All these proposed ADMs

Running Chess Tournaments
1

Whatever wording goes into these various proposed ADMs, I hope their final effect is along the lines listed below.

Note: I have used the word “executed” throughout, but you could just as well substitute the traditional “determined” (as long as you don’t confuse it with the “decided” meaning) or the FIDE version, “made”.

  1. A “regular” move is executed when the piece has been released on its new square.

  2. A capture is executed when the capturing piece has been released on its new square and the captured piece removed from the board (in either order). Once the player has touched both the capturing piece and the piece to be captured, the player is no longer allowed to make any move other than this capture.

  3. A promotion is executed when the new piece has been released on the promotion square and the pawn removed from the board (in either order). A promotion which is also a capture is executed when the new piece has been released on the promotion square and the pawn and captured piece removed from the board (in any order). Once a player has caused the new piece to touch the promotion square, the player is no longer allowed to make any move other than promoting to that piece on that square. Once a player has touched the pawn, the player is no longer allowed to make any move other than promoting that pawn.

  4. Castling is executed when both the king and rook have been released on their new squares. Once the player has released the king on its new square, or has touched both the king and rook, the player is no longer allowed to make any move other than castling.

  5. A move is completed when the player presses his clock. A player is not allowed to complete the move until it has been executed.

  6. Exception: A move which results in checkmate, stalemate, or a dead position is completed at the same time it is executed.

The above approach combines the “once [ something has happened ] the player is no longer allowed” clauses into the same paragraphs as the definitions of “executed”. This approach, IMO, clarifies the whole picture and improves the organization of the material.

Bill Smythe

2

This is almost perfect. Ship it.

(The only reason that I say “almost” is that it isn’t the FIDE wording. I think it is actually better than the FIDE wording, but being the same as the FIDE wording is advantageous.)

3

I agree with Brian. This is a model of clarity, and it improves on the FIDE Laws of Chess by establishing a clear description of the time between executing a move and completing a move.

4

I like it. I think the only proposed ADM of mine that it would affect is my change to Rule 9D to say that if the pawn is still on the board when the new piece is released on the promotion square that the move is not yet determined but the player cannot promote to a different piece or on a different square. If the proposal in this topic becomes an ADM and it incorporates that change then there will be no need for me to submit a separate ADM for it.

I’ve previously said that using another word for “determined”, such as “executed” or “made”, is a good idea but I thought the changes would be so extensive that they should go into a new edition of the rulebook. Now that I’ve looked at it I don’t think it’s so bad. Most of the changes would go into Rule 9, currrently titled DETERMIMATION AND COMPLETION OF THE MOVE. Other sections affected, mostly in minor ways, are: 6B, 7A, 8F6, 10H, 10I, 13A, 13A2, and 35F4. Instead of including the full text of the proposed amendments to the rules the ADM might instruct the Rules Committee to make the necessary changes.

5

Are there any cases when an external event can change the game state before a move is determined? I’m thinking of clocks and checkmate or stalemate I suppose. If so, an ambiguity exists between 1 and 4 with the case where a player moves the rook first. That can be either a determined rook move, or an undetermined castling move. Note that this cannot occur if the king must be moved first, since a king move of one square cannot be a castling move, and a king move of two squares cannot be a regular move.

6

That’s probably better taken of in some other rule, such as the rule that describes castling.

The FIDE version requires that the king be moved first. USCF is a bit more lenient, but if the player waits a long time after moving the rook before he moves the king, it would be reasonable for the TD (upon request by the opponent) to disallow the castling. If, however, the player moves both pieces within a second or two of each other, it would seem contrary to reasonable sportsmanship to try to force him to move the rook only.

Bill Smythe

7

I agree!

Bob

8

There is, indeed, an ambiguity, but unless the USCF rules are tightened to match the FIDE rules, the ambiguity is real: it could still be either one. But in light of 5, what difference does it make to the other player? The other player may not move until the clock is pressed, at which point the move will have been executed.

Bob

9

FEN “8/8/8/4N3/3p1p2/3PkP2/8/4K2R w K - 0 1”

If white places the rook on f1, black claims stalemate. A regular move has been determined.

10

I don’t know what this line means.

That’s an interesting point.

First, I can’t see anywhere in the current rulebook where it says that a claim of a stalemate has any significance. So the claim by black would be irrelevant.

Second, rule 10I2 clearly says that “If a player intending to castle touches the rook first, there is no penalty except if castling is illegal, the player must move the rook if legal.” Being forced to accept a draw could certainly be considered a penalty.

But you make an interesting point. In the case of a move that results in either a checkmate or a stalemate, the player who made the move does not have to hit his button in order for the move to be completed (rule 9E).

If there is a case where moving Rf1 (or Rd1) would generate a checkmate or stalemate but O-O (or O-O-O) would not, then rule 9E, as worded, would be ambiguous, since it would not be clear after the rook move whether 9A applied (in which case the move would have been completed) or 9C applied (in which case the move would not yet have been completed). At the moment, I can’t envision a board position in which what I’ve just described would be true (though I have seen a board position in which O-O-O would be checkmate but Rd1 would not).

The first question is whether there even is a board position where Rf1 (or Rd1) would generate a checkmate or stalemate but O-O (or O-O-O) would not. If there isn’t, then the point is moot. If there is, then a player who moved his rook to f1 (or d1) in that situation would, IMO, have to indicate in some way that he intended for the move to be over in order for the game to be over.

Bob

11

Having given this some additional thought, I am now convinced that this couldn’t occur for a checkmate for the following reason:

Black, after his previous move, couldn’t have been in check (because of rules 12C and 12E), so his king must have been put into check by the moving of the rook. And since there is no way that a rook moving from h1 to f1 (or from a1 to d1) could reveal a check, then it must be the rook at f1 (or d1) that is checking black’s king, giving rise to the checkmate. But this would not be affected by moving the king to g1 (or c1), so O-O (or O-O-O) would also be checkmate.

So (to envision a worst possible scenario) if white moved his rook to f1 (or d1) and then his flag immediately fell, it should be ruled that he won by checkmate unless he gave some clear indication (verified by one or more impartial witnesses) that he intended the rook move to be part of a castle (e.g., touching his king after moving the rook).

Bob

12

FEN is a notation which can be used to specify board positions. I don’t know how to post diagrams here - I’d probably need to put the diagram on a web site so I could include the URL in an “img” command - so I’ll describe the position in words. White has a knight at e5, pawns at d3 and f3, a king at e1 and a rook at h1. Black has pawns at d4 and f4 and a king at e3. White can castle kingside (king and rook haven’t moved). It’s White’s move.

If White plays Rf1 it’s stalemate while if he castles it isn’t.

It’s hard to specify exactly when a castling move has been executed when the rook is moved first. If White moves his rook to f1 and then immediately moves his king to g1 he’s castled moving the rook first, while if he hesitates too long he’s played Rf1 and stalemated Black. But how long is “too long”? It’s a judgment call.

13

Having ruled out the possibility that this situation could occur for a checkmate, I have now envisioned at least one situation in which it could occur for a stalemate.

Imagine that black has nothing but a king, which is at b2. And imagine that white has a pawn at c2, an unmoved king at e1, an unmoved rook at h1, and a queen, which he moves to b3 (Qb3+). And imagine that black responds to the check by moving Kc1. Rf1, in this situation, would result in stalemate, while O-O would simply be check (or, if d2 were under attack by another of white’s pieces, checkmate).

As I noted in a previous post, a claim of stalemate by black after the moving of the rook to f1 would have no significance (other than to insure that white knew Rf1 would be stalemate), since the rules do not recognize claims of this kind.

While it would be legal under the rules for white, if he wanted to stalemate, to move Rf1 without punching his clock, it would be somewhat surprising if he wanted a stalemate when a win was clearly possible, and in any case he would have no argument with black’s claim, if that were his intention.

It would be far more likely, in this situation, that either (a) white was in the process of castling and had not yet completed his move, or (b) white intended to move Rf1 without realizing that it would result in a stalemate, in which case this would become clear when he hit his button.

So in this situation, I would not rule that moving the rook to f1 constituted a stalemate unless (a) white agreed it was a stalemate or (b) white had pressed his button without moving the king, either of which would indicate that his intent had been to move Rf1. If black objected that he had called stalemate after white moved the rook to f1, I would explain to him that a claim of stalemate has no signficance under the rules, and that white would not have moved Rf1 unless he had intended to stalemate (in which case he wouldn’t be disputing the claim) or unless he hadn’t realized it would be stalemate (in which case this would have been confirmed when he hit his button without moving the king, if black had kept his mouth shut).

Bob

14

Forsyth-Edwards Notation (Wikipedia) is a notation for describing a chess position in text. The first part describes the contents of each rank, starting with the eighth and working back to the first. Each rank is separated by a slash. In the given example, the 6th, 7th, and 8th ranks are empty (“8” means “8 vacant squares”). The fifth rank has four vacant squares (a5 through d5), then a white knight (uppercase “N”) on e5, then three blank squares. The fourth rank has three vacant squares (a4, b4, c4), then a black pawn (lowercase “p”) on d4, one blank square, a black pawn on f4, and two blank squares (g4 and h4). And so on …

At the end, the “w” indicates “white to move”. The “K” means “white may castle kingside, black may not castle.” (The letters “K”, “Q”, “k”, and “q” are used. If neither side may castle, this field contains a hyphen.) The “-” means “no en passant capture possible”. (Otherwise, that position would contain the letter of the file containing the pawn that could be captured en passant.) The “0” indicates the number of half-moves (“plies”) since the last capture or pawn move, and the “1” indicates the move number. Often, when constructing a position in software, the move number will be 1, with no previous plies.

15

The point is that according to #6 above, it’s the execution of the move that immediately ends the game (for the stalemate situation shown in the FEN). Anything the player does after that is immaterial, including moving his king. Since Rf1 is a normal move execution, the game would be over at that point.

Since I (personally) like the FIDE “move the king first” rule, my inclination would be to declare the game a draw, and tell the player who made the move that perhaps he should choose the unambiguous move order. TD judgement call.

Along these lines… what was the original intent of the “immediately ends” wordings for these situations? If the player had to hit the clock in the situation above, this ambiguity wouldn’t exist.

16

Sorry. I hadn’t seen your post when I started my last post. But now we have two scenarios! :slight_smile:

In either scenario, if white moves his rook to f1, there are 3 possibilities:

  1. White intended to castle, but hasn’t yet finished his move.

  2. White intended to move Rf1, and didn’t realize it would be stalemate.

  3. White intended to move Rf1, and did realize it would be stalemate.

In case 3, white will not disagree that it is stalemate, since white intended to stalemate.

In cases 1 and 2, the situation will be resolved when (a) white moves his king, completing the castle, or (b) white presses the button on his clock, establishing that the move was Rf1.

The only time I can envision a problem would be if white moved Rf1, intending to stalemate because his flag was about to fall, and didn’t stop the clock, and black waited for the flag to fall and then tried to claim a win on time, claiming that white was really in the process of castling. But that is covered by rule 14A3.

Bob

17

The problem with citing rule 14A3 is that if you’re saying that the Rf1 move hasn’t been determined (decided, executed) until White presses the clock, then it isn’t stalemate until White presses the clock, so 14A3 doesn’t apply and White should lose on time.

It’s a difficult situation to describe with a precise rule. FIDE gets around this by saying that the player must move the king first when castling. If USCF rules say that the rook can be moved first then I think the best we can do is to say that it’s up to the TD to determine the player’s intent. In this situation if the TD believes that White intended to play Rf1 then he should rule the game a draw by stalemate (9E - no need for White to press or stop the clock), and if the TD believes that White intended to castle he should rule that White loses on time.

18

If #6 were to be a replacement for rule 9E, this would be valid. But the actual wording of rule 9E is:

“In the case of a legal move which produces checkmate (13A, 4A) or stalemate (14A), the move is determined with no possibility of change upon release as described in 9A, 9B, 9C, or 9D, whichever applies. The move is completed simultaneously with its determination.”

So to enforce rule 9E, the TD must determine which of the four rules, 9A, 9B, 9C, or 9D applies. In this case, if the player was moving Rf1, the move was completed when he released the rook on the f1 square. But if he was moving O-O, the move would not be completed until he released the king on the g1 square.

If black wants it to be statemate, he needs to wait. If white intends to move Rf1 and doesn’t realize this will be stalemate, he’ll hit his button after making the move. If he intends to castle, he’ll move the king.

Situations like this are presumably why the FIDE rule was written the way it was. And if you are running a tournament and like the FIDE rule better, feel free to use it! But announce in advance that you are planning to use it, or else you will be violating rule 1B2. Keep in mind that players who have developed the habit of moving the rook first may not want to participate in your tournament if you use the FIDE rule.

Otherwise, remember rule 10I2: “If a player intending to castle touches the rook first, there is no penalty except if castling is illegal, the player must move the rook if legal.” Being forced to accept a draw could certainly be considered a penalty!

I’m guessing it was so that a player who checkmated or stalemated his opponent shouldn’t lose on time because he forgot to hit his button due to the game being over.

But it’s not clear to me why this is a legitimate concern. If a player who is about to checkmate his opponent picks up his checkmating piece, but his flag falls before he can release it on its destination square, he loses. So why should it be any different if it falls after he releases the piece but before he has time to reach over and hit his button? As for forgetting, he could forget to hit his button after any move, so why should this one be any different?

Bob

19

I (or, really, the USCF rulebook) was saying that in the case of a stalemate, the Rf1 move (if that’s what it is) has been determined as soon as white releases the rook on the f1 square.

What black should do is:

  1. If white’s flag falls before white touches his rook or after he touches it but before he releases it on the f1 square, black should immediately call the fallen flag.

  2. If white has released his rook on the f1 square and then his flag falls, black should wait to see whether white touches his king, and if he does black should immediately call the fallen flag.

But this still leaves some interesting possibilities. For example, white might be intending to castle, but after releasing the rook on f1, he might glance up and see his flag fall and simply leave the move as Rf1, and black would have no way to prove that white had originally intended to castle, so white would draw instead of losing.

Agreed.

Bob

20

My reading of the current rules is that the same problem exists, which is why I brought this up in the first place. If people want to rewrite wording on this stuff to make things more clear, I was hoping we’d take the opportunity to remove ambiguity. I don’t think TD’s should unnecessarily be placed in the position of having to judge intent.