Just the Rules: Some Half-Point Claim Mysteries Solved

Tim Just is a National Tournament Director, FIDE National Arbiter, and editor of the 5th & 6th edition of the US Chess Rulebook. He is also the author of My Opponent is Eating a Doughnut & Just Law, which are both available from US Chess Sales and Amazon/Kindle. Additionally, Tim recently revised The Guide To Scholastic Chess, a guide created to help teachers and scholastic organizers who wish to begin, improve, or strengthen their school chess program. Tim is also a member of the US Chess Rules Committee and his new column exclusive to US Chess, “Just the Rules” will help clarify potentially confusing regulations.

While the time on your clock approaches zero, you decide to make your move 39. That move will repeat a position for the third time (Triple Occurrence of Position). You record your move 39 and then pause your clock before claiming the draw (pausing the clock before making any claim is a good idea). Your opponent, playing the black pieces, points out that, since there was no perpetual check or a repeat of the position three times in a row that, the game must continue–there is no draw. It is time to get help from a TD. Before the director investigates the claim, he reminds both of you that a draw claim is also a draw offer. Your opponent tells the TD to investigate your claim before he decides on the draw offer. The TD correctly points out to your opponent that there is no “perpetual check” or “three-times-in-a-row” rule, so your claim will now be looked into.

After checking the score sheet, the TD rules that the claim was incorrect. Why?

The exact same position with the same pieces sitting on the same squares for both players (think of a photo) was verified, but all of the pieces did not have the same moves available to them as required by the Triple Occurrence rule. In the first two positions, on your moves 32 and 34, castling was available to your opponent. By move 39, the third position in the claim, your opponent’s rook had moved away from its starting square and then back again before the third position occurred, i.e., while all of the pieces for both players were on exactly the same squares, not all of those pieces were able to make the same legal moves. The game continues. The draw offer remains in effect until you make your announced move 39 and your opponent accepts/rejects the offer (see my “Draw Offer Blues” column).

Let’s take a peek at the two rules at work here:

Three positions exactly the same? You have the right to make a draw claim.

Imagine you are taking a picture of each position on the board when each move is determined. When three of those pictures are exactly the same, you may claim a draw by “Triple Occurrence of Position” (make sure that is how you word your claim). Make your claim on your move just before or just after that third snapshot is taken. Typically, the player making the claim is allowed to first write the move down on the scoresheet without making the actual move on the board before making the claim; however, if the move is made first, but the clock is then paused—not pressed, the claim can still be made. Remember, all the pieces have to be able to make the same moves (even castling) in each picture-perfect position. Your accurate scoresheet is often your best source of proof. However, under some circumstances a TD is able to observe or re-create the threefold repetition.

And the wrong way to make the claim: Make your move, press your clock (which starts your opponent’s clock), and then make your claim. The TD should deny your claim. You need to pause your clock —so that neither your clock nor your opponent’s clock is running—before you make your declaration; otherwise, it is not your move, so you have no right to make a claim. Oddly, even though this routine is not proper, the improper draw claim still doubles as a draw offer until your opponent rejects or accepts the draw.

Make a draw offer or make a draw claim: Either way, they are just the same.

Making any draw claim—even an improper one—is the same as making a draw offer. You can’t take it back. That means your opponent can accept a draw before the TD investigates and makes a ruling on your claim. And the draw offer, via your claim, is still in effect even if the TD rejects your claim.

Check out this example: The game has been long and hard. By your count, no pawn has been moved and no piece has been captured in the last fifty moves. As you make that last move to reach that magic count of 50 moves, you press the clock and make your claim. Even though you pressed your clock (a big no-no when making claims) instead of pausing it to make the claim, your opponent accepts your draw offer. The game is over, and you split the point with your opponent.

Here’s a closer look at that 50 move routine:

After fifty moves, a draw is proven if no pieces were taken, and no pawns have been moved.

For fifty moves, neither you nor your opponent moves a pawn or takes a piece. It’s time to claim a draw. Just make sure your score sheet is accurate. And, please remember to pause your clock when you make a claim, so that a fallen flag does not create chaos.

If you are in time trouble, you may ask the TD to silently count off those fifty moves for you. Most TDs will help you out with this task. Sometimes the TD will assign an assistant or a bystander to do the counting while they attend to other duties. Don’t rely on your clock’s move counter to make this claim. Remember, if a piece gets taken or a pawn moves, then the fifty-move count starts all over again.

Next time around we will explore how some of the rules automatically hand you a half-point without almost any effort at all on your part.


  1. For Triple Occurrence of Position, (in addition to castling capabilities) en passant capabilities must also be considered. If the first occurrence of the position had a pawn with the capability of an en passant capture, and the other two positions did not, then there has only been a Double Occurrence of Position and the claim is not valid. Correct?

  2. Correct. ALL features of the position must be the same for the claim to be accepted. The last time a draw was claimed to me based on triple occurrence of position I denied it because the first time the position occurred it was White to move, but the second and third times the position occurred it was Black to move.

    It should also be pointed out that if a TD or someone deputized by the TD is counting moves in a sudden death time pressure situation in preparation for a draw claim based on the 50 move rule neither player has a right to know how many moves have been counted until the 50 move threshold is reached (Rule 15F4c.) Many players do not appear to know this, and assume the person doing the counting will tell them how close to 50 they are if asked.

  3. In the case of castling in a triple-occurrence situation, the TD should also make sure that all possible FUTURE moves are the same. If, the first time, castling is temporarily not possible because the opponent’s bishop is attacking the f1 square, but the second and third times, castling is also permanently impossible because the king has moved (and then moved back) since the first occurrence, then the TD should deny the triple-occurrence draw claim.

  4. We were involved in a recent Chicago Industrial League Ches Match. Both players were were in time trouble and no longer taking notation as they were both under 5 minutes. With 13.7 seconds left on his clock, one of the players stopped the the clock, claimed 3 fold repetition and began moving the pieces around. The other player refused the draw. Is this possible under rules?

    • The CICL is neither US Chess nor FIDE rated, it appears, so there is no compulsion to rule as dictated by either rule set.

      That said, in the situation you describe, if US Chess rules apply, and the claim is not agreed, the TD may only award the draw on independent verification of a properly claimed threefold repetition. That verification is usually accomplished by direct TD verification.

      If I were the TD, and had verified the repetition in my mind when the claim was made, I would be inclined to award the draw.

      In the absence of direct observation, I would be disinclined to award the draw, particularly as the pieces had been moved by the claimant. Moving pieces on the board of record to demonstrate a claim is NEVER a good idea and is not contemplated by the rules. Ruling on the outcome of the game or even restoring the position often requires a balancing of equities after such an event. I would be inclined to balance equities against the position disturber as much as I reasonably could.

      • Per an independent witness, the player made the move while making the claim and then paused the clock. That witness was wondering if the player would realize that a valid claim was possible. There was some confusion over whether or not a player could make the move before claiming (not valid in FIDE) and whether or not a valid scoresheet was needed per 14C8 (the 14C9 exception was the one in force). Questioning what happened was quite understandable but, based on a witness who was a member of both involved clubs and was not on either team, the draw claim was valid. The final result was an agreed draw without the need for a ruling.

        It is situations like that which make Tim’s columns valuable.

        • Agreed, Jeff. Would only add that a lot of these scenarios are truly “You had to be there” situations. Context is key.

  5. It is not abundantly clear that what Mr. Smythe says is correct. This precise scenario was debated on the Forums, and the TD’s commenting did not reach agreement. The relevant rule is 14C, and the relevant portion of it reads,”…the position is considered the same if pieces of the same kind and color occupy the same squares and if the possible moves of all the pieces are the same, including the right to castle…”. In the example Mr. Smythe gives the possible moves of all the pieces are the same. In the first instance castling is not allowed because the King would have to move through check. In the second and third instances castling is not allowed because the Rook has previously moved.
    The reasons why castling is not allowed differ, but the salient point is that castling is not allowed in all three positions, so the possible moves of the pieces are the same. Does this then mean that “the right to castle” is the same in all three cases? Here is where the “rules lawyering” comes in. How exactly do you define a “right”? I’m not sure that Mr. Smythe is right here, but I’m also not completely sure that he isn’t.

    Since this is extremely unlikely to transpire in practice it’s basically just a theoretical exercise. However, since this is Mr. Just’s column I would be very interested to hear his take on this.

    • Scott Parker writes: “It is not abundantly clear that what Mr. Smythe says is correct. This precise scenario was debated on the Forums, and the TD’s commenting did not reach agreement.”

      If I recall correctly, the Forums debate wasn’t about that type of case. It was about an even trickier case, where neither the king nor the rook had ever moved, but the player was in check in such a way that moving the king was the only way out of check. (In other words, no interpositions or captures existed that would get out of check.)

      Now, that’s a LEGITIMATELY debatable question.

      (1) If you define “castling is permanently illegal” to mean “there does not exist a sequence of legal moves leading to the player castling”, you get one answer.

      (2) If, on the other hand, you define “castling is permanently illegal” to mean “either the king and/or the rook has moved previously”, you get another answer.

      In practice, chess engines probably use definition (2), because in general it would not be practical for an engine to generate all possible sequences of legal moves (not just single legal moves) and try to see if one of them involves castling later.

      Definition (1), however, is more satisfactory to the purist in me.

      But I would not blame any arbiter who decided to use definition (2). There might even be puzzle-like positions where it is difficult to establish definition (1).

      Obviously Tim Just is aware of such examples. Equally obviously, he didn’t want to clutter up his explanation with complicated cases, as this column is intended for relative “rules newbies” who may not have a good grasp even of the basics.

      This weekend in Chicago, for example, on board 1 in the last round, one of the two near-masters in the game was not even aware that the same player had to be on the move in each of the three occurrences! It took a lot of explaining by the TD, the organizer, and the opponent to set the player straight — and he still didn’t seem convinced.

  6. Mr. Whited,

    I’m not exactly sure what actually happened in the example you give. You say that a player claimed a draw, began moving the pieces around and the opponent refused the draw. However, you don’t claim to an opponent. You can offer a draw to an opponent, but you can only claim a draw to a TD. If what happened was that the player claimed a draw to a TD, the TD then informed the opponent that a claim of draw is an implied offer of a draw, the opponent rejected the offer, and the TD then ruled the draw claim valid, then yes, the player can move the pieces around. The draw having been declared by the TD the game is over. If the player claiming the draw began moving the pieces around before the TD had validated the claim, then no, it was not legal to do this.

Leave a Comment

  • Categories

  • Archives

  • Announcements

  • US Chess Press