Draughts rules

[ildjarn]

In chess there are similar interpretations of the 50-move rule. See e.g. http://www.janko.at/Retros/Masterworks/Part1.htm composition #3 (J. Sunyer). I've also seen several Plaksin compositions where white deliberately allows a mate in 1 to claim the draw by the 50-move rule.

[ildjarn]

The exception rule can only be applied when the game is still going on. Capturing the last piece (or leaving your opponent no playable moves) immediately ends the game, so the exception rule doesn't kick in.

The order is:
Move - rule - exception rule
In the case of capture of the last piece: Move - rule (which ends game). Exception rule isn't applied because it's never reached.

[GuidoB]

Does the sequence 1.23-40 2-11?? 2.40-44 11-16 3.44-49 16-2 4.49-35 2-11 5.45-34 11-2 6.34-39 2-7 7.50-45 7-2 8.35-49 2-16 9.45-40 16-7 10.39-34 7-11 11.34-1 11-2 12.40-34 2-16 13.34-29 16-11 14.49-44 11x50 15.1-6 50-45 16.6-1 45x23 lead to an immediate draw? I don't think so. After 28 ply a new situation occurs. 6.3 does no longer apply and one should start counting again concerning 6.4. 5 ply later the game has finished in a win for white.

Not agreed! The way I read rule 6.3 is that each player has 16 moves left after the first occurance of a 3 vs 1 situation. The 2 vs 1 counter also applies in your example from move 14 onwards, but only in a cumulative sense, not as an overriding rule! Of course, because the numbers 32 and 10 were derived from optimal play endgame databases, rule 6.4 cannot endanger a win under rule 6.3 as long as the majority side plays optimally.

One could disagree whether it is reasonable that these rules require optimal play in small endgames. Of course a balance has to be struck between avoiding endless shuffling of pieces in drawn endgames, but in my opinion article 6.2 is already taking care of that and rules 6.3 and 6.4 are superfluous and just plain annoying.

6.3 does not say anything like 'two kings plus one man at most', it just says 'two king plus a man'. The position after 28 ply is no longer embedded in rule 6.3, so one cannot claim a draw by this rule. The 2 vs 1 position lasts for 4 ply only, so one cannot claim a draw by rule 6.4. The 1 vs 1 position lasts for 1 ply only, so one cannot claim a draw by rule 6.4. Therefore, it is a win.

[TAILLE]

In chess there are similar interpretations of the 50-move rule. See e.g. http://www.janko.at/Retros/Masterworks/Part1.htm composition #3 (J. Sunyer). I've also seen several Plaksin compositions where white deliberately allows a mate in 1 to claim the draw by the 50-move rule.

The equivalent in draughts may be the following position:


White to play

I found this position on Dragon site in the endgame statistic part.

1. 9-36 34-40 2. 33-28 40-45


White to play

Now I begin to increment the 25 moves counter
3. 26-17 4-15 4. 17-6 15-29 5. 41-47 29-7 6. 47-38 7-12 7. 38-20 12-8 8. 20-3 8-35 9. 3-21 35-49 10. 21-8 49-35 11. 8-2 35-40 12. 2-30 40-7 13. 30-43 7-12 14. 43-25 12-21 15. 25-3 21-16 16. 36-13 16-7 17. 13-2 7-18 18. 2-16 18-1 19. 3-17 1-29 20. 16-2 29-18 21. 2-24 18-36 22. 24-47 36-18 23. 17-21 18-9! (and not 18-7? as indicated on the above site) 24.21-26 9-18 25.47-43 18-36 26.33-50 36-18 27.06-01 18-36


White to play

The 25 moves counter has reached the 25 value.
The game is a draw, even if 28.28-23! is now a winning move.

Dramatic!

[TAILLE]

The exception rule can only be applied when the game is still going on. Capturing the last piece (or leaving your opponent no playable moves) immediately ends the game, so the exception rule doesn't kick in.

The order is:
Move - rule - exception rule
In the case of capture of the last piece: Move - rule (which ends game). Exception rule isn't applied because it's never reached.

I do not see any word in the rule saying that 6.3 apply only if it remains some moves to play. Capturing the last piece with the 16th move surely concludes the game for two reasons : "General rule" + "Exception rule".

Now I cannot understand the order "Move - rule - exception rule". Do you have an exemple of such order on a subject different from draughts ?

[Rein Halbersma]

6.3 does not say anything like 'two kings plus one man at most', it just says 'two king plus a man'. The position after 28 ply is no longer embedded in rule 6.3, so one cannot claim a draw by this rule. The 2 vs 1 position lasts for 4 ply only, so one cannot claim a draw by rule 6.4. The 1 vs 1 position lasts for 1 ply only, so one cannot claim a draw by rule 6.4. Therefore, it is a win.

Code: Select all

6.3. If only three kings remain, two king plus a man, one king and two men, against one king, the game shall be considered a draw when the players have each played another sixteen moves maximum. 

IMHO, the rule says that you can play at most another 16 moves each after the first occurance of a 3 vs 1 situation with at least one king each. The rule doesn't say that a 3 vs 1 situation has to be on the board during that entire sequence. So I find it hard to interpret the above as "the 16 moves counter is reset when the 3 vs 1 endgame no longer exists".

[ildjarn]

Now I cannot understand the order "Move - rule - exception rule". Do you have an exemple of such order on a subject different from draughts ?

Some chess equivalents:

If the 100th move without capture/pawn move delivers mate, the game is ended, and the 50-move rule doesn't apply.

There's a rule 'If there is no possible legal way the game can be won, the game is a draw' (e.g. King+bishop against king or stalemate, but not king+2 knights against king, because in the latter, although it's a draw endgame, there are legal ways the game can be won). This draw immediately ends the game. If someone doesn't see the draw, and keeps on thinking until his flag drops, this doesn't change the fact that the game is still a draw.

[TAILLE]

Now I cannot understand the order "Move - rule - exception rule". Do you have an exemple of such order on a subject different from draughts ?

Some chess equivalents:

If the 100th move without capture/pawn move delivers mate, the game is ended, and the 50-move rule doesn't apply.

There's a rule 'If there is no possible legal way the game can be won, the game is a draw' (e.g. King+bishop against king or stalemate, but not king+2 knights against king, because in the latter, although it's a draw endgame, there are legal ways the game can be won). This draw immediately ends the game. If someone doesn't see the draw, and keeps on thinking until his flag drops, this doesn't change the fact that the game is still a draw.

If you find in the chess rules a statement like "If the 100th move without capture/pawn move delivers mate, the game is ended, and the 50-move rule doesn't apply" then it is fine because it is an explicit "Exception rule" to another "Exception rule" which is of course acceptable. But in the draughts rules you cannot find the equivalent explicit new exception. That is my point but I guess I missed something.

[Vraagje]

In the KNDB-rules the order of the rules is important: a rule that is presented earlier in the 'rule-book' is more important then a rule that is presented later. In the KNDB-rules, the rules for win come before rules for draw, so I suppose the position would be winning. In the FMJD-rules, the rules for draw seem to be in article 6, and the rules for win in article 7 (however there is referred to draws in that article also). So I doubt if there is a difference between KNDB and FMJD-rules. (I do not know if the FMJD also uses this principle of the order of the rules.)

[TAILLE]

In the KNDB-rules the order of the rules is important: a rule that is presented earlier in the 'rule-book' is more important then a rule that is presented later. In the KNDB-rules, the rules for win come before rules for draw, so I suppose the position would be winning. In the FMJD-rules, the rules for draw seem to be in article 6, and the rules for win in article 7 (however there is referred to draws in that article also). So I doubt if there is a difference between KNDB and FMJD-rules. (I do not know if the FMJD also uses this principle of the order of the rules.)

I am not sure to understand your point. What is the differnce between the two different orders :

Exception rule : when you reach a winning position in less than 16 moves then the game is a win otherwise (winning position reached after 16 moves or more) the game is considered a draw.
General rule : when you reach a winning position the game is a win

or

General rule : when you reach a winning position the game is a win
Exception rule : when you reach a winning position in less than 16 moves then the game is a win otherwise (winning position reached after 16 moves or more) the game is considered a draw.

In any case an "Exception rule" is a rule identifying a case where the "General rule" has not to be applied. The order is not relevant for me.
The only point causing problem is when the winning position is reach after exactly 16 moves, but by definition the "Exception rule" has always to be applied BEFORE the "General rule".


Black to play

Of course I hear the argument saying that in the above position there are no more black pieces on the board after 16 moves and, as consequence, the game is not going on. That is completly true but it adds nothing to the fact that the winning position is reached after 16 moves. In any case the losing side cannot make a move as soon as a winning position is reached because in a winning position the losing side as no more pieces or have all its pieces blocked.

It is then a non sense to say that the "Exception rule" does not apply if the game is not going on. If it is written that the winning position must be reached in less than 16 moves that means that the game is a draw if the winning position is reached after exactly 16 moves and in this last case the opponent has obviously no more move left by definition of a winning position.

If it was the intention to declare a win if the winning position is reached after exactly 16 moves then the "Exception rule" would have been written accordingly. Unless you change the "Exception rule" a winning position reached after 16 moves implies a draw game.

IMO the above position was not discovered when the exception rule has been written. It's only a mistake with no great imporance because the probabilty that this position occurs in a real game is almost null.

[ildjarn]

The exception rule can only be applied when the game is still going on. Capturing the last piece (or leaving your opponent no playable moves) immediately ends the game, so the exception rule doesn't kick in.

The order is:
Move - rule - exception rule
In the case of capture of the last piece: Move - rule (which ends game). Exception rule isn't applied because it's never reached.

I just realised that this order is wrong. The correct order should be:
Movecount rule - move - Result rule

In the case the 32nd ply is the capture of the king of the minority party:
- Are there 32 plies since starting the 3x1? no
- Move: Capture the king
- Result rule: Is the game decided? Yes

So in this case, the majority party wins.

Next case, the 32nd ply is the move that gives away the king:
Ply 32:
- Are there 32 plies since starting the 3x1? no
- Move: Sacrifice the king
- Result rule: Is the game decided? No
Ply 33:
- Are there 32 plies since starting the 3x1? yes ==> Minority party can claim the draw

For the practical implementation there are two details that are important. I take the KNDB text:

Code: Select all

De partij is remise als één van de spelers of the arbiter tijdens de partij constateert:

(Translation: The game ends in a draw if one of the players or the arbiter, during the game, observes that:)
- The draw has to be claimed
- The draw must be claimed during the game. If, in the case that black sacrifices his piece on the 32nd ply, he allows white to capture the piece, then the game is finished, and the draw can no longer be claimed.

[Rein Halbersma]

The exception rule can only be applied when the game is still going on. Capturing the last piece (or leaving your opponent no playable moves) immediately ends the game, so the exception rule doesn't kick in.

The order is:
Move - rule - exception rule
In the case of capture of the last piece: Move - rule (which ends game). Exception rule isn't applied because it's never reached.

I just realised that this order is wrong. The correct order should be:
Movecount rule - move - Result rule

It's simpler to do it like this: 1) result rule (FMJD 7.1 and 7.2), 2) drawing rule (FMJD 7.3 and 6.4), 3) moving/capturing rule (FMJD 3 and 4). Your order actually has different rules being evaluated at different moments in the game.

So if you can't play a legal move, you've lost. If you can play a move but if the game history has reached a critical counter (3, 25, 16 or 5 from articles 6.1 through 6.4) it's a draw. Otherwise, you play any of your legal moves.

This is how I have programmed my search function in my draughts engine.

[ildjarn]

Whether the result rule is applied as last step of a ply or as a first step of a ply is of course irrelevant (except for the first ply).

[TAILLE]

The exception rule can only be applied when the game is still going on. Capturing the last piece (or leaving your opponent no playable moves) immediately ends the game, so the exception rule doesn't kick in.

The order is:
Move - rule - exception rule
In the case of capture of the last piece: Move - rule (which ends game). Exception rule isn't applied because it's never reached.

I just realised that this order is wrong. The correct order should be:
Movecount rule - move - Result rule

In the case the 32nd ply is the capture of the king of the minority party:
- Are there 32 plies since starting the 3x1? no
- Move: Capture the king
- Result rule: Is the game decided? Yes

So in this case, the majority party wins.

Next case, the 32nd ply is the move that gives away the king:
Ply 32:
- Are there 32 plies since starting the 3x1? no
- Move: Sacrifice the king
- Result rule: Is the game decided? No
Ply 33:
- Are there 32 plies since starting the 3x1? yes ==> Minority party can claim the draw

For the practical implementation there are two details that are important. I take the KNDB text:

Code: Select all

De partij is remise als één van de spelers of the arbiter tijdens de partij constateert:

(Translation: The game ends in a draw if one of the players or the arbiter, during the game, observes that:)
- The draw has to be claimed
- The draw must be claimed during the game. If, in the case that black sacrifices his piece on the 32nd ply, he allows white to capture the piece, then the game is finished, and the draw can no longer be claimed.

OK Ildjarn now I understand why our interpretations are different : in KNDB rules it exists really a new explicit exception rule saying that "in the case that black sacrifices his piece on the 32nd ply, he allows white to capture the piece, then the game is finished, and the draw can no longer be claimed". If I understand correcly the picture for KNDB it looks like the 3 following rules :

General rule : when you reach a winning position the game is a win
Exception rule : when you reach a winning position in less than 16 moves then the game is a win otherwise (winning position reached after 16 moves or more) the game is considered a draw.
Exception to exception rule : in the case that black sacrifices his piece on the 32nd ply, he allows white to capture the piece, then the game is finished, and the draw can no longer be claimed

With this "Exception to exception rule" I agree with you. In addition it seems to me that this "Exception to exception rule" brings a real difference of 2 plies (and not only 1 ply). Do you know why this rule is in KNDB rules and not in the international rules?

BTW that brings me another question.
Imagine that the 32th move is not the capture of the black king but a move that BLOCKS this black king (it is the second case in the definition of a winning position). In that case, because the "Exception to exception rule" do not apply the result is changed. Is it a good understanding ?

[Vraagje]

...

There is not made an exception-rule. In the KNDB-rules the order of the rules is important. The first rule is stronger than the second rule (unless the second rule is clearly meant to overrule the first rule).

There are two rules that are important.
1. The game is lost when one has no pieces left to play.
2. The game is draw after 16 moves in 3x1.

This is the order in which these rules are in the KNDB-rules. Because rule 1 is mentioned earlier, a winning position after exactly 16 moves is considered a win. Should rule 2 be mentioned first, the game would be a draw.

---------------------------------------------

I'm starting to doubt now. In the KNDB-rules you lose when you have no pieces left at your own move. So if the 16th move (capture) is made by the player with the ''winning position'', I would suppose first the 16th move is completed, and only after that the other player seems to have no pieces left.

But... you should notice during the game that 16 moves have been played, but you have no time to do so between completion of the move and your own turn. Conclusion: I do not know.