) IIRC, a few days later that group came to Berlin. I went to a go club to meet a few of them. Then suddenly Hahn explained that they wanted to have a meeting elsewhere in the city to talk about likely sponsor prize money for the then planned German KPMC preliminary. Ok, why not, I thought. So a few Berliners incl. me went to that meeting and we had related discussions. (The tournament started a few years later.)

3) Was it the same travel group? They asked for a few Berliners to accompany the Korean sightseeing travel tour through Berlin and Potsdam. I participated for a full day. Well, I tried. I like to be punctual. In Korea, I had seen that there punctual meant 25 minutes late. So I thought: If I am 5 minutes early, this is good enough even if I miss a local city transport connection. And sure enough, for the first time in years, my local bus was missing and I had to take the next one 20 minutes later. So I arrived ca. a quarter of an hour late, the bus with the Koreans had already gone, I called them, the bus came back and I entered to explain them their preferred sights.

4) A few years(?) later, another Korean travel group with Hahn in Berlin asked for sightseeing help again, but I had no time at that date.

5) Probably again a few years later, Hahn visited a tournament in Berlin (was it the KPMC preliminary?) and there the incident translated by you occurred. I do not recall the exact contents of my short talk with Hahn, but in my memory the Hahn system was mentioned by one of us, he asked "Do you like it?" and I replied "No!". Presumably I also gave a short explanation for my opinion. - I do not recall who might or might not have said "Hello". Maybe neither of us. It can be like this at go tournaments when everybody sees everybody else but does not run to everybody saying "Hello" but at some time a talk can occur nevertheless.

The reported description "studying only the go rules" is, of course, false, Hahn knows it better, and so I do not think that he would have described me so (except maybe as a joke). It reads more like an anecdote having lost connection to its source and being exaggerated into a one-sided report by a go "journalist".

Ask minue622 what he would say about my occurrence in Korean or Chinese newspapers. According to him, I would be more popular in Seoul for my ICOB speech than I am among Western players. Chinese newspapers appear to have reported my fair and neutral contribution to the International Go Rules Forum. Well, this would partly explain why I am sometimes known as a rules expert even in Asia:)

tj86430 · #9

Bill Spight wrote:

The Hahn point system sounds something like International Match Points (IMPs) in contract bridge team events. Team events can also use total point scoring or what is called board-a-match scoring, where the team that wins a hand gets one point and the opponents get zero. In bridge the form of scoring does not particularly affect aggressiveness. For instance, at board-a-match one may take chances because at worst you only lose one point.

Each scoring method has its pluses and minuses, but the World Championships use IMPs.

OTOH, I believe in pairs tournaments a system basically similar to board-a-match - Match Points - is widely used (I haven't participated in tournaments for almost 20 years, but I believe it still is the norm)

Bill Spight · **#10**

tj86430 wrote:

Bill Spight wrote:

The Hahn point system sounds something like International Match Points (IMPs) in contract bridge team events. Team events can also use total point scoring or what is called board-a-match scoring, where the team that wins a hand gets one point and the opponents get zero. In bridge the form of scoring does not particularly affect aggressiveness. For instance, at board-a-match one may take chances because at worst you only lose one point.

Each scoring method has its pluses and minuses, but the World Championships use IMPs.

OTOH, I believe in pairs tournaments a system basically similar to board-a-match - Match Points - is widely used (I haven't participated in tournaments for almost 20 years, but I believe it still is the norm)

Right.

To be clear, I was talking about team events.

Krama · **#11**

Difference Winner Loser

0.5 ........... 60 ..... 40
1.5 ........... 70 ..... 30
2.5 ........... 80 ..... 20
3.5 ........... 90 ..... 10
4.5+ (resign) 100 ..... 0

RobertJasiek · **#12**

Uberdude, what you call Continuous-Hahn, is not really continuous but goes in direction of becoming continuous and so better than Hahn.

Quote:

RobertJasiek wrote:

Another great advantage of go is the open end of maximal scores. Hahn abandons this by setting a maximum; my ability to kill tremendous groups in tremendous moyos is punished.

I don't see how Hahn is any worse than normal Go, in fact it's better.

This sentence of yours sounds like you think each aspect of Hahn was better than normal go but in the context of where you state the sentence it reads more as if you want to say only that Hahn was better WRT rewarding killing big groups. I cannot guess what you want to express, please clarify.

Quote:

your game record can say you won by 100.

The game record also includes the series of moves, from which tactics and strategy can be derived. Therefore I do not understand why you mention the game record. Maybe you do not mean the game record but only the game result (statement of the outcome and possibly statement of the score)?

Quote:

RobertJasiek wrote:

Hahn abandons this in order to invide[invite] taking increasingly great risks; however, one is not rewarded for playing for the greatest possible score but one is punished for the risk driven to its unreasonable limit.

In normal Go there is no reward (other than smugness etc) for going for a big win, so it seems a bit odd to complain about Hahn rewarding you for a 40 point win but not more for a 100 point.

I say different things about

1) normal go

2) Hahn

because, for Hahn, the presupposition of allowing varying scores having varying impact on the outcome is introduced at all. Hahn-with-maximum and Hahn-without-maxmimum both rely on the presupposition. So when I express different opnions on Hahn-with-maximum and Hahn-without-maxmimum, this my opinion does not affect normal go.

schawipp · **#13**

As a side note, in the recent Berlin "Go to innovation" tournament there was no 10 point truncation - e.g. if you won a game by 13 points you got +73 points in tournament score and not 70 or 80. One effect of this system is that more games are played till the end. Even if you are behind by about 10-20 points, you will still get some "score" (40 - point difference). This has the further side effect that a higher fraction of games could be reversed by late blunders...

Bill Spight · **#14**

Something like this might work all right.

Code:

Winning margin   Winner Points   Loser Points

5 -  1.5         12                8
5 -  4.5         13                7
5 -  9.5         14                6
5 - 18.5        15                5
5 - 35.5        16                4
5 - 68.5        17                3
5 +             18                2

Uberdude · **#15**

RobertJasiek wrote:

Uberdude, what you call Continuous-Hahn, is not really continuous but goes in direction of becoming continuous and so better than Hahn.

I know, I did not claim it to be continuous in the mathematical function sense, but it is as continuous as you can get with Go's integer spaced scores (with possible half offset from komi). And I'm glossing over the possibility(?) of bizarre end game situations like that bulky five under the stones uncaptured thingy in the corner that had some ruling over it and making one of those with a non-integer count.

RobertJasiek wrote:

Quote:

RobertJasiek wrote:

Another great advantage of go is the open end of maximal scores. Hahn abandons this by setting a maximum; my ability to kill tremendous groups in tremendous moyos is punished.

I don't see how Hahn is any worse than normal Go, in fact it's better.

This sentence of yours sounds like you think each aspect of Hahn was better than normal go but in the context of where you state the sentence it reads more as if you want to say only that Hahn was better WRT rewarding killing big groups. I cannot guess what you want to express, please clarify.

Each point I make is in respect of your quoted passage above. I think Hahn Go is an inferior game to normal Go, though I haven't played in a tournament with it so am open to changing my mind, and indeed being an inferior games doesn't mean some tournaments with it is a bad idea; I have enjoyed watching Inseong Hwang give 5 stones to a 4d and win in the Go to Innovation for example.

RobertJasiek wrote:

Quote:

your game record can say you won by 100.

The game record also includes the series of moves, from which tactics and strategy can be derived. Therefore I do not understand why you mention the game record. Maybe you do not mean the game record but only the game result (statement of the outcome and possibly statement of the score)?

By game record I mean some data storage mechanism by which you record what happened in the game with more details than that which is retained for the result for the tournament. This could be traditional grid kifu paper, some software, your head or whatever. You could record the moves, or just player names and score. My point is in normal Go that you won by 13 points is lost to the tournament result, so complaining that Hahn is inferior to normal Go because it loses the fact you won by 13 points makes no sense. In both cases you can record the fact you won by 13 points in whichever manner you choose, accompanied by the moves or the weather on the day or the trouser colour of your opponent.

RobertJasiek wrote:

Quote:

RobertJasiek wrote:

Hahn abandons this in order to invide[invite] taking increasingly great risks; however, one is not rewarded for playing for the greatest possible score but one is punished for the risk driven to its unreasonable limit.

In normal Go there is no reward (other than smugness etc) for going for a big win, so it seems a bit odd to complain about Hahn rewarding you for a 40 point win but not more for a 100 point.

I say different things about

1) normal go

2) Hahn

because, for Hahn, the presupposition of allowing varying scores having varying impact on the outcome is introduced at all. Hahn-with-maximum and Hahn-without-maxmimum both rely on the presupposition. So when I express different opnions on Hahn-with-maximum and Hahn-without-maxmimum, this my opinion does not affect normal go.

In that case your criticism of Hahn versus normal Go as below is illogical, it should be a criticism of Hahn as he proposed versus a better implemented variant of Go in which the winning margin on the board is reflected in the tournament score for the game, as normal Go is inferior rather than superior to Hahn in those two respects.

RobertJasiek wrote:

A great advantage of go is the linearity of scores: every adjacent two scores are 1 point apart. Hahn abandons this by making 10 points steps. Another great advantage of go is the open end of maximal scores. Hahn abandons this by setting a maximum; my ability to kill tremendous groups in tremendous moyos is punished.

Knotwilg · **#16**

A pro who knows that he'll lose by a few points will go all out at a crucial stage, possibly losing a group. In a system where close losses are rewarded more than big losses, the incentive will be much smaller. So I'd think that the system leads to less spectacular games, not more.

DrStraw · **#17**

Knotwilg wrote:

A pro who knows that he'll lose by a few points will go all out at a crucial stage, possibly losing a group. In a system where close losses are rewarded more than big losses, the incentive will be much smaller. So I'd think that the system leads to less spectacular games, not more.

That is one of my objections to this perversion. To have a guaranteed loss by a small margin to be preferable to going all out for a win is just anathema to my way of thinking. Imagine the situation of needing needing a birdie on the 18th hole of the US Open to win but deciding that the risk is too great so you settle for a bogie to guarantee second place. I don't think so.

Bill Spight · **#18**

DrStraw wrote:

Knotwilg wrote:

A pro who knows that he'll lose by a few points will go all out at a crucial stage, possibly losing a group. In a system where close losses are rewarded more than big losses, the incentive will be much smaller. So I'd think that the system leads to less spectacular games, not more.

That is one of my objections to this perversion. To have a guaranteed loss by a small margin to be preferable to going all out for a win is just anathema to my way of thinking. Imagine the situation of needing needing a birdie on the 18th hole of the US Open to win but deciding that the risk is too great so you settle for a bogie to guarantee second place. I don't think so.

In the system that I lay out above, suppose that someone faces a small loss of 3.5 points of territory and risks around 15 points of territory to gain around 5 points of territory. In converted points that is a risk of 2-3 points to gain 5-6 points: a good bet.

mitsun · **#19**

DrStraw wrote:

Imagine the situation of needing needing a birdie on the 18th hole of the US Open to win but deciding that the risk is too great so you settle for a bogie to guarantee second place. I don't think so.

What about the situation where a bogie on the final hole wins the tournament (or preserves a lead, in an earlier round)? Would any professional go for the birdie, if it carries more risk of ultimate loss? Golf sounds rather analogous to Hahn Go to me. Each hole has a margin of victory, not a simple win/loss result, and the final winner is decided by the sum of the marginal scores.

Hahn pointing system

Who is online