Draws Declared By Remaining Mating Material Rather Than Possibility Of Checkmate Is Illogical

"hey we'll just make kbnk a draw coz no one knows it, its an edge case"

jk but still

knnk has a mating position

so why is knn insufficient?

it literally passes all your rules!

He said the site finds it easier to just count the material on one side. I'd argue that it's just as easy to do it by FIDE, since only 3 or 4 more piece combinations need to be programmed in, and if they really didn't want to program it, I'd suggest an even easier system than either of these. The side running out of time only gets a draw if the opponent has a lone king.

even IF we count for one side, knn should be allowed

thats my point

it is illogical, even under chess.com logic, for knn to be insufficient

Exactly, KNN is sufficient even if the opponent has a lone king, so in any position with a king and 2 knights should be a win. Since when does it depend on I'd it's forced or not (and it can be forced if there's a pawn on the board) but with their logic why should we have to keep playing KQKQ since a win usually can't be forced? Makes absolutely no sense even from their own reasoning.

This is the only type of position where 2 knights wouldn't be able to win, and here there are pawns on the board so it might be void anyway, but still, this shows the flaming contradiction. In a position where 2 knights can never win such as this, if white loses on time white loses because black has pawns that can't do anything ever, but in a position like below, where white has mate next move:

Black can let his time run out and get a draw out of a forced mate, no logic or consistency whatsoever.

It's amazing how many people don't understand this lol, if you got your opponent down to a king and minor piece and you only have a long king, 1st of all that would immediately trigger insufficient mating material draw. 2nd, even if it didn't and it let you play it out, both players have secured a draw since they can no longer lose, given any amount of time, which is why the player gets a draw. If your opponent cannot beat you in the position you achieved, you earned a draw within the time allotted, so even if time runs out, that's the worst possible result you could get. The result when you flag should be the worst possible result you could get from the position you achieved before you flagged. But the positions you mentioned are irrelevant anyway, since those would be immediately declared drawn, game over, end of discussion. This thread I am only talking about positions where checkmate is theoretically possible even if not forced or would require the opponent to help you win.

Edge cases as in they occur so rarely it isn't a huge impact.

In the discussions previously, when the rules currently in effect were being decided, it was mentioned that the US Chess implementation was considered better. The only case where that implementation is not in effect is on those rare cases when there's a forced mate on the board on timeout.

That second case is a win on timeout here. I actually thought it required a pawn for the side without time, like the US Chess rule, but apparently not. But apparently two knights versus a king plus any other material, is a win on timeout if the side with time.

All mate possibilities should be considered even if not forced. If they can do it for king and 2 knights they can do it for the other piece combinations I mentioned. I honestly don't know why they thought USCF rules were the best with this. And how exactly is "demonstrating a forced mate on the board" implemented? Does the player who still has time left have to explain it to the arbiter? What if he doesn't see it? It should go by what is possible in the position, not whether either player is aware of it or can tell the arbiter the moves. Yet another reason FIDE is better.

They're edge cases. Stop being stupid.

I could understand that argument (still wouldn't agree though) if it were help mates being discussed, like KNKR where the king and rook would have to box themselves into the corner and wait for the knight and king to come checkmate them, but in endgames like the ones I mentioned, they can occur frequently in chess games at any level. Say KPKN the player with the pawn is just rushing to promote the pawn hoping the opponent will make a mistake on low time:

White accidentally boxes himself in and black knows the mating net, and plays it out. This could easily happen in a blitz or bullet game, even between high rated players. The fact that white can deliberately let his time run out once he realizes this and obtain a draw is absurd. This isn't a help-mate or self-mate or position where the odds of a player blundering into it are astronomically low, it's a simple endgame that an occur and easy forced mate. I don't agree with using USCF to begin with, but how they are classifying "drawn sufficient piece positions" is even more insane. So KQKQ both players have to scramble making random moves not to flag, but in KNKP where the guy with the pawn is about to checkmate, he can flag and its a draw?

No, they are arbitrarily picking and choosing what piece combinations qualify as "insufficient losing chances". If KNKB and KNKP and KNNK and KNKN all get away with draws after flagging, then why not KQKQ or KRKR? I'd argue those are just as drawing, especially at my rating level. Yet I have to make a bunch of random premoves to not flagging, so why should the former not have to do the same? Heck, why not got even further and declare KNBKQ a draw if the KNB side reaches this position:

A well known fortress, just as easy to maintain as any minor piece random shuffling. It's either all or nothing. Either mate is possible or it isn't. I don't understand the double standard at all.

Sure, that'd be an improvement, have a "force draw" button that's different from offering a draw, and it only works for certain positions such as pawnless RvR and pawnless QvQ, and only after an engine verifies there's no instant win Searching for Bobby Fischer style.

But chess.com's code is a wet sack of sh*t so don't expect anything like that in the next century.

I suggested a compromise that would require even less code. It's only a draw if the opponent only has a lone king left. A king + even 1 pawn or piece will always have checkmating possibilities against the enemy king if he has 1 other piece, with only 3 exceptions:

KNKQ

KBKQ

KBKR

Anything else is either an immediate insufficient material draw for both sides, or checkmate is possible in some way, whether forced or an absurd 30 move help-mate. With the lone king parameter, only 3 exceptions need to be programmed. Anything else is either an immediate draw by insufficient material or a mate is possible. Unless of course this site forgot to program bishops of same color + king = draw, which would be pretty hilarious.

Pawnless R vs B has some winning positions, and it can be hard for the bishop to draw.

Anyway, I dislike your idea of letting endgames like pawnless R vs R be a win on timeout.

So opponents should be able to "force-draw" the game in this position:

Nice way to throw some basic endgame theory and underpromotion studies out the window! KRKR are heavy pieces, so in a blitz or bullet game, part of the skill is being careful not be skewered or checked away from the defense of the rook, within the time allotted. The KRKB example was to show how the bishop could never win, not the other way around. The rook should always be given the win if the bishop side runs out of time. I think some of the confusion here is that there is a difference between insufficient mating material and timeout draws. Insufficient mating material immediately ends the game even if both sides have 23 minutes left. Neither side could ever win. Timeout draws are where only one side can't win but the other side runs out of time. KNKQ, KBKQ, and KBKR are the only positions where the lesser material could never win.

Play FIDE tournaments then

problem solved

I'm talking about online chess. This site uses some weird hybrid version of them that makes even less sense than either alone.