I've noticed that, whenever a player have so called "insufficient mating material", and his opponent loses on time, the "insufficient" player deserves only a draw, regardless of total number of pieces remained on board.
Why "only", shouldn't the "rule of drawing because of insufficient material" be applied?
Well, the answer, and the actual fact is "NO", if we respect the updated official rules by FIDE, which say that INSUFFICIENT MATING MATERIAL OF YOUR OPPONENT IS NOT ENOUGH FOR YOU TO NOT TO LOSE THE GAME (BUT DRAW), IF YOU RAN OUT OF YOUR OFFICIAL TIME!
What it means in practical play is that even if I have (as white), say, only a white-square bishop and my opponent have (as black), ONLY a black-square bishop, he LOSES (and NOT draws), if his time is up!
To illustrate that rule with the simplest position possible where such situation could occur, put the white king on a6 (or b6) square and white bishop on c6. Ok, now put the black bishop on b8 and the black king on a8.
It is mate, isn't it?
Note that, for example, the position with the same pieces involved, but where our opponent has a SAME (white) - SQUARED bishop, the game would be declared DRAW if our opponent runs out of time - now ANY matting pattern is NOT possible, king has enough squares available to avoid mate.
Keep in mind that, of course, our opponent could have additional material there - it does not change the matters if that material have zero influence on illustrated mating position. The point is - we got only a bishop ("an insufficient mating material", by old definition), and we WIN if our opponent spend all his(her) time, regardless of opponent's material.
So, here we have THE POSSIBILITY to mate an opponent even with a minimal material available on board - the rules are not concerned with questions like "would my opponent actually allow such a position to appear in real game continuation"- it is very clear about the fact that IT'S ENOUGH that such a mating position is POSSIBLE!
Any thoughts of implementation of "matting pattern" rule here?
You're right, and some slight changes might be done with simple endings (like K+B v K + xxx, K+N v K+ xxx); not sure if it's the top priority.
However, it's almost impossible to do it properly:
Imagine the following situation:
White: P b4, e4, h4, Ka1
Black: P b5 e5 h5, Ka8
According to the FIDE rules, it's a draw. But you won't find any engine capable of saying that it is.
@dougthehead, I'm afraid that the position from your example is NOT a draw according do FIDE rules IF, say black's time is up, although either of the kings cannot penetrate the "pawn barrier" and therefore it's an unavoidable theoretical draw. The reason is simple: there is more than enough mating material on board (3 pawns - and let's not forget that any pawn is always treated as potential queen), for, say, the white to mate black.
Even if white's pawns are treated as "only" a pawns, there are possible matting pattern with them alone. for example: White: Ka6, Pawns: b6, c7, d6; black: Ka8, Pawns: f7, g7, h7. Now, with white to move, 1.b7 is mate (of course, 1.c8=Q is mate as well).
Of course, in practical game the opponents would draw your position as soon as they reach it, in 99% of cases (unless one of the opponents is complete S.O.B. and wish to move aimlessly his king around and wait for his opponents' flag to fall before reaching the 50-moves rule).
But it should be clear that such "S.O.B." player would be - deservedly and clearly, by the rules; let's put aside the morality of his behavior aside for the moment - awarded with WIN if in that particular endgame such situation (i.e. his opponent runs out of time before reaching 50 moves from there) really appears.
Talking about engine's estimation, even though their estimation would probably be about 0.00 from the moment your position appears on, they cannot of course declare the game as draw for the same reason - if the time's up for any of the opponent before they 50-moves rule and they didn't agree on draw before that happened, he clearly loses (sufficient mating material).
#2 - In fact, the example position would be declared as win - if one opponent runs out of time, either by "old" ("sufficient mating material") or "new" ("possible mating pattern" - which could as well be seen as "helpmate") rules.
#1 That subject has been hotly debated, with many players arguing that we should do better than FIDE:
http://en.lichess.org/forum/lichess-feedback/no-mating-material-shouldnt-win?page=7#63That said, other endgames (including K+R vs K+R, for example Ka1,Rb1 versus Kd1,Rc1) might seem "obviously drawn" but by what algorithm can we declare which positions are "obviously drawn" and which ones aren't?
Thanks for the link, I wasn't aware the topic was already posted some time ago.
Should the topic be continued, or is it considered as "closed"? In my humble opinion, the matters very much deserves to be clarified, with a final verdict.
I believe now all goes down to decide "FOR or AGAINST implementation of FIDE rules?" They (FIDE rules) (as Thibould himself clearly explained as well) seem way much logical - because "Insufficiant material" is a confusing term: the point is, it IS sufficient, IF the opponent has some material left, HIS material can actually HELP him to be mated.
We can call it paaradoxical as we please, but it's the fact. Chess is a kind of game of paradoxes, as it's the life itself, and in the case of discussed topic, it's (chess') paradox is quite obvious.
FIDE Rules of chess 5.2.b
The game is drawn when a position has arisen in which neither player can checkmate the opponent’s king with any series of legal moves. The game is said to end in a ‘dead position’. This immediately ends the game, provided that the move producing the position was legal. (See Article 9.6)
Is there something you didn't understand in this article? It looks pretty clear to me thant the position I gave you is within its scope.
#3
"Talking about engine's estimation, even though their estimation would probably be about 0.00 from the moment your position appears on, they cannot of course declare the game as draw for the same reason "
and that is the Problem
White: P b4, e4, h4, Ka1
Black: P b5 e5 h5, Ka8
is a draw, because of the rule FIDE Rules of chess 5.2.b (read #7)
But it is not easy to "see" that neither side can make any progress. Sure, you can write a algo for that single type of endgame. But there are lot more specialpositions in a lot of endgame types. So you can not jugde automatic all of them. And bacause of that, in my opinion, it is better to judge KB vs KB or KR vs KR endgames like gentlemen would do. With a draw.
#7 - You are right, there IS such rule (Article 5.2, I checked abut it in "Chess Handbook", written by International FIDE Arbiter I happened to know personally), and if such position arises in actual game, game COULD be pronounced as draw, but there is, we call call it so, a little "twist" there (which is confirmed in a little interesting conversation I just had with another friend of mine, Bojan Birk, which also happened to be a FIDE arbiter).
The "twist" is following: A quoted, valid FIDE rule will come into force in practical game ONLY if the player whose time is about to be up very soon calls the arbiter and REQUEST the game to be called as draw.
If he FAILS to do so in the moment when his time is over, he LOSES by the mentioned "mating pattern" rule.
In other words, if player was previously unaware (or ignorant) of the existence of "Article 5.2." before his time is up, his ignorance of mentioned Article doesn't exclude him of losing by mating pattern rule if he fails to call the arbiter timely (i.e. before his time was up).
I apologize, I forgot to mention FIDE Arbiter's (author of "Chess Handbook) name, so it can be verified if needed; it's Eduard Piacun.
