I have just purchased, from USCF Sales, an interesting book entitledChess Handbook, Book for Arbitersby two FIDE Lecturers, IA Zoran Bojovic and IA Branislav Suhartovic.
It’s not an official FIDE publication, and it was published in 2017, so it may be one or two FIDE rules cycles behind the times, but it still has a lot to offer. It frequently challenges the reader to balance his thoughts between the letter of the rules and the FIDE Preface, the latter being roughly equivalent to U.S. Chess rule 1A. In fact, I think a good alternate title for the book would be “Food for thought for aspiring Arbiters”.
There are also several of what I like to call “arbiter puzzles”, which are similar to chess problems (chess puzzles), except that the central question is always “What should the arbiter do here?”. Here is one of my favorites:
To quote exactly from the book:
“In the diagrammed position … White to move overstepped the thinking time. What’s the result of this game?”
Have at it, everybody! What, indeed, is the result of this game?
I don’t see how stalemate of white is even possible. Note that the problem says white to move. If it were black to move, stalemate of white would be possible.
I see only a white checkmate on the horizon. But what’s the FIDE rule? Does white actually win because that is the only possible result? Can black win since he has mating material even though there is no possible sequence of moves leading to a checkmate for him? And while I’m sure White can’t win under US Chess rules, I’m not sure whether mating material means a win or not. It shouldn’t, of course.
Sorry, I ran through it fairly quickly. I looked at the option of Black pushing the b pawn past, but there is no way to get there without White being forced to make the capture. (Shift the White pawn to d2 and you can also have White being stalemated).
If White flags, White can’t win, so the only question is time forfeit loss or draw by rule. FIDE’s rule is that you can’t lose on time if the flagging player can’t even be “helpmated”. (There are positions, for instance, where the only way you could get checkmated is to promote a pawn to a Bishop, and maneuver it to block your King’s escape square).
It doesn’t quite say, “White is to move.” It says White overstepped. I suppose it could be the moment between when White makes his move, and he presses his clock.
Nope. The FIDE position is that white can’t win because his time has expired, and black can’t win because white can’t be checkmated. Therefore, draw.
This “mating material” business is a vague concept, and everybody seems to have a different definition. If you define it as “either a queen, or a rook, or a pawn, or two bishops, or a bishop and knight, or two knights versus at least one pawn”, then you open up a whole new can of worms, which (for example) U.S. Chess rule 14E deals with in a not-too-satisfactory manner.
FIDE wisely avoids the use of the term “mating material” altogether (I think – maybe somebody could correct me on this).
In this case (and in many cases), the U.S. Chess rule has the same effect as the FIDE rule, but the former is worded more vaguely so that people can argue about it.
I think FIDE (and U.S. Chess) should adopt some new terminology:
A position iswin-excluded for a playerin case there does not exist a sequence of legal moves leading to that player checkmating his opponent.
Then adead positioncould be defined as one that is win-excluded for both players (and hence drawn).
" … if a player does not complete the prescribed number of moves in the allotted time, the game is lost by that player. However, the game is drawn if the position is such that the opponent cannot checkmate the player’s king by any possible series of legal moves."
How many times does the phrase bolded above in blue appear in the FIDE rules? With my proposed definition, all of these rules could be shortened to " … However, the game is drawn if the position is win-excluded for that player."
While we’re at it, what would be a reasonable definition of “draw-excluded”? I say the following would work:
A position isdraw-excludedin case there does not exist a sequence of legal moves leading to either player stalemating the opponent AND there does not exist a sequence of legal moves leading to an eventual repetition of position.
The purist in me would like the rule to be that if a position is win-excluded for a player and is also draw-excluded, then the game is an automatic and immediate win for the opponent.
Another rule should state that, if a position is draw-excluded, both players should be prohibited from making or accepting a draw offer. Play it out to checkmate or resignation.
However, my practical side tells me that the FIDE rule is just fine the way it is.
Under your definitions, the position is both win and draw excluded for black. You did not included “out of time” as a criteria for being win or draw excluded, although presumably that is adequately covered under the time forfeiture rules. However, in this case it says “white to move overstepped”. That means at some point after black’s last move, white still had time. Under your proposal since black is both win and draw excluded, he immediately lost, because white still had time at that moment.
Aside from being rather pointless, isn’t that (logically) equivalent to all legal sequences of moves lead to a checkmate for one of the players–since chess has a finite number of possible positions, based upon the movement rules, you will eventually get checkmate, stalemate or a repetition.
As compositions go, this one is fine for beginners or even people who know little to nothing about chess. The moves are forced from beginning to end. A player looking at this is simulating real calculation. It is similar to other early compositions where the inferior side must move and allow a checkmate.
If White’s flag did fall, he should lose as his next move, the completed move, did not result in immediate checkmate. Black’s extra pawns could constitute mating material if they are able to promote, which can occur in this composition. What FIDE would rule is up to FIDE, but this is one of the instances where USCF rules and FIDE rules differ. I recall seeing a game from a Women’s Candidates event where each side only had a Knight and King and were told to play on as it was “possible” for a mate position to occur even though unlikely. There was some hubbub about this, but the game result, a loss for the player who ran out of time, was allowed to stand. The game was played without delay or increment.
In postmortems, one can always demonstrate a series of “forcing moves” that would inevitably lead to checkmate. But if your flag fell when executing move 5 of a twenty move sequence, you lose in a practical game even if you, your opponent, Fritz, and Alphazero can see the end. As long as we use clocks, these results are going to occur. In compositions, the clock is not a factor.
I strongly disagree, for two reasons (and the two reasons are at opposite ends of the spectrum, sort of).
First, it is absurd to declare a player lost if his opponent is win-excluded.
Second, in a more “normal” situation, if a player has mate on the move, he must actually play (i.e. determine) that move before his time expires, otherwise he loses on time. “Determine” in most situations means that the player must have actually executed the checkmating move and released the checkmating piece on its new square.
As I said upthread, I wish everybody would stop using the term “mating material”, since its meaning may vary depending on who is saying it.
Under FIDE rules a player definitely cannot win on time if his position is win-excluded.
I disagree again, but here there is room for reasonable people to come to different conclusions due to the poor construction and wording of U.S. Chess rule 14D, and specifically 14D4:
[b]14D. … The game is drawn when one of the following endings arises, in which the possibility of a win is excluded for either side: …
14D4. No legal moves leading to checkmate by opponent. There are no legal moves that could lead to the player being checkmated by the opponent.[/b]
The above wording first appeared in the 4th edition of the U.S. Chess rulebook.
14D talks about a win being excluded for “either” side, but then 14D4 switches horses midstream by talking about “the player” and “the opponent”.
If 14D4 had been intended to apply only when both players are win-excluded, then the words “by opponent” should have been omitted from the first sentence, and the second sentence should have ended with “… lead to either player being checkmated by the other”. The wording actually used convinces me that the intention was that 14D4 should apply also to the case where only one player (e.g. the opponent of a player whose time has expired) is win-excluded.
OTOH, to remove any doubt, 14D4 should have been worded differently, perhaps along the following lines:
14D4. No legal moves leading to checkmate by opponent. A player cannot lose, even if his time expires, if there are no legal moves that could lead to the player being checkmated by the opponent.
This was probably a poor ruling by the arbiter. FIDE has a rule allowing an arbiter to declare a draw if neither player is attempting to win “by normal means”, e.g. if one player is simply trying to run the opponent out of time. With K+N vs K+N, an arbiter could easily rule a draw because there are no “normal means” in such a position. Issues like this (perhaps even this exact case) are discussed in the book I mentioned at the top of the thread.