Claiming threefold repetition/disturbing neighbouring boards

[Alistair Campbell]

But what happens if you haven't been keeping your score-sheet up to date, but your opponent has? Do you just write down your proposed move any-old where and advise the arbiter?

Perhaps Jack or another arbiter can advise us, but my understanding was that you have to be recording the moves to make any claim. After all, how else could you prove what you say has happened? In practice I pretty much only play in 30-second increment events now so the issue of not-recording doesn’t arise for me.

The scenario I envisaged was something along the following lines (under a time-control without increments).

I keep my score-sheet up to date for, say, 60 moves.
The key position has arisen on, say,moves 56 and 58.
My opponent avoids the three-fold and I stop recording as I have under 5 minutes left on my clock.
Several moves later my opponent inadvertently allows me to the opportunity to claim. I don't know which move it is.

I note the law mentions writing my proposed move "on the scoresheet" (so not any-old where, but anywhere on the scoresheet, I guess).

Having the evidence of a partially complete scoresheet and the position on the board, I would hope a correct claim would be upheld.

A variant to this would be where all three instances of the key position arise after I have stopped recording, but my opponent has a complete scoresheet. Can I subpoena this scoresheet as evidence? What happens if my opponent "accidentally" spills his coffee over his sheet rendering it illegible? Can an arbiter collect other evidence such as eye-witness reports or DGT print-outs?

A related question is the standard of proof required to make such decisions - on the balance of probabilities, beyond reasonable doubt or something else?

With regard to noise on neighbouring boards - I think as a player you have to expect this and make allowances. Of course polite remarks at the end of games often degenerate into conversations - "Well played (upon resigning) - I should have swapped rooks when I had the chance" - "I was lucky, I thought you were much better - maybe you had to play h6..." might be acceptable; anything more is too much.

I have long thought that the loudest, most distracting behaviour is that of people trying to put equipment away "quietly" whilst the last few games are in progress..

[Tim Harding]

You can claim it

But precisely how? What do you do?

Cry? Scream? Stand on the table?

[NickFaulks]

I have long thought that the loudest, most distracting behaviour is that of people trying to put equipment away "quietly" whilst the last few games are in progress.

My votes goes to the players who can be relied to go "SSSHHHH" at maximum volume whenever a whisper can be heard from the other side of the room.

[NickFaulks]

Cry? Scream? Stand on the table?

None of these is ideal. I take it you're saying that you don't know either.

I seriously don't believe that anyone does know, and this seems to me an important gap in the regulations. Unless, of course, it really is the intention that the right to claim a draw by threefold repetition does not exist in rapid games or the late stages of quickplay finishes.

[Ian Kingston]

I'll offer the following story to illustrate that the problem can be resolved if there is good will and reasonable behaviour on both sides.

The Birmingham League has a time control option with a 10 second increment. I'd reached the wrong end of K+Q vs K+Q+P, with the pawn a long way from queening, so there was the potential for a very long game. Both players had less than five minutes left, so we weren't keeping score. There was no arbiter present, of course.

Fortunately, the position was very favourable for the defence, as I could start checking without allowing the enemy queen to interpose usefully. It only took a few moves to reach the same position for the third time, but I knew very well that I couldn't back up my claim. So what did I do? I made the claim anyway, saying 'My next move will be ... and I claim a draw by repetition' (the closest approach to the proper way to claim that I could think of). My opponent was equally well aware of the laws, and (quite properly) rejected the claim. So I played the move and offered a draw in the normal way. He turned it down. We played one more move and then he offered the draw, which I accepted.

Overall, I think that was a decent solution to the problem. We both made our point, and the right result was achieved.

[Angus French]

While we are on the subject, if you have a perpetual but are seconds from losing on time (if there are no increments, of course), can you stop the clock and demonstrate the perpetual to the arbiter in support of a 10.2 claim (or whatever the numbering is now). Or would the arbiter (if the right clocks are available) reset the clocks to increment play and ask for play to continue?

Appendix G replaces the old 10.2 (and also the old Appendix D).

G.4 states "If the player having the move has less than two minutes left on his clock, he may request that a time delay or cumulative time of an extra five seconds be introduced for both players, if possible."

It's perhaps worth pointing out that Appendix G "appl[ies] to standard play and rapidplay games without increment and not to blitz games" (G.3).

(FWIW, the Chess Arbiters' Association has an annotated version of the 2014 Laws of Chess which I often consult.)

[NickFaulks]

My opponent was equally well aware of the laws, and (quite properly) rejected the claim.

If you are saying that your opponent knew perfectly well that the position was occurring for the third time but rejected the claim because he could, I regard that as, to say the least, unsportsmanlike. Am I out of touch with the modern game?

[Ian Kingston]

My opponent was equally well aware of the laws, and (quite properly) rejected the claim.

If you are saying that your opponent knew perfectly well that the position was occurring for the third time but rejected the claim because he could, I regard that as, to say the least, unsportsmanlike. Am I out of touch with the modern game?

No - I'm saying that he may not have been sure at that point (the third occurrence followed a different move sequence), and perhaps a bit surprised. There's no suggestion of unsporting conduct whatsoever.

[NickFaulks]

No - I'm saying that he may not have been sure at that point (the third occurrence followed a different move sequence), and perhaps a bit surprised. There's no suggestion of unsporting conduct whatsoever.

It seems that everything was fine, as I'm sure it is in 90% of such cases, probably more. However, you and your opponent achieved this despite the unhelpful nature of the regulations, which say quite clearly that somebody has to write something on a piece of paper.

[Ian Kingston]

No - I'm saying that he may not have been sure at that point (the third occurrence followed a different move sequence), and perhaps a bit surprised. There's no suggestion of unsporting conduct whatsoever.

It seems that everything was fine, as I'm sure it is in 90% of such cases, probably more. However, you and your opponent achieved this despite the unhelpful nature of the regulations, which say quite clearly that somebody has to write something on a piece of paper.

Exactly. I have no idea how the laws can be changed to deal with this.

[Mike Gunn]

To claim you stop the clock and summon the arbiter. In the absence of a completed scoresheet you could demonstrate the sequence of moves which lead to a 3 fold repetition. If you can't do that (or the opponent disagrees), you will probably be unsuccessful in your claim. If you have an "unsporting" opponent you will almost certainly be unsuccessful in your claim. Your only chance of success if the opponent says "oh yes, that's right".

[Jonathan Bryant]

If you have an "unsporting" opponent you will almost certainly be unsuccessful in your claim.

My opponents weren’t in the least bit unsporting they just didn’t know the rules. They were acting correctly according to the circumstances as they understood them.

If my situations at Gatwick had arisen in circumstances as described above I wouldn’t have got my draw(s). However, I think when you play under such conditions - guillotine finishes, I mean - you have to accept what comes along with them. That includes maybe not being able to get a valid claim for a threefold repetition accepted.

It’s one of the reasons why I prefer a 30 second increment and don’t like 10 second increments.

[NickFaulks]

To claim you stop the clock and summon the arbiter. In the absence of a completed scoresheet you could demonstrate the sequence of moves which lead to a 3 fold repetition. If you can't do that (or the opponent disagrees), you will probably be unsuccessful in your claim. If you have an "unsporting" opponent you will almost certainly be unsuccessful in your claim. Your only chance of success if the opponent says "oh yes, that's right".

The above makes sense, but has no grounding in the regulations, which insist that a move should be written on a piece of paper. If it is universally agreed that the regulations must be ignored because they are daft, then they need changing.

[Stewart Reuben]

Christopher Kreuzer (upthread) > It is only when one person doesn't realise that the position has been repeated three times that things get tricky. <sigh<

A bigger sigh. The claim can be made when the position has OCCURRED. or is about to OCCUR, three times. It does not have to be repeated three times.

An anecdote regarding Korchnoi.
In 2000 at the Olympiad he claimed a draw because the position had occurred or was about to occur three times. His clock was going. I was the arbiter and made no effort to look at Viktor's scoresheet. They were invariably illegible. I stopped the clock, I looked at Eugenio Torre's scoresheet which was easy to read and easy to determine that Viktor was correct. I asked the Filipino whether he agreed. He did, so the game was drawn.

If the same position has appeared after several moves it is much more difficult. Then the arbiter has to move away from the games in progress and preferably re-construct the game eventually on three boards. Use a coin to indicate who is to move for each instance. One of the reasons for such a complex way of doing things is to convince the players that the claim is correct/incorrect. In the absence of an arbiter this is very difficult to achieve.

[Christopher Kreuzer]

Christopher Kreuzer (upthread) > It is only when one person doesn't realise that the position has been repeated three times that things get tricky. <sigh<

A bigger sigh. The claim can be made when the position has OCCURRED. or is about to OCCUR, three times. It does not have to be repeated three times.

I've noticed this wording used a couple of times. To me, there is no difference between a position being repeated three times or occurring three times. The meaning (linguistically) is the same. I get the distinction between a position being repeated (a draw) and the moves being repeated (not necessarily a draw as the position may be different), but the distinction is between the words 'position' and 'moves', not between the words 'repeated' and 'occurred'.

Or am I missing the point here? Are you saying that when a position has been repeated three times, that it has occurred four times? If so, then would the wording that a position has been repeated two times be the equivalent of a position occurring three times? Some people would consider the first occurrence of a position to be one of the three repetitions, but I can see now why the wording 'occurrence' is preferred (actually, the rules use the wording 'appeared').

It is easiest to actually demonstrate with example games. If the players develop and undevelop their kingside knights as follows:

1.Nf3 Nf6 2.Ng1 Ng8 3.Nf3 Nf6 4.Ng1

At this point, Black can write 4...Ng8 on his scoresheet and declare his intention to make this move and claim a draw as the starting position will have appeared on the board for the third time. The move is then made, White records it, and the draw is recorded on both scoresheets. Whether the arbiter would accept it is another matter...

Black could also play 4...Ng8 and not make a claim, but White could make the claim instead at this point.

1.Nf3 Nh6 2.Ng1 Ng8 3.Na3 Nc6 4.Nb1

As above, Black can write 4...Nb8 on his scoresheet and declare his intention to make this move and claim a draw as the starting position will have appeared on the board for the third time. Or Black can make the move and see if White will make the claim.

You can add lots of different moves to the examples above, as long as the three positions you end up repeating are the same with the same player to move and there are no castling or en passant differences between the three positions (really, simple examples of those should be provided, and I am sure some explanations or arbiter-expanded versions of the rules do include such examples). You can even swap the kingside and queenside knights, as the pieces don't have to be exactly the same ones in the three positions (ditto you can swap rooks if two of the same colour are on the board).

The relevant bit of the rules is: "pieces of the same kind and colour occupy the same squares and the possible moves of all the pieces of both players are the same".

Things that definitely break the possibility of a position repeating are: pawn moves (including pawn promotions), as pawns can't move backwards or reappear on the board once promoted, and piece captures. Castling doesn't mean a position cannot be repeated, as you can 'uncastle' by hand, but the king or rook will have lost castling rights, so the position count needs to restart after the castling takes place (or after a king or rook moves for the first time, which by definition is the case when castling takes place). So the thing that breaks the repetition sequence is a rook or king moving for the first time.

You can construct examples that appear to be edge cases. For example:

Position 1: White is in check but has not yet castled. However, the only way to evade the check is to move the king. Does the White king still technically have castling rights before it has moved? My reading of the rules, which explicitly state 'The castling rights are lost only after the king or rook is moved' is that yes, even if the loss of castling rights is forced, the rights are still there and 'possessed' by the king and rook up until the point that the move with the king or rook is made.
Position 2: After a series of checks (with nothing else changing), the White king returns to e1 and the same position occurs as above. The position is different, because the White king has now lost castling rights.
Position 3: After another series of checks (again with nothing else changing), the White king returns to e1 and the same position occurs as above. This looks like the same position for a third time, but the first occurrence is not eligible because the castling rights were still there (though the loss of castling rights was forced).
Position 4: After another series of checks (again with nothing else changing), the White king returns to e1 and the same position occurs as above. The position has now occurred for the third time with the same (lack of) castling rights.

Change the above positions slightly by constructing a position where White has two options when in check: either move the king (for the first time), or interpose a piece to block the check. Do the latter, and the position can occur three times no problems. Once the king is moved, the count needs to restart.

You also need to consider *both* kingside and queenside castling. A king that can castle both sides is different from one that can only castle one side and not the other. This can happen when a rook moves away from its starting square and back again. When the king moves, the castling rights of the king and both rooks (if both are still on the board) are affected. When a rook moves (or is captured), the castling rights of that rook and the king are affected (but not the castling rights of the other rook). I say 'or is captured', but of course any capture resets the position count for threefold repetition, so that doesn't really matter.

This sort of case, where the queenside rook may have moved away from and back to its start square, and the possibility of queenside castling is not obvious (e.g. the king is obviously ready to castle kingside) and the position appears to have repeated three times (but the castling rights have subtly changed), is one where I suspect even experienced players might get confused.