Talk:Threefold repetition

Old talk

 * In another amusing game, one of the players claimed a draw by repetition which his opponent disputed by saying that in the previous instances of the postion the positions of the two white rooks had been interchanged! What followed was a painstaking replay of the game by tournament officials on an alternate board to verify the claim.

I have removed this, as even if this very unlikely situation were to occur, such an interchange of rooks would not be an issue under modern FIDE rules. To quote part of article 9.2 of the FIDE laws of chess (see ):


 * Positions as in (a) and (b) are considered the same, if the same player has the move, 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.

Note in particular the part I have italicised. It does not say "the same pieces occupy the same squares" but rather "pieces of the same kind and colour". Therefore, even if such an interchange of rooks as described occured in a game, a draw could still be claimed.

I suspect, therefore, that the story is apocryphal. If it really did occur, the game must have been played under some other ruleset (or the arbiters were incompetent, which wouldn't really make it worth mentioning in the article). If this is so, there should be a note of explanation saying under what rules this incident occured, and that under current FIDE rules it would not. I think the story could also do with a source, as it sounds rather unlikely to me.

I also have my doubts about the Karpov - Miles story. Do we know when and where this game was played, or do we at least have a source for the story? A quick check through my database (which is not completely comprehensive, but which is pretty large) turns up no games between the two where this sort of thing happened. --Camembert


 * Hi,


 * Totally sure about the Karpov-Miles game. I remember that I read it in the "Chess Mate" magazine, so I could find the reference, but I have something like 200 copies of the mag, which would take hours, so I'm unwilling to do it.


 * About the other game, it was some local tournament in India; either my memory is wrong or the arbiters were incompetent, either way, as you say, it shouldn't be in the article.


 * Arvindn 13:30, 28 Mar 2004 (UTC)

No problem on Karpov-Miles - if you're sure it's right, I believe you. Do you have any idea about what year it happened, roughly? I want to try and track down the score (it would be good to quote in the article). --Camembert

Aha! I think I've found it: Karpov-Miles, Tilburg 1986. I'll add the reference to the article. --Camembert


 * Wow, nice find :) Still, I have a lingering doubt: I seem to remember that it was Karpov who had lost the right to castle, whereas in the game you gave it is Miles! (and therefore the article is inconsistent as it stands.) Miles is apprently a jolly guy, is there a chance of emailing him or something? :) Arvindn 15:34, 28 Mar 2004 (UTC)


 * Ah, good point. Should've noticed that really. I'll take the reference back out for the time being, and see if I can dig up something more comprehensive. No chance of mailing Miles, I'm afraid - he died a couple of years ago - but I'll have a poke around and see if I can come up with anything. We'll get this sorted out eventually, I'm sure :) --Camembert


 * Oh, sorry, didn't know Miles died, about 7 years since I played tournaments; haven't kept up with news since then. Yes, it'd be real nice to get this sorted out :) -- Arvindn 15:52, 28 Mar 2004 (UTC)


 * You were right - It was indeed Tilburg 1986. So that issue is resolved :) -- Arvindn 09:16, 11 September 2005 (UTC)

Castling - temporary
The article says "Positions are not the same if a pawn that could have been captured en passant can no longer be captured or if the right to castle has been changed temporarily or permanently. "

I don't see any way for the right to castle being changed temporarily if the position is the same. If that is correct, "temporarily" should be removed from that sentence. Bubba73 (talk), 23:57, 21 December 2005 (UTC)

Karpov-Miles
According to Chessgames, Karpov had white, so it was Miles' king that could castle in the first of the three positions. The article says it was Karpov's king. Chessgames.com is probably right. Bubba73 (talk), 01:41, 17 February 2006 (UTC)


 * I changed it from Karpov to Black. Another thing, in the position shown, with the moves I got from ChessGames, under the rules, it should have been Black that was claiming the draw, not White (Karpov).  Right?? Bubba73 (talk), 03:14, 17 February 2006 (UTC)


 * Well, it depends on whether the last move shown was the one written on the scoresheet and shown to the director, and if it was actually made on the board. Bubba73 (talk), 02:00, 19 February 2006 (UTC)

Karpov-Miles, Tilburg 1986, position after 22. Nb5
As far as I know, it is impossible to have a white pawn on the A row. Is there something wrong with this game? Erikina 09:22, 16 May 2006 (UTC)


 * That was an error, I fixed it. Bubba73 (talk), 14:51, 16 May 2006 (UTC)

Alekhine-Lasker analysis
Can someone check the ref for that game to make sure this story is correct. It seems to me that Alekhine is winning this game quite handsomely, so I suppose he offered the draw? And the analysis cannot be right (even if it is as it reads): after 16. ...Qe8, White has Qh7#! Did they mean 16. ... f5? Baccyak4H (Yak!) 18:09, 19 July 2007 (UTC)


 * There was an error in the position, should be a black knight on f6 instead of a pawn. Probably my error; thank you for pointing it out.  Bubba73 (talk), 18:27, 19 July 2007 (UTC)


 * Great, thanks. Actually, I messed up f6 inadvertently when I tried to fix the position before (when Lasker's King was on e6!): when playing through the moves, I captured on f6 rather than h6 on Alekhine's 13th.  Baccyak4H (Yak!) 13:59, 20 July 2007 (UTC)


 * OK, you can take credit for the f6 error and I'll take credit for the two prior errors. Thank you.  Bubba73 (talk), 14:06, 20 July 2007 (UTC)

Fischer - Petrosian
History entry says, "...at first Pet did not even understand what was going on, and it was only when Schmid ... that the ex-champion began studying his scoresh)"

That sounds to me more like that Petrosian was surprised, not that he didn't know the rule.

Would you be able to provide a fuller quote? I still find it very hard to believe Petrosian was ignorant of the rule. He might have been taken by surprise by the claim (I've often seen people surprised by a 3-fold repetition claim), but that is very different from being ignorant that such a rule exists. Peter Ballard 04:09, 23 October 2007 (UTC)


 * I didn't write the last sentence of the paragraph. I wrote the part up to that point.  The two sources are almost identical in their wording. "At first Petrosian did not even understand what was going on, and it was only when the arbiter went behind the stage to check the position on a spare set that the ex-champion began studying his scoresheet. Fischer jumped up and followed the arbiter to make sure of the draw as soon as possible. Petrosian was distressed by the result..."  I think he was just surprised by the rule, not ignorant of it.  I didn't write the part about it being his first encounter with the rule, and I don't know where that came from, or if it is an assumption by an editor.  Bubba73 (talk), 04:24, 23 October 2007 (UTC)


 * OK, I get it. He was taken by surprise by the draw claim. Perhaps a slight reword is in order.


 * It happens all the time down at my level. I remember, many years ago in the days of adjournments, I helped a friend analyse a win in a difficult ending against a strong opponent (a future IM). But what happened was that they made about one move each and his opponent claimed a 3-fold repetition. It had not occured to either of us to check the scoresheet for that possibility. Petrosian's reaction sounds quite similar to that of my friend. Peter Ballard 04:30, 23 October 2007 (UTC)
 * Petrosian himself said it was the first 3-fold repetition of his career, but doesn't say whether he knew the rule. If you look for the book Russians v Fischer on Amazon.com, the text is searchable.  Searching for "studying his scoresheet" will direct you to page 284.  You can look at the discussion of the game on pages 283 through 285.
 * (p. 283) Petrosian: "But for the first time in my life I ran into a three-fold repetition of position..." (elipses in source text)
 * (p. 284) Baturinsky: "When Fischer asked the arbiter to come over and told him in English that with his next move 34 Qe2 he was claiming a draw, at first Petrosian did not even understand what was going on,"
 * It seems that it was a combination of unfamiliarity with 3-fold repetition and English that caused Petrosian's confusion. P.S. I highly recommend Russians versus Fischer for anyone interested in Fischer. Quale 04:31, 23 October 2007 (UTC)


 * Thanks for the tip. But Amazon will not let me view the page because I've never made a purchase from them. Peter Ballard 05:00, 23 October 2007 (UTC)


 * Sorry, the amazon suggestion wasn't as helpful as I hoped. You both are almost surely correct that Petrosian was familiar with the rule.  Repeating moves/positions was a common way to gain time on the clock before adjournment, and Petrosian would have had many years experience with it.  Since the current wording in the article is a little unclear, maybe it should instead say that Petrosian was surprised or caught off guard by the draw claim. Quale 14:52, 23 October 2007 (UTC)


 * He may have been caught off gaurd by the fact that the intermediate moves/positions were different. Bubba73 (talk), 16:45, 23 October 2007 (UTC)


 * Certainly he must have been familiar with the rule, because he started studying his scoresheet, as if to check. He must have just been surprized that it came up.  Bubba73 (talk), 04:52, 23 October 2007 (UTC)

Peter (or anyone), yes it would probably be a good idea to reword it. Bubba73 (talk), 11:58, 23 October 2007 (UTC)

Fischer-Petrosian, move 33
Plisetsky & Voronkov, Kasparov, and ChessBase give 33. Qh5 but ChessGames.com gives 33. Qd3, as the article was recently changed. 33. Qd3 is certainly plausable, and may have contributed to Petrosian's failure to recognize the repetition. What is the correct white move 33? Bubba73 (talk), 04:54, 6 December 2007 (UTC)


 * In the disucssion at ChessGames, it quotes a book by Petrosian in Spanish that gives 32...Re5 33 Qd3 Rd5, draw. Bubba73 (talk), 05:00, 6 December 2007 (UTC)


 * Actually, ChessBase gives moves 32 and 33 as duplicates of 30 and 31 (queen moves instead of the rook moves). The two books above give the same as the article currently has, except for 33. Qh5 instead of 33. Qd3.  Bubba73 (talk), 05:39, 6 December 2007 (UTC)


 * Chess Life, Nov 1971, page 619, gives 32. Qe2 Re5 33. Qd3 Rd5 34 Qe2 draw. Bubba73 (talk), 17:55, 6 December 2007 (UTC)


 * Chess Life, December 1971, page 682 agrees with the moves, and comments "Petrosian did not seem to realize that he was allowing a three-time repetition of the position. Of course, what is confusing is that, in reaching the same position three times, Black's move was different each time, but tht has no effect onthe position repeated. Nor did it have any effect on Bobby whonoticed it right away.'".  So I believe that 33. Qd3 is correct, not Qh5 as in the two books and in ChessBase.  Bubba73 (talk), 18:27, 6 December 2007 (UTC)

Two questions
I guess the answers to these may be pasted into the article.

1) "b. has just appeared, and the player claiming the draw has the move"

Does the word just mean that if the player doesn't claim the draw immediately (that is, before he has done his own move), he (and his opponent) loses the right to do so?

2) "The rule Positions as in (a) and (b) are considered the same, if the same player has the move, 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".

Note the italicized part. Could anybody explain me: is it possible that the pieces, given that the position is exactly the same, have not the same moves? As for me, it is possible only if we take into account en passant and castling - but they are explicitly mentioned below. What's the point of having the words "and the possible moves of all the pieces of both players are the same" here?

Thank you very much.93.183.227.110 (talk) 06:36, 21 March 2009 (UTC)


 * (1) - yes for that player, for that move. If he doesn't claim it then, he loses the right to claim it.  If the position comes up again (i.e. for the fourth time) he again has the oppertunity to claim it.  Or if any other position comes up three times the draw can be claimed.


 * (2) yes, but only because of en passant and castling. There is an example in the article from a game by Karpov where the castling ability changed.  Bubba73 (talk), 14:10, 21 March 2009 (UTC)

Capablanca versus Lasker
Was the Threefold respetition a rule at the time of this match in 1921? Regards, SunCreator (talk) 00:34, 12 April 2010 (UTC)


 * I don't know for sure, but I assume it was since the rule was used as early as 1883 and in the 1886 World Championship match. Bubba73 (You talkin' to me?), 01:07, 12 April 2010 (UTC)
 * The rules then was a 6 move repetition? I would of thought two world title contenders would of had a good grasp of the rules. Regards, SunCreator (talk) 01:24, 12 April 2010 (UTC)


 * At one time it was 5 or 6 repetitions. (Petrosian did not seem to have a good grasp of the rule.)  The game is covered in My Great Predecessors, vol 1, pp. 264-67, and all it says is "After checking to gain time on the clock, Capa makes a move ...".  If it was 5, Capa could have gained more time by doing it again.  But it isn't clear.  Perhaps Lasker didn't want a draw, I don't know.  Bubba73 (You talkin' to me?), 01:31, 12 April 2010 (UTC)
 * Lasker was playing for a win according to the game notes. Regards, SunCreator (talk) 01:47, 12 April 2010 (UTC)
 * Do you have a reference for that? It would be good to explain why Lasker didn't claim a draw in the article.  Bubba73 (You talkin' to me?), 03:44, 12 April 2010 (UTC)

An example from 1883
Brooklyn Daily Eagle, Feb 14, 1883, p. 2:
 * A Short Game and a Drawn Battle.
 * The sixth and last game of the exhibition series between Messrs. Steinitz and Mackenzie was played at the Manhattan Chess Club rooms on February 13, on which occasion the contest ended at the fifteenth move. Steinitz had the opening play, and for the second time selected his own gambit opening, which was played as follows:
 * WHITE, STEINITZ   BLACK, MACKENZIE
 * 1. P-K4   P-K4
 * 2. Kt-QB3 Kt-QB3
 * 3. P-KB4  PxP
 * 4. P-Q4   P-Q4
 * 5. PxP    Q-KR5 (ch)
 * 6. K-K2   Q-K2 (ch)
 * On his seventh move Steinitz played his king to KB3 and Mackenzie varied his play in the second game by moving Q to KR5 instead of Kt to KB3 as he did in the same opening of the seconf game of the series. Steinitz then went back to K2, and Mackenzie again checked him at K2. This time, however, Steinitz varied his play by moving his king to KB2.  His opponenet again checked him, and they moved as before until the 15th move had been completed, when a draw was declared, as Steinitz would not change his play and neither would the Captain.

If I understand the moves correctly, I think under modern rules either player could have claimed a draw at any point starting with Steinitz's 10th move. Quale (talk) 07:49, 20 August 2012 (UTC)

Eightfold ... No, Twelvefold Repetition required?
I seem to remember a chess column that explained the rule and offered an elaborate scenario like this (I’ve attempted to reconstruct.)
 * Player A: I claim a draw. The same position occurred at [shows scorecard], moves 21, 32 and 44.
 * Arbiter: [Analyzes.] Yes, but at Move 21, your opponent had the right to take your queen bishop pawn en passant, which right was obviously gone by Move 32.  So you only have two repetitions.  Keep playing.
 * Player A: [Later.] All right. Move 59 and the same position for the third time!
 * Arbiter: No. You castled on Move 46, and laboriously “uncastled” after that. This is the first occurrence for this particular dynamic position.  Keep playing.
 * Player A: [Later.] Okay, we repeated the position on Move 61 and then on Move 73.
 * Arbiter: No, your opponent castled on Move 62 and then inexplicably uncastled. This is the first occurrence for this particular dynamic position.  Keep playing.
 * Player A: [Later.] We have repetition on Move 77 and again on Move 90.
 * Arbiter: Okay, you have a draw.
 * Player B: Wait a minute: within the last ten moves, I swapped the position of my knights.
 * Arbiter: Doesn’t matter. The position is statically and dynamically the same.  Drawn.

It seems that the position may need to occur as many as eight times in the worst case before a claim is valid. Can anyone improve upon that? (I wouldn't recommend putting this in the article, but we should have an answer ready of anybody asks for the worst case.) WHPratt (talk) 17:18, 9 December 2012 (UTC)


 * Yes, that seems right. I think the point he making is about how the rights to en passant and castling change it, so they appear to be the same position but they are not considered the same because of e.p. and castling.  But that does seem to be the maximum number, if someone asks.  Bubba73 You talkin' to me? 17:48, 9 December 2012 (UTC)


 * I was wrong! Each side can lose a castling privilege twice.  In an extreme case, each side moves a rook and loses the right to castle on one side, then moves the other rook or the king and can no longer castle on the other side.  Even if the rooks and the kings end up back to where they were, these can be false positives re repetition.  In a worst case, you could have no draw until the tenth repetition! WHPratt (talk) 23:09, 9 December 2012 (UTC)


 * Castling is already in thee twice. Bubba73 You talkin' to me? 23:33, 9 December 2012 (UTC)


 * It should be in there four times in the worst case.
 * 1st Rep: Assume that either player may castle on either wing. Also, an e.p. capture is possible.
 * 2nd Rep: Same position, but no e.p. possibility.
 * 3rd Rep: Same, but only the last two count.
 * 4th Rep: Same, but say that White had moved his king's rook away and back. The count is at 1.
 * 5th Rep: Same, but say that White had moved the king or the other rook away and back. Count remains at 1.
 * 6th Rep: Same, but say that Black had shifted his queen rook. Count remains at 1.
 * 7th Rep: Same, but say that Black had moved his king or other rook, Count remains at 1.
 * 8th Rep. Same, positionally and dynamically. No way around that. Count goes to 2.
 * 9th Rep. Same, positionally and dynamically. Count goes to 3. Now a draw claim can be sustained.

I think that nine is max. Can anyone top this? WHPratt (talk) 23:50, 9 December 2012 (UTC)


 * Right - each side can lose a castling right twice. But I think you undercounted.  Between 4th and 5th above, there can be another rep, getting the count up to 2 before it goes back to 1.  Same thing between your 6th and 7th above, right?  That would make it 11.  Bubba73 You talkin' to me? 00:03, 10 December 2012 (UTC)


 * also another rep between your 5th and 6th where the count goes up to 2 before going back to 1, total 12. Bubba73 You talkin' to me? 01:30, 10 December 2012 (UTC)


 * How many chess addicts does it take to count to 12? Okay, this should do it:
 * In the key position, assume that at every checkpoint, the pawn and piece placement is exactly the same.
 * 1st Rep: The key position arises. Assume that at this point, either player may castle on either wing. Also, an e.p. capture is possible. Besides the obvious, potential moves include: e.p. capture, 0-0, 0-0-0, ...0-0,  ...0-0-0.
 * 2nd Rep: Same position, but no e.p. possibility.  (There can be no more e.p. captures fro0m now on, as obviously, no pawns have been moved since the last time.)Potential: 0-0, 0-0-0, ...0-0,  ...0-0-0. Count is 1.
 * 3rd Rep: Same as above. However, but only the last two count, so Count is 2.
 * 4th Rep: Same, but say that White had moved his king's rook away and back. Potential: 0-0-0, ...0-0, ...0-0-0. The count is back at 1.
 * 5th Rep: Same as above. Count goes to 2.
 * 6th Rep: Same, but say that White had moved the king or the other rook away and back. Potential: ...0-0, ...0-0-0. The count is back at 1.
 * 7th Rep: Same as above. Count goes to 2.
 * 8th Rep: Same, but say that Black had shifted his queen rook.Potential: ...0-0. The count is back at 1.
 * 9th Rep: Same as above. Count goes to 2.
 * 10th Rep: Same, but say that Black had moved his king or other rook. Potential: no more castling need be considered. The count is back at 1.
 * 11th Rep: Same, positionally and dynamically. No way around that. Count goes to 2.
 * 12th Rep: Same, positionally and dynamically. Count goes to 3. At last, a draw claim can be sustained!
 * 12th Rep: Same, positionally and dynamically. Count goes to 3. At last, a draw claim can be sustained!

WHPratt (talk) 01:37, 10 December 2012 (UTC)


 * I agree. Bubba73 You talkin' to me? 01:45, 10 December 2012 (UTC)

Could it be stretched out further by triangulation? Like 11th rep: Same, but other guy to move. 12th rep: Same, but with original guy to move. (2) 13th rep: Same, but with other guy to move. (2 of those) 14th rep: Same, with original guy to move. (3, draw)

67.61.185.177 (talk) 14:23, 27 June 2013 (UTC)
 * That would make the same position come up with the other player to move. Bubba73 You talkin' to me? 18:22, 27 June 2013 (UTC)

Max Euwe
"The sequence has been discovered independently many times, not always by professional research mathematicians; for example, Max Euwe, a chess grandmaster, who held the world championship title from 1935 to 1937, and mathematics teacher, discovered it in 1929 in an application to chess: by using its cube-free property (see above), he showed how to circumvent a rule aimed at preventing infinitely protracted games by declaring repetition of moves a draw." From the Wikipedia article Thue-Morse sequence. - Eroica (talk) 07:58, 24 November 2018 (UTC)

It's not called "threefold position"
It's "threefold repetition, or in full "threefold repetition of position". It's probably better to paraphrase than to copy/paste FIDE laws. The latest revision did not actually change the rules, just made them clearer. MaxBrowne2 (talk) 11:42, 5 June 2020 (UTC)
 * You are correct on all points: the term is "threefold repetition", and copy/paste of FIDE laws into Wikipedia articles is not a good idea. Quale (talk) 08:05, 7 June 2020 (UTC)


 * I agree. I brought the issue of copying so much directly from FIDE up at Copyright problems/2020 June 4, but so far only one person has put in an opinion.  Bubba73 You talkin' to me? 16:35, 7 June 2020 (UTC)

Too much copied directly from FIDE
You are taking too much directly from FIDE, see Copying text from other sources and Copyright problems/2020 June 4. Bubba73 You talkin' to me? 13:36, 8 June 2020 (UTC)

Few obervations
I find it a bit confusing to overall start mentioning the 1972 World Championship match first, but then starting the examples with the 1971 Candidates Final Match for it. The examples must not in chronological order, but for the first two examples, it would be better. The diagrams sometimes are mentioning the location, sometimes not. I don't find the location of primary interest; I suggest to remove the location. In the game Hillsbury versus Burn, the first threefold repetition first occurs after 50... Kg7 but the diagram only mentions the second threefold repetition after 51.Qb2. These are just some vague comments, in case this was already discussed, please let me know. Dlbbld (talk) 06:10, 1 July 2020 (UTC)


 * I don't see any benefit to have them in chronological order. The 1972 championship is mentioned first because it had three such draws, which was unusual.  But none of them are illustrated with diagrams or moves.  Then the 1971 match shows a good example.  There are reasons for Capablanca-Lasker and Alekhine-Lasker to be later, because they illustrate different aspects of the rule. And Pillsbury versus Burn is later, under History, because it shows how the rule was different at that time.  Bubba73 You talkin' to me? 06:36, 1 July 2020 (UTC)


 * Thanks for the feedback. It's just that Fischer qualified in the second listed event 1971 Candidates Final for the first listed event, the 1972 World Championship. But the article does not connect the events. For the average readers that might not be clear, and I find it worth mentioning. For pointing this fact out, the chronological order would be more natural. As there are other criteria for the order, in this respect the order is ok. But I would mention it. Agree, for the other examples chronological order is not of importance, as the order is for illustration. Dlbbld (talk) 08:45, 1 July 2020 (UTC)


 * World Chess Championship 1972 gives the history. This article is about the three-fold repetition rule, with examples.  Bubba73 You talkin' to me? 21:00, 1 July 2020 (UTC)


 * Ok, I can agree with this reasoning, thanks for the comment. Dlbbld (talk) 18:36, 4 July 2020 (UTC)

Adams versus Ponomariov, 2005
Per my view there is a typo here, taking into consideration the article text, as the title refers to the match in Wijk aan Zee, where Ponomariov played the white pieces. However, in 2005 there was another match in Sofia between the two. From the game, this must have been threefold repetition. This time Adams played the white pieces. Any ideas to consider or not consider both matches? Dlbbld (talk) 08:54, 1 July 2020 (UTC)


 * I don't see any reason to consider both matches. Bubba73 You talkin' to me? 18:30, 2 July 2020 (UTC)


 * Yes, that would overload the article. Also the example is about time trouble as the source confirms. To add the second game as such would not make sense, being most likely not time trouble related. Dlbbld (talk) 18:44, 4 July 2020 (UTC)

Another use
I don't have a reference for this, but it is said that in tournaments or matches with a "no agreed draws before move 30" type of rule, that the players will repeat a position if they want to draw earlier. Bubba73 You talkin' to me? 22:06, 2 July 2020 (UTC)


 * I also heard this, but without example. Notably tournaments cannot deactivate the rule in this "no agreed draws before move 30" case, as the rules leave no option for doing so for 9.1.1 Dlbbld (talk) 06:56, 5 July 2020 (UTC)


 * On spot checks, this is very much the case per my opinion. If this is regularly performed or these are just peculiarities I don’t know. Even with a source stating as such, I would not add it to the article. For me mentioning such a topic would stray too much from the main topic. For example, all the following games played in 2019 have a threefold repetition with the last move, so potentially ended by a draw claim, which was non-forced per my opinion for any of the games:
 * Vachier-Lagrave vs Ding in Norway Chess Tournament ended already with the 20th move of Black (20...Qg6 18...Qg6 16...Qg6 is a threefold repetition).
 * Karjakin vs Nepomniachtchi in Sinquefield Cup ended already with the 21st move of White (21.Nb5 19.Nb5 17.Nb5. is a threefold repetition).
 * Grischuk vs Giri in Gashimov Memorial ended already with the 22nd move of White (22.Kf1 20.Kf1 18.Kf1. is a threefold repetition)
 * But these are only speculations. This topic, for me, is too far away from the main topic. I have never ready any discussion to amend the rules to disallow such draws, which should eventually arise if this is a problem. As already mentioned, even with a minimum move restriction for general draw claims, the rules explicitly still allows draw claims based on threefold repetition with no restriction to the move number. Dlbbld (talk) 14:43, 9 July 2020 (UTC)

In the opening
I find the example from the Giuoco Piano line not suited. I only found one source from vanity press leading to this position, as added.

The line is "1. e4 e5 2. Nf3 Nc6 3. Bc4 Bc5 4. d4 exd4 5. c3 Nf6 6. cxd4 Bb4+ 7. Nc3 Nxe4 8.O-O Bxc3 9. d5 Bf6 10. Re1 Ne7 11. Rxe4 d6 12. Bg5 Bxg5 13. Nxg5 O-O 14. Nxh7 Kxh7 15. Qh5+ Kg8 16. Rh4 f6 17. g4 Re8 18. Bd3 Kf8 19. Qh8+ Ng8 20. Bh7 Kf7 21. Bg6+ Kf8 22. Bh7".

As such per my understanding, this is a very exotic line and as such, not suited for an example. In chess365.com for the Giuco Piano line after 20...Kf7 there are two games. For the Pirc Defense line, after 9.Ng5, there are several hundred games. As such, for the example with the Giuco Piano line, a justification should be found to keep, removed or replaced with a more common line. I think one example is enough, the whole article is already on the longer side for the definition of a term. Dlbbld (talk) 15:08, 11 July 2020 (UTC)


 * One of the references for the Giuoco Piano line was Modern Chess Openings. Bubba73 You talkin' to me? 22:09, 13 July 2020 (UTC)

Since then no reference was found for Giuoco Piano; also no other opening example added. As such changed general title "In the opening" to more specific title "Opening line". If this statement is generally true, somebody can find an example and revert. Dlbbld (talk) 15:29, 8 September 2020 (UTC)


 * I think the Giuoco Piano three-fold repetition is (or was) in ''Modern Chess Openings", and also elsewhere, I think. Bubba73 You talkin' to me? 22:07, 8 September 2020 (UTC)

There is *no* consensus to give fivefold repetition (or seventy five move rule) such a prominent place in the article
It was not "discussed and agreed" as claimed by that fivefold repetition should feature so prominently in this article. The so-called "fivefold repetition rule" is barely even a "thing", how often do you even see this specific combination of words in reliable sources? The main source for the rule is the FIDE website and it is rarely featured in any other RS. This is a clear example of WP:UNDUE weight and WP:POV pushing.

In particular the quote under WP:PROPORTION is relevant here: "'An article should not give undue weight to minor aspects of its subject, but should strive to treat each aspect with a weight proportional to its treatment in the body of reliable, published material on the subject.'" Fivefold repetition is not featured prominently in any reliable sources, so wikipedia should not feature it prominently either. Consensus has not been achieved that it should get a whole new section, let alone a standalone section after a section on repetition rules in games other than chess, and this editor's apparent obsession with the rule is frankly eccentric. Nor is there consensus that fivefold should be treated as some kind of distinct and unrelated thing with respect to the threefold repetition rule, a contention that flies in the face of common sense regardless of how FIDE's rules are worded or organized. All this has been explained many times, only to be met with massive WP:WALLSOFTEXT, WP:IDHT and attempts at intellectual one-upmanship.

All of these comments apply equally to the "seventy-five move rule".

I accept the consensus (but mildly disagree with it) that fivefold repetition and seventy-five move rule should have entries in the glossary. They could maybe have small subheadings within the rule explanations but in my opinion even this is undue weight, in view of the low level of coverage these rules actually get in reliable sources. MaxBrowne2 (talk) 00:06, 30 July 2020 (UTC)
 * I don't know that there was a consensus, but I think this was a good-faith attempt at reaching a compromise.  wanted separate standalone articles,  wants no mention of these new FIDE rules anywhere in wikipedia at all.   thinks that the rules aren't important enough (not enough WP:RS sources to demonstrate WP:N) for a standalone article, but they are significant enough to deserve brief mention in rules of chess, glossary of chess, threefold repetition and fifty-move rule.  It would be undue weight to write a lot about these rules, but really the requirement for WP:RS reliable sources practically ensures that can't happen.  As far as I know, there just hasn't been much written about these rules.  So perhaps I was the only one satisfied with the compromise since for me it wasn't a compromise at all.  I think chess coverage in Wikipedia is better if the new rules are mentioned, as long as it is in the appropriate places and given the appropriate weight.  Four sentences in this article did not seem to me to be undue weight.  The wikitext in the edit window looked much longer because inline cites in wikipedia are very bulky, but all that should matter here is how the text renders to readers of the article.  I actually think https://en.wikipedia.org/w/index.php?title=Threefold_repetition&oldid=967547043 which gives 5-fold repetition a small section to itself near the end of the page was the better approach.  Dlbbld tried that, but Max rejected it.  Quale (talk) 00:35, 30 July 2020 (UTC)
 * "MaxBrowne2 wants no mention of these new FIDE rules anywhere in wikipedia at all. " Not correct, I want them to be treated with an appropriate level of coverage. Given that they don't get very much coverage in reliable sources, the appropriate level of coverage is "not very much". MaxBrowne2 (talk) 00:39, 30 July 2020 (UTC)
 * Well, I don't want to misinterpret your complaint. I think since you kept the text on 5-fold repetition but moved it into another section perhaps you think five-fold repetition doesn't deserve a separate subsection?  I think it's a little out of place under "Statement of the rule" because 5-fold repetition is a different rule than 3-fold repetition.  Obviously they are very closely related, but they are different rules, so a small separate section seems appropriate to me.  But since much of the rule is the same (namely the definition of a repeated position) it also makes sense to keep it close to the definition of 3-fold repetition.  I guess I could convince myself either way and don't have as strong an opinion about it as you do.  Quale (talk) 01:52, 30 July 2020 (UTC)
 * I am saying something on my own here. There is a lot of talk about WP here and WP there. For everything, one can find a WP, that is even stated. Particularly, MaxBrowne2 mentioned that I am not following WP here and there. That is one thing and I tried to argue where appropriate. But I remember MaxBrowne2 here of Please do not bite the newcomers and claim, that he himself did not follow this rule.That I am relatively new also does not mean I cannot request a new article, I am neither new to chess nor markup language and made substantial corrections to several chess articles. Dlbbld (talk) 14:53, 10 September 2020 (UTC)

I am returning to this discussion after seeing fivefold mentioned recently in Talk:Chess. I do not disagree that fivefold is not a very important rule. But since there is a redirect for Fivefold repetition that sends you to Threefold repetition, and at least one article (Chess) has a wikilink to that redirect, the helpful thing for that redirect to do would be to send you to the brief section of Threefold repetition that talks about fivefold, rather than sending you to the head of the article. We could add a subsection header, or if people would prefer, we could add an anchor without adding the actual header text. Bruce leverett (talk) 16:07, 14 July 2023 (UTC)
 * Redirect to a subsection sounds like a good option. MaxBrowne2 (talk) 21:20, 14 July 2023 (UTC)

General statement of the rule
I have changed the wording in this version. I see the following problems:

Dlbbld (talk) 21:42, 8 September 2020 (UTC)
 * The article exposition is unclear. Should this be an article about the threefold rule at a general level? Then the introduction details do not make sense, for this is paraphrasing a rulebook. But paraphrasing a rulebook as now and not mentioning (FIDE, USCF or whatever) leaves me unclear if the intention of the article is the threefold repetition in general (whatever that means) or the rule form the rulebook.
 * "In chess, in order for a position to be considered the same": There is a general use of "position" in chess, this is the visual position on the board. The rule specifies position specifically for the rule and this makes only sense for the rule itself.
 * "each player must have the same set of legal moves each time": This is the rule intention which results in factual conditions (turn, en passant, castling). The article mentions the conditions, so must not mention the rule intention like a condition.
 * "including the possible rights to castle and capture en passant": This wording suggests that there is a "right" to capture en passant in the same sense as there are castling rights. This is not the case. Using "at the start of the sequence a pawn could have been captured en passant" from the rules is correct.
 * Paraphrasing of the rules leads to a long paragraph. What about omitting the paragraph and keeping it with rule citation?


 * I'm against omitting the paragraph. The full rule should be in the text, not a footnote.  You might use a footnote explaining when positions are considered the same.  Bubba73 You talkin' to me? 22:11, 8 September 2020 (UTC)


 * I mentioned FIDE in the paragraph paraphrasing the rules so the article in this respect is now consistent for my understanding. Dlbbld (talk) 14:57, 10 September 2020 (UTC)

Wrong paraphrasing of the rule
Since 11th September, the paraphrasing of the FIDE rules for position repetition is wrong.

The FIDE rules under 9.2.2 give a general rule. 9.2.2.1 and 9.2.2.2 are specific. It is simplest to take 9.2.2 as an informal idea and not to interpret and take 9.2.2.1 and 9.2.2.2 as concrete guidelines. 9.2.2 on itself needs interpretation which leads to problems, but one can assume that 9.2.2.1 and 9.2.2.2 are what is actually meant, this is how I have seen the rules applied everywhere.

The wording "Two positions are by definition "the same" if the same types of pieces occupy the same squares, the same player has the move, and the same set of legal moves are available to the player taking into account the castling and en passant rules.""

goes wrong for the "legal moves". Two positions where castling is temporarily disabled and the castling right changes have the same set of legal moves but are not considered the same per 9.2.2.2, this is where the wording goes wrong. A wording like "the same set of current and future legal moves" would be correct but nobody would understand. I find it best to directly refer to 9.2.2.1 and 9.2.2.2 and avoid any general statement. Dlbbld (talk) 16:11, 12 September 2020 (UTC)
 * If you think you can improve the wording go ahead, but please don't quote FIDE paragraph numbers or make fine technical distinctions because honestly nobody gives a fuck. (Yes I'm weary of your constant nitpicking). The point of the lead section is to summarise the article in language ordinary people can understand. See MOS:LEDE. MaxBrowne2 (talk) 17:02, 12 September 2020 (UTC)
 * I have simply explained what I see as a problem. It is not nitpicking. If you talk of apples you talk of apples. If you mean oranges and don't say so that is your problem disregarding how aggressive you try to put it on someone else's back. Dlbbld (talk) 17:16, 12 September 2020 (UTC)

Lead sentence
I recently received this statement in an edit summary from IHTS:

" 'Threefold repetition' fundamentally & specifically a shorthand reference & meaning about the rule. (And 'rule' is no longer a word in the lead sentence after this change.) MOS:FIRST supports that 'rule s/ be included in the lead sentence. (What part of MOS:FIRST supported this change?)"

To answer the question, the part of MOS:FIRST that I drew upon is as follows:

"If possible, the page title should be the subject of the first sentence."

In the previous version of the first sentence, threefold repetition was not the subject; the rule describing threefold repetition was the subject.

I think the statement "MOS:FIRST supports that 'rule s/ be included in the lead sentence" means that, if there is a rule declaring the existence of the article's subject, then it is preferable to include that rule in the first sentence rather than to include the subject itself. However, I could not find any part of MOS:FIRST that corroborates this. I'm pretty sure I'm reading this statement wrong, so I would appreciate clarification.

I am surprised to hear that the term "threefold repetition" is fundamentally meant to be shorthand for the rule. I have never heard someone use a construction like, "threefold repetition states that [...]". Alternatively, IHTS meant that the only reason threefold repetition matters is due to the rule that says what threefold repetition entails. That's true, but it doesn't explicitly make it impossible to make "threefold repetition" the subject of the first sentence. If it did, then we'd also have something like, "The checkmate rule states that a player loses when they have no legal moves and their king is in check."

The main concern here seems to be that the new version split off some important information into the second sentence. I agree that that information should probably be in the first sentence.

IHTS, I hope there's no hard feelings. I know tone is difficult through text, but I interpreted kind of an aggressive tone from your message, given its length and the intensity of its language. I'm hoping that wasn't your intention; I only want peace. ISaveNewspapers (talk) 06:39, 15 January 2023 (UTC)


 * My point was Threefold repetition implies the rule, that is the only reason we have a WP:CHESS article on it, and by that title. If there were no rule, there wouldn't be an article named 'Threefold repetition', 'threefold repetition' has no meaning outside the rule. The same can't be said re 'checkmate', which is so obviously a rule it needn't be stated. That's not the case w/ 'threefold repetition', so you're comparing apples & oranges. So the page title is the subject of the first sentence. By concluding otherwise you are dropping said implication, interpreting too literally. BTW I never suggested or implied the two phrases c/ grammatically supplant or be exchanged w/ one another in complete sentences. Cheers, --IHTS (talk) 07:22, 15 January 2023 (UTC)


 * p.s. My editsum was completely objective, so to get personal by stating on Talk page it was jeopardizing your personal sense of peace was uncalled for and is itself passive aggressive. Please stop. --IHTS (talk) 07:29, 15 January 2023 (UTC)

I agree with IHTS that "threefold repetition" is meant to be a shorthand; in other words, as a chess player, when I use the word "repetition", it is always and only to indicate that an invocation of the threefold repetition rule either is possible, or will be possible after one or two more repetitions.

For this reason I think that the present wording of the first sentence is satisfactory. I understand the distinction drawn by MOS:REFERS, but as with some other articles in the chess canon, such as Castling and En passant, this article is really about a rule and the consequences of that rule. Bruce leverett (talk) 03:58, 16 January 2023 (UTC)

Perhaps we could include a new example of how threefold repetition was AVOIDED in the recent 2023 world chess championship?
In the recent 2023 world chess championship match between Ding Liren and Ian Nepomniachtchi, game 18, the last game of the rapid time controls, saw an instance where Ding Liren, playing with the black pieces, was given an opportunity to draw the game by threefold repetition. Had he done so, they would have tied the rapids and proceeded on to the blitz time control games.

In the game, 44.Qe4+ Kg8 45.Qd5+ Kh7 46.Qe4+ was the first two instances of the repeated position. Had Ding Liren (black pieces) played 46...Kg8, that would be how he goes for a draw. Instead, Ding Liren plays the unexpected 46...Rg6! This puts the rook into a slightly awkward pin in order to advance the game: Ding wanted to keep going. This bold choice actually led to him *winning* game 18 and thus the World Champion title.

Primary source: https://www.fide.com/news/2365 Source was found from this wikipedia article. INSANITYISAVIRTUE (talk) 04:04, 7 September 2023 (UTC)

Succession
Why would the identical positions need not be in succession for a valid claim? I thought they could. 174.103.211.175 (talk) 23:20, 29 May 2024 (UTC)


 * I am not sure I understand your question.
 * "need not be" means that the identical positions do not have to be in succession. But of course, they can be in succession, and often they are. Bruce leverett (talk) 00:42, 30 May 2024 (UTC)
 * They can be in succession, but they don't have to. I think that basically no progress is being made.  You could take one course from the position, get nowhere, and get back to the same position.  Then you could try a different approach, get nowhere, and wind up back in the same position, for the third time.  Bubba73 You talkin' to me? 01:16, 30 May 2024 (UTC)
 * I wondered "why would being in succession invalidate the claim"? 2600:1008:B135:E6E5:504D:CB74:4273:E547 (talk) 19:10, 7 June 2024 (UTC)
 * I had never heard of this. 2600:1008:B135:E6E5:504D:CB74:4273:E547 (talk) 19:11, 7 June 2024 (UTC)