Insufficient mating material

Hello,

in the last game I played, my opponent lost on time. And it was considered a draw.

We were both low on time and I messed up a winning position. The last move of my opponent was taking a pawn close to promotion, with his Knight. Instead of taking the knight, which left the position with two knights for me and one knight for my opponent I did not recapture the knight and the time of my opponent ran out.

I did not recapture, because you cant mate with two knights and the king only, but with the knight of the opponent it is definitly possible. So I dont know why it is still considered a draw.

You can see the final position here:

https://www.chess.com/livechess/game?id=1513151157

Is there any specific rule, when it is a draw, after you lose on time, or is this just a bug?

Thanks in advance

The site uses a variation of the US Chess Federation insufficient material rules and not FIDE. So, it doesn't matter if there is a theoretical but non-forcible win or not.

The site just looks at the material that the player with time has. Since two knights can not force mate, it is declared a draw.

The last line of the following gives the material combinations that will result in insufficient material draws here.

https://support.chess.com/customer/en/portal/articles/1444798-how-do-i-claim-a-draw-

We've had previous threads on this or something similar.  Seems the USCF and FIDE have slightly different interpretations of the insufficient material rule and chess.com follows the USCF rule.  Maybe an arbiter or someone equally knowledgeable can quote you the chapter and verse; I can't... .  I do remember a thread or two going on and on though while arguments were presented for all sides.

Day Late and a Dollar Short Again...

Yeah, the USCF one isn't followed strictly either and could be (easy code check). Mainly the decision is to ease programming adjudication. FIDE rules would require more complex coding to determine if mate is possible or just assume in most positions mate is possible if enough pieces exist on the opponent's side so that help-mates would be feasible.

Thanks for your reputations.

Two Knights cannot force Checkmate on a bare King, but two Knights against one sometimes can. Probably not in this position, though, but I wonder if this website would also adjudicate positions that are forced mate in 3 or 4 as draws...

FICS is known for even adjudicating mate in 1 as draw, in KNKN.

1 fxg8=Q Bxg8! 1/2-1/2

for post 6 if...bxg8 then ng6#.Or I'm I missing something there?

Well, FICS is missing something. By assuming that KNKB is insufficient material under USCF rules, and thus declaring a draw after Bxg8, because mate cannot be forced. I just wondered if chess.com would similarly miss forced wins in KNNKN (which could be longer mates).

Yeah, some forced mates might be lost. My guess is the problem might impact a miniscule number of games (the forced situations that is).

It actually would impact a lot of games, as far as I know, especially in bullet, where it is common to reduce the opponent to just a minor piece or other non-mating material.  In particular, it would be a shame if I just had a knight against a rook pawn, but had my opponent run out of time, even though in many cases there is Stamma's mate; my opponent should not be able to extricate himself simply by waiting for the clock to run down. The mate need not be forced under FIDE rules - one minor piece against another is not considered insufficient material under any circumstances, even if there is not a forced mate. I think Chess.com, given its international reach, should at least consider implementing FIDE rules - a mate search isn't necessary.

It is easier to check material programmatically than it is to determine if a specific material combination can be arranged in a mating position (mate possible by any legal series of moves).

That is the crux of the situation. It is still my assertion that the number of games where the FIDE implementation would result in a win rather than a draw, as it does now, is small. I would guess less than a tenth of a percent of games, but that is just pure conjecture. 

A player with a sufficient number of games could probably come up with a better estimate by looking at their own database and finding how many insufficient material draws here would have been wins under FIDE rules. It would be anecdotal but better than guessing 

Just checked your bullet stats. 3% of your 31,000+ games were drawn. Of those draws, how many were for insufficient material and of those how many were wins under FIDE rules (for either side)?

1. https://www.chess.com/live/game/1611128000
2. https://www.chess.com/live/game/1379554480 (game could've been drawn much earlier, and without risk of losing on time, if FIDE rules were implemented - in the final position there is absolutely no way for either side to make progress)
3. https://www.chess.com/live/game/1298650854
4. https://www.chess.com/live/game/1234457699 (helpmate exists for Black - white sacrifices the queen for free and promotes pawn to knight which then can be made to obstruct the king)
5. https://www.chess.com/live/game/1249915381
6. https://www.chess.com/live/game/1241914318
7. https://www.chess.com/live/game/1218747984
8. https://www.chess.com/live/game/1208706107
9. https://www.chess.com/live/game/1136730159
10. https://www.chess.com/live/game/1132899401
11. https://www.chess.com/live/game/1102212726
12. https://www.chess.com/live/game/1084897520
13. https://www.chess.com/live/game/1046400894
14. https://www.chess.com/live/game/1045204028
15. https://www.chess.com/live/game/1026891376
16. https://www.chess.com/live/game/1025764923 (again, helpmate exists for White - Black promotes pawn to knight, gives away rook for free, hems in own king with knight)
17. https://www.chess.com/live/game/1025753089 (same type of helpmate)
18. https://www.chess.com/live/game/1001754020
19. https://www.chess.com/live/game/933203049
20. https://www.chess.com/live/game/889040091
21. https://www.chess.com/live/game/889012419 (again, helpmate possible)
22. https://www.chess.com/live/game/830082388
23. https://www.chess.com/live/game/816752169
24. https://www.chess.com/live/game/798860627
25. https://www.chess.com/live/game/776860134
26. https://www.chess.com/live/game/774382606
27. https://www.chess.com/live/game/770634245
28. https://www.chess.com/live/game/758338648 (if I had not played ...Bf4 but a king move instead as my last move, I could have possibly claimed a win under FIDE rules if he timed out)
29. https://www.chess.com/live/game/745267625 (helpmate)
30. https://www.chess.com/live/game/723818308
31. https://www.chess.com/live/game/663905741
32. https://www.chess.com/live/game/663899238
33. https://www.chess.com/live/game/663888954 (yes, this and the last two were with the same opponent on the same day)
34. https://www.chess.com/live/game/587102417 (the most obvious example)

and I probably missed a few examples. These amount to 34/909 = 3.7% of my drawn games, but make up a substantially bigger percentage of my games drawn by insufficient material (at least 1/3).

In a number of these games you link to, it was you who ran out of time. What are you complaining about ?

In many of the others you have a very poor position compared to your opponent. Do you really expect your opponent to queen a pawn and then use it to blockade his own king ?

It has already been made clear. This site uses a "hybrid" USCF rule which relies on a simple piece count to determine the result on a time-out.

If you dont like it go and play on FIDE's own site.

In case you didn't notice, I think it's equally practical for Chess.com to enforce FIDE's rules, which take precedence over USCF's on a global scale. I just wanted to make the point that these are not as uncommon as one might think.

And in case you didn't notice, I was pointing out that:

1. your 3.7% figure is not true, and

2. The rules on this site are well known, often discussed, and no-one is forced to play by them.

So, around 0.11% of your games could have been won, by the player with time, using FIDE rules. About 1/10th of a percent. 

I know that is anecdotal but I'm guessing that number is probably close over all games.

And FIDE's rules take precedence in FIDE rated competition. Not here and not in USCF events unless they are also rated under FIDE.

The site decided to use the much less complex method of forced mating material, a variation even of USCF rules. And for the number of likely games that are different it probably isn't worth the programming time to change the implementation significantly.