Which scoring method?

[amnal]

Still though, it doesn't seem like deciding life and death is that difficult, though I understand that it's probably much more difficult in pro (and even high amateur) games and that there needs to be a standardized way to decide these things for tournament purposes - right?

It is usually not difficult to ascertain the status of a settled life and death position, regardless of the level of the game. If a player disputes its status, generally there isn't a problem as this player can dispute it by trying to kill their opponent!

I suppose the biggest (maybe only?) problem is that there are some amusing positions where one player would lose points by proving it (generally this occurs in territory scoring but not area scoring). There is one relatively common example where 'the bent four in the corner is dead'. Confusingly, this doesn't refer to the bent four eyeshape, but to shapes like the following:

$$c
$$ -------
$$ | O . X O .
$$ | O X X O .
$$ | O X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .

Click Here To Show Diagram Code

[go]$$c
$$ -------
$$ | O . X O .
$$ | O X X O .
$$ | O X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .[/go]

In this shape (assuming there are no other moves left to play on the board), can you work out the status of the group? In case it isn't obvious, this is something of a trick question, and is quite a hard problem because of this.

I won't post the answer here, because it's well analysed at http://senseis.xmp.net/?BentFourInTheCornerIsDead . It is a good example of a relatively common shape where the specifics of a ruleset can make a big difference in a life and death situation. However, it's worth emphasising that it's really not worth worrying about this, it's no kind of problem, and you're fine using whatever ruleset you find comfortable. Area scoring has some advantages, but I've personally never seen it used except to show someone how it works. I like AGA rules for the way they neatly allow you to count whichever way you prefer, but use verbal Japanese scoring in any non-tournament game. Counting territory is just the way almost everyone does it here (in the UK).

[RobertJasiek]

it doesn't seem like deciding life and death is that difficult

It can be easy or difficult in every game. This is secondary for correct rules application. Primary is whether you always determine / perceive the life and death statuses correctly.

As for the couple of Go books I have, they do use territory scoring and they do explain how to create two-eye formations.

Really? The beginner books I have seen mostly showed how a group with minimal life looks like but did not explain that every independent life could be transformed to a two-eye-formation, how to do that and how to consider several variations. Have the recent beginner books finally become better as a consequence of my efforts to explain two-eye-formation?

I can't say that what I've learned is the same as the rules you posted above,

I posted simplified rules. You don't want to know the official Japanese rules, don't you? I have linked only a commentary on them (but of course you COULD find all on my webpage;) )

but the books are Korean - perhaps this is a difference,

There are differences. See my webpage if you can't resist.

http://home.snafu.de/jasiek/rules.html

Are the life and death rules different then?

Their fine print, yes.

A question - are you trying to say that I should switch to area scoring?

Sure. There is a minor obstacle though: Go is a 2-player game and your opponent needs to agree:)

Anyways, I'm not trying to focus on the rules too much...

Right. Rules study prevents from becoming stronger because one lacks time to study strategy and tactics:)

I simply want to know that I'm scoring correctly.

If you use territory scoring, there is no simple answer. You would need to provide games or sample positions, I guess, and hope that somebody takes time for checking.

[RobertJasiek]

I suppose the biggest (maybe only?) problem is that there are some amusing positions

The biggest problem is the rather frequent shapes with fake obvious status:

$$
$$ -------
$$ | . . X O .
$$ | . . X O .
$$ | . . X O .
$$ | X X X O .
$$ | O O O O .

Click Here To Show Diagram Code

[go]$$
$$ -------
$$ | . . X O .
$$ | . . X O .
$$ | . . X O .
$$ | X X X O .
$$ | O O O O .[/go]

In a Go introduction in German TV ("Sendung mit der Maus"), it was taught wrongly...

[HermanHiddema]

For a beginner it matters very much because of the extremely different difficulties of rules understanding.

The evidence suggests otherwise.

The evidence is that still nobody could show me any beginner with a reasonable understanding of territory scoring rules. Therefore the evidence does not suggest otherwise. In particular, beginners tend to overlook simple facts such that filling liberties for final removals is a mistake. Almost all beginners are having great difficulties with reading more than one move deep or with the idea of playing inside an eye; this is the contrary to having an ability to distinguish life from death. Beginners reading the wrong introductions don't even know that sekis exist. Etc.

I've taught dozens of beginners, and they've all been able to grasp the rules and have gone on to become happy, active go players. Furthermore, millions upon millions of people in Japan and Korea have managed to successfully learn the rules and play the game for hundreds of years.

@scutheotaku: I wouldn't worry too much about the differences between Territory and Area scoring, they are pretty trivial. Stick to the rules that the people around you play, because ultimately, the purpose of learning this game is to have fun. It is better to play against other people, than to get into arguments with them about rules trivia.

[RobertJasiek]

I've taught dozens of beginners, and they've all been able to grasp the rules and have gone on to become happy, active go players.

Right. I forgot that lucky beginners having a good personal teacher (and even with solid rules knowledge) at all have a much better chance.

Furthermore, millions upon millions of people in Japan and Korea have managed to successfully learn the rules and play the game for hundreds of years.

Those mythical millions of millions have not understood the rules well. As I know from many talks, only very few actually do understand the rules reasonably. It is more like they have understood life and death than the rules. So what they have is rather an implicit rough approximation of what the rules' effect means in ordinary cases. If they were asked to write down the rules, mostly they would create such nonsense as J1949 or WAGC79.

[HermanHiddema]

Furthermore, millions upon millions of people in Japan and Korea have managed to successfully learn the rules and play the game for hundreds of years.

Those mythical millions of millions have not understood the rules well. As I know from many talks, only very few actually do understand the rules reasonably. It is more like they have understood life and death than the rules. So what they have is rather an implicit rough approximation of what the rules' effect means in ordinary cases. If they were asked to write down the rules, mostly they would create such nonsense as J1949 or WAGC79.

If they understand the rules well enough to play, they understand the rules.

In a similar vein:

I understand the rules of football (soccer), but I have never even looked at a referee's handbook. There may be dozens of obscure edge cases that I have never heard of.

Numerous people understand computer programming, but if you ask them a question about, say, the pump lemma, they'll have no idea what you are talking about.

There is a difference between "understand" and "have studied in depth every possible edge case".

[scutheotaku]

Still though, it doesn't seem like deciding life and death is that difficult, though I understand that it's probably much more difficult in pro (and even high amateur) games and that there needs to be a standardized way to decide these things for tournament purposes - right?

It is usually not difficult to ascertain the status of a settled life and death position, regardless of the level of the game. If a player disputes its status, generally there isn't a problem as this player can dispute it by trying to kill their opponent!

I suppose the biggest (maybe only?) problem is that there are some amusing positions where one player would lose points by proving it (generally this occurs in territory scoring but not area scoring). There is one relatively common example where 'the bent four in the corner is dead'. Confusingly, this doesn't refer to the bent four eyeshape, but to shapes like the following:

$$c
$$ -------
$$ | O . X O .
$$ | O X X O .
$$ | O X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .

Click Here To Show Diagram Code

[go]$$c
$$ -------
$$ | O . X O .
$$ | O X X O .
$$ | O X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .[/go]

In this shape (assuming there are no other moves left to play on the board), can you work out the status of the group? In case it isn't obvious, this is something of a trick question, and is quite a hard problem because of this.

I won't post the answer here, because it's well analysed at http://senseis.xmp.net/?BentFourInTheCornerIsDead . It is a good example of a relatively common shape where the specifics of a ruleset can make a big difference in a life and death situation. However, it's worth emphasising that it's really not worth worrying about this, it's no kind of problem, and you're fine using whatever ruleset you find comfortable. Area scoring has some advantages, but I've personally never seen it used except to show someone how it works. I like AGA rules for the way they neatly allow you to count whichever way you prefer, but use verbal Japanese scoring in any non-tournament game. Counting territory is just the way almost everyone does it here (in the UK).

Thanks, that really cleared some things up

it doesn't seem like deciding life and death is that difficult

It can be easy or difficult in every game. This is secondary for correct rules application. Primary is whether you always determine / perceive the life and death statuses correctly.

As for the couple of Go books I have, they do use territory scoring and they do explain how to create two-eye formations.

Really? The beginner books I have seen mostly showed how a group with minimal life looks like but did not explain that every independent life could be transformed to a two-eye-formation, how to do that and how to consider several variations. Have the recent beginner books finally become better as a consequence of my efforts to explain two-eye-formation?

I can't say that what I've learned is the same as the rules you posted above,

I posted simplified rules. You don't want to know the official Japanese rules, don't you? I have linked only a commentary on them (but of course you COULD find all on my webpage;) )

but the books are Korean - perhaps this is a difference,

There are differences. See my webpage if you can't resist.

http://home.snafu.de/jasiek/rules.html

Are the life and death rules different then?

Their fine print, yes.

A question - are you trying to say that I should switch to area scoring?

Sure. There is a minor obstacle though: Go is a 2-player game and your opponent needs to agree:)

Anyways, I'm not trying to focus on the rules too much...

Right. Rules study prevents from becoming stronger because one lacks time to study strategy and tactics:)

I simply want to know that I'm scoring correctly.

If you use territory scoring, there is no simple answer. You would need to provide games or sample positions, I guess, and hope that somebody takes time for checking.

Yeah, the book series I have explains two-eye formations pretty well, I think. While I'm not sure how reknowned they are in the Go world, while looking for a good Go book I found that Janice Kim and Jeong Soo-hyun's Learn to Play Go series (the books I'm referring to) are the ones that seem to get recommeneded the most. As far as how new they are, the original printing was 1996, second edition is 1998.

I know what you mean on bad instruction books though. When I first picked up Go, I bought one of those set with the half-size (though still 19x19) board and the little, impossible-to-pick-up plastic stones. The set's Go manual was my introduction to the game, and well... Let's just say that it was pretty bad... It used territory scoring too, but it barely even mentioned life and death and certainly didn't explain it. In the end, the book had somehow led me to believe that stones were only alive if, and only if, they had some sort of connection (whether by open territory or by other stones of the same color) to the edge of the board... I thought this was the way to play for a couple of weeks until I picked up the Learn to Play Go series.

It seems like the full-length Japanese rules (not that I've read them) is mainly so long because of the technicalities (e.g. for resolving disputes). If you could please outline the general differences between the simplified Japanse rules you posted and the far-longer officila rules - or are the differences more minor and/or based on specific issues?

I'd rather not compare Korean and Japanese rules looking for differences, but I appreciate the link - nice website!

At the moment, all of the players I play in person are the people I have taught - so if I switch to area scoring, then I don't see them not.

I agree on rules study - I am focusing almost entirely on strategy and tactics, but I think it's important to an extent to get the concepts of the scoring method being used. Maybe I'm wrong though, I'm far from an expert at the game.

Thanks again for your lengthy responses!

[amnal]

I agree on rules study - I am focusing almost entirely on strategy and tactics, but I think it's important to an extent to get the concepts of the scoring method being used. Maybe I'm wrong though, I'm far from an expert at the game.

I'm not sure what you really mean here in terms of how much you focus on what, but it seems a little dubious. You can play perfectly good go (including all the subtlety and ideas of any high level go game) with nothing more than the concept that at the end of the game you must have more stones on the board than your opponent (well...plus that you take turns to play, have the capturing rule, and a simple ko rule). The idea of territory follows on this when players realise that they don't need to fill in some areas because their opponent can be killed if they play there themselves, so 'territory' is invented to save time.

Whilst there is a little more than that to the normal rulesets, it's all you need to play the game, and studying the ruleset in more detail seems of very minor value for improvement itself. That is to say, rules investigation may be interesting on its own, but it's something almost completely separate to strategy and tactics.

[scutheotaku]

I agree on rules study - I am focusing almost entirely on strategy and tactics, but I think it's important to an extent to get the concepts of the scoring method being used. Maybe I'm wrong though, I'm far from an expert at the game.

I'm not sure what you really mean here in terms of how much you focus on what, but it seems a little dubious. You can play perfectly good go (including all the subtlety and ideas of any high level go game) with nothing more than the concept that at the end of the game you must have more stones on the board than your opponent (well...plus that you take turns to play, have the capturing rule, and a simple ko rule). The idea of territory follows on this when players realise that they don't need to fill in some areas because their opponent can be killed if they play there themselves, so 'territory' is invented to save time.

Whilst there is a little more than that to the normal rulesets, it's all you need to play the game, and studying the ruleset in more detail seems of very minor value for improvement itself. That is to say, rules investigation may be interesting on its own, but it's something almost completely separate to strategy and tactics.

Yeah, I guess you are right! To be honest, I'm not really sure what I was thinking

Still though, it doesn't seem like deciding life and death is that difficult, though I understand that it's probably much more difficult in pro (and even high amateur) games and that there needs to be a standardized way to decide these things for tournament purposes - right?

It is usually not difficult to ascertain the status of a settled life and death position, regardless of the level of the game. If a player disputes its status, generally there isn't a problem as this player can dispute it by trying to kill their opponent!

I suppose the biggest (maybe only?) problem is that there are some amusing positions where one player would lose points by proving it (generally this occurs in territory scoring but not area scoring). There is one relatively common example where 'the bent four in the corner is dead'. Confusingly, this doesn't refer to the bent four eyeshape, but to shapes like the following:

$$c
$$ -------
$$ | O . X O .
$$ | O X X O .
$$ | O X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .

Click Here To Show Diagram Code

[go]$$c
$$ -------
$$ | O . X O .
$$ | O X X O .
$$ | O X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .[/go]

In this shape (assuming there are no other moves left to play on the board), can you work out the status of the group? In case it isn't obvious, this is something of a trick question, and is quite a hard problem because of this.

Looking at this again, it doesn't really seem to me that a problem like this would be hard to decipher. Right away I was able to see that black was dead - or is this wrong?

[RobertJasiek]

If they understand the rules well enough to play, they understand the rules.

Playing is trivial: Alternation, removal of libertyless stones, no immediate ko recapture. Rather understanding the rules well enough means to understand the reasoning behind scoring well enough.

[RobertJasiek]

Yeah, the book series I have explains two-eye formations pretty well

Do they explain how to decide between variations suggesting life and variations suggesting death, when trying the assess one particular string or group?

I know what you mean on bad instruction books though. [...] to the edge of the board...

Similar to my (worse) experience.

It seems like the full-length Japanese rules (not that I've read them) is mainly so long because of the technicalities (e.g. for resolving disputes).

The official Japanese rules are not particularly long, although they contain much superfluous / flawed stuff. Their much greater problem is their gaps concerning the core of the scoring description.

If you could please outline the general differences between the simplified Japanse rules you posted and the far-longer officila rules

The major difference is: While the Simplified Japanese Rules use one status analysis move-sequence for all strings on the board together, the official Japanese Rules rely on (but fail to mention and explain) arbitrarily many move-sequences combined by arbitrarily many strategies for each string on the board.

- or are the differences more minor and/or based on specific issues?

There are differences of all degrees from trivial or specific to general.

I think it's important to an extent to get the concepts of the scoring method being used.

For that, my commentary on Verbal-Japanese Rules scoring is the best because it concentrates on the decision making underlying the territory scoring. You can ignore the special positions there.

[RobertJasiek]

black was dead - or is this wrong?

The more interesting discussion is: Why is Black dead according to Japanese rules? The first step of the answer is: The reason depends on which Japanese ruleset is used. (You don't want to know the details, I assure you;) )

[amnal]

$$c
$$ -------
$$ | O . X O .
$$ | O X X O .
$$ | O X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .

Click Here To Show Diagram Code

[go]$$c
$$ -------
$$ | O . X O .
$$ | O X X O .
$$ | O X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .[/go]

Looking at this again, it doesn't really seem to me that a problem like this would be hard to decipher. Right away I was able to see that black was dead - or is this wrong?

You are correct, but I am very impressed if you also got the reason right.

$$Wc
$$ -------
$$ | O 1 X O .
$$ | O X X O .
$$ | O X O O .
$$ | 2 X O , .
$$ | X X O . .
$$ | O O O . .

Click Here To Show Diagram Code

[go]$$Wc
$$ -------
$$ | O 1 X O .
$$ | O X X O .
$$ | O X O O .
$$ | 2 X O , .
$$ | X X O . .
$$ | O O O . .[/go]

$$Wc
$$ -------
$$ | 4 5 X O .
$$ | 3 X X O .
$$ | . X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .

Click Here To Show Diagram Code

[go]$$Wc
$$ -------
$$ | 4 5 X O .
$$ | 3 X X O .
$$ | . X O O .
$$ | . X O , .
$$ | X X O . .
$$ | O O O . .[/go]

If white wants to take black's stones off the board and prove that black is dead, he has to do it like this, which makes a ko. But in area scoring rules, he has a way to kill black, and in japanese rules black is simply defined as dead (that is, the rules specify 'in this shape black is dead'). The problem is whether you got this reason right.

[Redundant]

Amnal's last diagram, fixed.

$$Wc
$$ -------
$$ | 4 5 X O .
$$ | 3 X X O .
$$ | . X O O .
$$ | X X O , .
$$ | X X O . .
$$ | O O O . .

Click Here To Show Diagram Code

[go]$$Wc
$$ -------
$$ | 4 5 X O .
$$ | 3 X X O .
$$ | . X O O .
$$ | X X O , .
$$ | X X O . .
$$ | O O O . .[/go]

[RobertJasiek]

in japanese rules black is simply defined as dead

In the Japanese 1949 Rules, the World Amateur Go Championship 1979 / 1980 Rules and typical verbal Japanese rules. In the Japanese 1989 Rules, there is no such special shape definition in the rules but other concepts.