New Zealand Rules / Discussion

Sub-page of NewZealandRules

These are particularly simple rules, although the very similar Tromp-Taylor rules are even more concise.

The New Zealand rules use area scoring and situational superko, allow suicide, and specify a komi of 7.

The full text[1] of the rules is given in New Zealand Rules and an explantion is given in /Explanation.

Table of contents


Sandra: Hey, wait a minute… it took me years to think of this, but… the ruleset doesn't define "points" and it uses it in two different ways ("intersection" and "score unit") and it doesn't specify that a player's territory counts as that players points. It just says "The winner is the player with the most points at the end of the game." It says that White adds 7 points to the territory, which, generously, might be read in reverse and give White a bunch of points, but Black doesn't have any points.

Instead, I'd rewrite it to this:

Go is played on a board with a grid of 19 × 19 intersections, by two players, one using a set of (about) 180 white counters, the other a set of (about) 181 black counters. These counters are called stones.

Adjacent intersections are those intersections connected by lines of the grid, with no intervening intersections.

Two stones of the same colour are connected if they are on adjacent intersections or if they are both connected to a third stone.

A liberty of a stone is an unoccupied intersection adjacent to that stone or to any stone connected to that stone.

Territory of a player (at the end of the game) consists of all intersections occupied by that player's stones plus all unoccupied intersections adjacent to that territory.

A play consists of placing a stone (of that player's own colour) on an unoccupied intersection, then removing any of the opponent's stones that then have no liberties (if any), and then removing any of that player's own stones that then have no liberties (if any).

A move consists of

  1. making a play so that the resulting board position does not repeat the whole board position as it was after any of that player's previous moves or
  2. saying 'pass'.

A game of go begins with an empty board, and the players take turns to move (beginning with black).

The game is finished when both players agree that there are no more worthwhile moves. 'Dead' stones may then be removed from the board by mutual agreement. If they cannot agree which stones are dead they must play on. If they cannot then agree who shall move next, all stones stay on the board (are alive) and are counted.

In an even game, 7 points are added to the white players territory.

In a handicap game, white passes the first n - 1 moves where n is the size of the handicap.

The winner is the player with the biggest territory at the end of the game.

Or something like that. If I can have a positive impact on this elegant ruleset I can retire happy♥ That'd be the way I'd want to be remembered♥

You're right, those 'points' are rather ambiguous. And I would define 'Territory' as a number ("Territory is the number of intersections that...") instead of a set.


[1]Hikaru79: The full text? Don't New Zealand rules also have rules for ko, handicap placement, points in seki etc? Also, isn't NZ rules one of the few that allow suicides? I think a more comprehensive and complete rules text for New Zealand rules can be found at : [ext] http://homepages.ihug.co.nz/~barryp/rules.htm, although this link is more of an introduction to the game through New Zealand rules than a ruleset. (Thanks to Michael Goetze for the link.)

blubb: The main page lists the (core) rules completely but skips the subsequent section titled "Explanation of Rules" (see above link to the official text).


Unicyclist The rules need to be read fairly carefully, as they are rather terse. Yes, that is the full text of the rules, although Barry's page goes on to add a more reader-friendly explanation. But that explanation is not officially part of the rules.

barry: The split between rules and explanation is not quite that strong. The rulings in the explanation should also be treated as official.

Ko is handled by the restriction of not repeating a previous position.

Handicap placement is handled by white passing a number of times.

Suicide is handled under play above, i.e. removing your own stones that have no liberties (self-capture) after any capture of opponent's stone(s).

Seki is nothing special - just count the stones and the territory they surround. Note that any dame points not surround by a single colour are not counted.

Well, if I read the rules carefully I think that dame points count for both sides ;-) The rules have no need to define surrounded by a single colour. The recursive definition of territory seems to be enough. Nice rules.

Regarding counting, the general practice here is to count the board as it stands (after removing dead stones). Not moving the stones around means a recount can be done easily without dispute. Firstly, we guess who has the smaller amount to count, then count it off (2, 4, 6, etc...). Since the board has 361 points, 181 wins and 180 loses in handicaps, and in even games black 184 / white 177 is a tie and any more is a win.

Bill: Well, if there is a seki, then there are more than 361 points on the board. :-)


Ah, my mistake! You are right, if the rules are read carefully enough, they all come together! =) What an interesting way for the rules to be written-- one page completely explaining what rec.games.go took dozens of pages to say! I apologize. ^^;

Andrew Grant: Two questions:

How do NZ rules allow winning by resignation? I can't find anything in the rules (as opposed to commentaries, which are not part of the rules) that would imply a right to end the game by resignation.

barry: The rules could have explicitly allowed resignation, but that (and other things - eg how to handle illegal moves) more correctly belong to the explanation section. As above the explanation section is also official, but contains things that are more about conduct.

In the bit about settling disputes over dead stones, what does the statement "If they cannot then agree who shall move next, all stones stay on the board (are alive) and are counted" actually mean? All the stones, including those dead stones whose death is not in dispute? Or just all the stones that are the subject of the dispute?

barry: It is questionable whether this belongs in the rules section or the explanation. However ending the game by agreement is an accepted part of the game and the rules should cover it. This means that the possibility of resumption is necessary. It is a fact of life that agreement is necessary in go. There are 2 reasons for resuming. 1. the players cannot agree on life/death. 2. Both players have made a mistake and there is still a gainful move.

If 1) it doesn't matter who plays first and it is traditional to allow the "defender" first move. If 2) there is no good reason for allowing either player to go first. They either agree who plays first or leave the move unplayed and count normally or count without removing any stones.

If they count without removing any stones one player will usually be disadvantaged. that player will probably agree to allow their opponent first move.

mgoetze: They don't explicitly allow resignation. But in a casual game, if someone says, I resign, are you going to tell them, hey now, the rules don't allow that? As for a tournament setting, it is up to the tournament rules to recognize wins by resignation - or not, if the tournament happens to have some really exotic system based on margin of victory or something.

Andrew Grant: Well no, if my opponent resigns I'm not going to quibble. All the same, it seems odd to leave such a common event as resignation to the whim of a tournament organiser. Are you sure this isn't just an oversight on the part of the NZ Go Society? I've never heard of a tournament using the margin of victory for anything, and would not wish to encourage something like that which goes against the spirit of the game. A win by 1 point is a win. It is not inferior to a win by 100 points.

Neil: I don't see how resignation is within the scope of the rules. When you resign, you're saying that you do not wish to continue the game and want to give your opponent the credit for the win without playing it out. The rules can only cover the continuation of the game, not what happens when you stop playing and interact with outside things like tournaments and ratings.

Andrew Grant: By that argument you can't have rules about scoring or removing dead stones either, since this takes place after the game has been declared over.

Neil: How does that follow? Scoring is in the rules. Giving ratings and tournament points for incomplete games are not.

Andrew Grant: I said nothing about giving ratings or tournament points. My only point is that a resignation, if it occurs, should be considered as part of the game and therefore subject to the rules. (Incidentally, both the Chinese and Japanese rulesets explicitly allow a game to end by resignation, so I'm not alone in thinking this.) You said "The rules can only cover the continuation of the game, not what happens when you stop playing..." Well, you stop playing when you agree the game to be over; by your argument anything that comes after (including scoring) should not be in the rules.


As for the disputes, a literal reading of the rules would indeed suggest that all stones are counted as alive when the players cannot agree on whose turn it is, even those which are not disputed. It would make sense to allow those stones which are undisputedly dead to be counted as dead - but - it doesn't make sense for the players not to agree on whose turn it should be in the first place! At least not if the game really was over when they both agreed it was... ;)

DougRidgway: Play resumes after a dispute. The players will then play to explicitly remove all dead stones. Everything left on the board after that is defined as alive (unless the players agree). If it's not dead when your opponent goes first, it's not dead, so you should be happy to let your opponent go first in the resumption. I think the stuff about handling when the players can't agree on who is to play next is to deal with situations where the players realize that there are points that they missed before passing. Both players voluntarily gave up their right to move, so the rules should favor neither. It only matters when the points decide the game. This means of handling this situation seems arbitrary, but is symmetrical and therefore fair.

Andrew Grant: If that is really what it means, there seems to be a problem. The rule implies that this "all stones are alive" decision only applies if the dispute is over the life & death status of stones - in which case disagreement over who plays first would not occur, the player who thinks his stones are alive will be happy to let his opponent go first.

If, on the other hand, the dispute is because a territorial boundary somewhere is incomplete, my reading of the rules suggests that the play can not resume and all the vacant points in the incomplete territory are dame.

Bill: Wait till Robert Jasiek moves to New Zealand. ;-) Then you'll find that you need to spell out the final dispute resolution procedure more clearly.

Robert Jasiek: That has been established since Barry Phease and I discussed that on rec.games.go...

DougRidgway: Maybe I'm being dense, but I don't see the problem. A dispute is a dispute, isn't it? If I want to resume, I could claim a life-death dispute, and let my opponent go first. It's only a problem if who goes first affects the outcome. In J1989, both players would lose in this case. An arbitrary outcome doesn't seem unreasonable to me.

Warp: As far as I know it's not unseen that after both players have passed one of the players spots some weakness somewhere inside his opponent's groups (that is, he could play some tesuji which captures some stones or even perhaps kills a group) which both had missed during the game. Naturally it's the interest of the owner of the weakness to end the game as it is (specially if using territory scoring) but the other would want to continue the game in order to exploit that weakness and get profit (specially if he can win the game thanks to it).

I had the impression that in the Japanese rules the one who wants to continue cannot start but the other player gets to play first. This way he can (potentially) spot the weakness and defend it. (If there are no weaknesses he can simply pass, but the important thing is that it is him who gets to move first). I think this is very fair.

As I read, with the NZ rules the players must agree who gets to move first. This doesn't sound too fair because in this case both players would want to start (the owner of the weakness to defend it and not lose anything, the other player naturally to get profit from it). If they can't agree, then it seems that no dead stone removal is done. This means that if there's even one single dead stone inside a huge territory, the huge territory becomes basically a seki and the owner of the territory gets no points from it. Doesn't sound fair at all.

barry: Of course opinions about what is fair will vary. Some people might say that the attacker has to tell the defender there is a problem under Japanese rules. The short answer is that the player who has neglected a defensive move has made a mistake. If they suffer because of that mistake then it is not the fault of the rules.

Bill: Haven't both players erred by failing to make a gainful move before ending play? In that case, what impartial rule isn't fair?

I like the Japanese idea that either player may request a resumption of play, which must be granted, but the opponent plays first. (I do not like their idea of mutual loss. If nobody requests resumption, why not score the board as is?)

blubb: Unlike japanese rules, NZ rules don't define any group as alive or dead, and I like it that way. As far as I can see, the phrase "If they cannot then agree who shall move next, all stones stay on the board (are alive) and are counted." shows - as a very basic scoring rule - that scoring does not depend on agreement. If players agree about all points of the board immediately after passing twice, that's nice, and scoring can be done. If they don't, playing is continued upto the next two consecutive passes, and the former is applied again. Only if they don't play any further moves although disagreeing, that is, they "continue the game by passing", the game is counted as it stands. I suppose that's a good "worst case rule", being a rather good reason to "continue by moving".

However, as that paragraph is worded currently, I can't see a clear answer to the following question. Consider e. g. a position where dumb scoring with all stones regarded as alive would give jigo, but both players believe they should win, according to their differing L/D judgement. What to do if both sides want the other to move first? In such a case, giving in would entail a disadvantage. Because disagreeing is something mutual, you can't say "w is the one who disagrees, hence b moves first", either. See also my comment[1743] below.

rocky?: Another question is, now that some dead stones are removed, is the board position different enough to revisit some kos? I hate the concept of "dead" stones. All stones are equal. If you don't want them to score points in your areas you need to KILL them. ;) As far as I can tell, the main appeal of the zombie concept was not laziness, but the loss of territory to stones needed to kill off invaders. Area scoring solves that problem. The "clutter" is worth as much as the empty space. There is no reason not to kill off enemy stones in areas you control.

blubb: Yeah, I think, too, that could be ruled more clearly. Due to possible interactions accross the board, continuing with SOME captives already removed may give a bogus outcome. Therefore, NO prisoners should be removed unless the opponents agree about ALL of them.

For the same reason, I find the explanation "Where there is disagreement about who should have the first move in such a situation it is usual to allow the person whose stones are in dispute to have the first move to defend them." a bit awkward. In cases of disputes with stones of both colors involved, L/D treatment gets either arbitrary or localized thereby, which needlessly re-introduces one of the main issues with Japanese rules. [1743] This could be avoided by strictly preserving the original move order (w204, b205, wPass, bPass, w206, b207, ...) whenever the play continues.


Possible Exploit/Loophole in the NZ agreement phase handling

The rules in question:
The game is finished when both players agree that there are no more worthwhile moves.
'Dead' stones may then be removed from the board by mutual agreement.
If they cannot agree which stones are dead they must play on.
If they cannot then agree who shall move next, all stones stay on the board (are alive) and are counted.

Anonymous: The NZ rules have a loophole. Suppose Alice and Mallory play until there are no more worthwhile moves, both knowing what stones will eventually be captured. Both pass. Mallory has much more dead stones than Alice. Mallory: "All my stones are alive." Alice: "Do you want to play next or shall I?" Mallory: "I want nobody to play next." Alice: "But I want that at least one of us plays next." Mallory: "So we disagree on who shall move next. Let's count."

Herman Hiddema: This loophole assumes that "I want nobody to play next" is a valid choice. Logically, Mallory would have to choose a player, wouldn't she?

If one player has dead stones but does not agree, the other player will just play and kill. Under New Zealand rules playing in your own territory does not cost you points.

Bill: Nobody is not a who.

Flower I agree that this part of the NZ Rules poses a possible Loophole. As Hermann said 'common sense' kind of dictates that each players chooses either player A or player B to make the next move. Yet 'common sense' is a poor companion if one wants to create concise and conclusive rules that give raise to few disputes. As such I think that the agreement phase handing is lacking in the NZ Rules. In order for them to "catch exceptions" we would need to add three more rules:

  • The players must choose either player A or player B
  • The players must not change their choice after announcing it
  • In case of disagreement the game continues as soon as one of both players yields to the other ones choice. Otherwise the position will be scored as it is.

Of course this impacts upon the NZ Rules conciseness and might be added to the NZ Rules commentary instead. Yet perhaps it would be simpler to reword the agreement phase handling. As such I think that the following three options are roughly equivalent (please comment). Especially the third option is easier to understand and prevents above exploit to my mind.

  • If they cannot then agree who shall move next, all stones stay on the board (are alive) and are counted.
  • the one who claims that a group should be removed needs to defer to the wish of the other player in order to show that the he can capture the group as otherwise he would disagree which would cause scoring with the group staying on board.
  • in case of disagreement the game is continued with the defendant's (the one claiming that his group will not be captured) prerogative to make the first move.

Anonymous: I wonder why the finishing is not defined in a logical way, for example: The game is finished when both players pass (formally) in succession. All stones stay on the board and the score is counted. The players may cooperate to avoid uninteresting game ending moves, for example by stopping the game informally and removing dead stones by mutual agreement.

Herman Hiddema: Finishing is defined in a logical way in these rules. The first two passes are an indication that neither player feels there are any worthwile moves left. This is virtually equal to your "stopping the game informally and removing dead stones by mutual agreement". By formalizing this trough passes, you avoid problems like language barriers.

Flower: I fail to see why you (Anonymous) think that the current alternation ending conditions are inconclusive. Can you elaborate?

Flower: Above content in this thread safe for my comments (which I just added) had been deleted. I revived it as a) I wanted to comment and b) I think the thread and the question it raises is important. (after all this is the discussion page and there is no need to delete threads that contain info :)


New Zealand Half Counting

Flower: While reading Robert's IGRF reports I encountered the term 'New Zealand Half counting'. It seems to have the advantage that the counting process does not disturb the final position. I wondered how exactly it is done. (My first guess would be, that one just counts all stones and territory without rearranging, which would be quite tedious but perhaps deemed an acceptable sacrifice for the verifiable result)

RobertJasiek: Read my [ext] pages and [ext] here, where Point By Point Half Counting is what you are looking for.

Flower: Thanks for both links :) your efforts that undoubtedly flowed into these pages are greatly appreciated.

LuizBorges?: I understand how Point-by-point is supposed to be done, but in real games, I found it very easy to lost track of what was counted and what was not. Do you recomend any particular method RobertJasiek to avoid that? What I tried was counting in twos, group by group, but when a groups had "jagged" edges is easy to get lost.


Same Number of Points

New Zealand Rules do not define the outcome of a game in which both players have the same number of points. What is the result of such a game? Shouldn't the rules state that explicitly?

Herman Hiddema: It's a draw. See /History.

Velobici: Thank you. Now I see why its a draw. Its an interesting arguement that perfect play should result in a draw. I will read the NZ rules more carefully...so far I can't find in the rules where its stated to be a draw.

Herman Hiddema: In this context, it is also interesting to read the Tromp-Taylor rules ([ext] http://homepages.cwi.nl/~tromp/go.html ), which are "essentially the New Zealand rules, re-worded to be as simple and elegant as possible. --Bill Taylor"


Peculiar Definition of Territory

MrMoto?: The definition of territory seems strange to me: first of all, it includes the word 'territory'; second, according to the definition of 'adjacent', it includes only those empty spaces immediately beside your stones. Thus, if you have a group enclosing a large area you won't get points for much of the interior. Of course, you can get the points back by playing sufficiently many times inside your own groups, but that's so vulgar ...

MrMoto?: Ah! I see that the definition is *supposed* to be recursive. That really wasn't clear when reading it the first time -- maybe the wording could be improved?

Martin?: Exactly, it's recursive. There is another "fault" in the definition, however: shared dame in seki are counted for both players, which seems weird. It doesn't change the score, but it's more intuitive for most go players to not count those points for either player. My suggested wording is something like

Territory of a player (at the end of the game) consists of all intersections occupied by that player's living stones plus all unoccupied intersections adjacent to that territory, that are not adjacent to the opponent's territory.


New Zealand Rules / Discussion last edited by 213.89.125.245 on April 1, 2021 - 12:51
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
RecentChanges
StartingPoints
About
RandomPage
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Goproblems.com
Login / Prefs
Tools
Sensei's Library