King and Knight v King and Knight

[Roger de Coverly]

The forum is always happy to discuss 10.2 claims - but what about a form of chess where 10.2 does not apply, namely blitz chess?

Usually this is social rather than competitive so local conventions can apply. However blitz is used as a tie break in knock-out events - the structure is that one player is given more time than the other but the rule is that the player with less time will win the tie break in the event of a draw.

So what should happen if the position comes down to K+N v K+N as in the recent women's world championship and one player runs out of time. The relevant article in the rules of chess is

9.6

The game is drawn when a position is reached from which a checkmate cannot occur by any possible series of legal moves, even with the most unskilled play. This immediately ends the game, provided that the move producing this position was legal.

In the recent event the term even with most unskilled play was interpreted to mean that a win with K+N with K+N was possible because a helpmate could be constructed.

I suppose my point is that players need to know how these totally drawn (but not legally drawn) positions will be handled before they can assess whether the time odds are fair in these shoot-outs.

[Nigel Wright]

That's certainly an interesting and valid point, Roger, and also one which I have never come across in my 14 years of chess.

[Sean Hewitt]

K+N v K+N might seem an extreme example, but it's really no different to K+P winning on time versus (say) K+Q+R+R!! K+P can only win against helpmate there too. It's simply very unusual.

[John Upham]

In the same vein, can a King vs King ending be won by flagging ones opponent?

In one of our local leagues a hapless weak player was brow beaten into resigning with K vs K & 2 x Knights by a spectator from the opposing club.

What was really funny was that the stronger side captured the last pawn of the weaker side changing a won ending into a drawn one. As everyone no doubt knows, 2 knights versus a pawn can be won in 155 moves (according to Lamprecht & Muller).

How we (or at least I) laughed!

[David Shepherd]

I once lost a dead drawn blitz game (basically K+R v K+R) as there was no two minute claim available - I believe that the rules in this area need revising - I repeatedly offered my opponent a draw after every move just to show my opinion of the fact he was playing on. The controller of the tournament could do nothing other than to state they would never run another blitz tournament without time added per move


I think that to have this situation in a world championship match brings the game into disrepute. The game should clearly have been a draw based on common sense (although the rules may not say this), I still find it amazing that the tie beaks in such Championships are not played with time added per move - even if it is just for example 2 seconds, this would stop this from happening.

[Matthew Turner]

I am posting a couple of comments from the BCM blog, the first is mine

The rules are open to interpretation, but I have always believed that the intention of the rules was that the least skilled player should mean one who gives all their pieces away and not one who cleverly positions them to allow checkmate. Therefore, K + N vs K + N should be a draw. However, I think there is a complicating factor here. If the moves as reported are correct then Black declined a draw, by not playing NxN. In these circumstances White could claim that Black was playing for a win and by doing so risks losing as indeed they ended up doing.
I tend to feel that the balance of evidence suggests the game should be declared a draw.

These are Stewart Reuben's comments

Armageddon games can be very important, as here in the Women’s World Championship. In the US Rules, a bare bishop or knight cannot win on time unless there is a forced mate. This can be several moves deep. e.g. black pawn h4, black king h1, white knight e3, king f2. 1 Ng4 h3 2 Kf1 h2 3 Nf2 mate.

FIDE Laws do not provide for this possibility. I am hoping FIDE will agree to a new Law in Dresden.

• Where there is adequate supervision of a blitz game, the Arbiter may step to declare a draw if commonsense requires it.
• Examples include where neither player makes any claim:
• setting the clock incorrectly
• setting the pieces up incorrectly
• a player leaving his king in check and his opponent not noticing
• both players being in check
• a pawn being promoted, but no replacement piece being put on the board
• a pawn being promoted to a queen, but an upside-down rook being used
• bare bishop
• bare knight
• bare two knights against bare king
• multiple repetitions, four or more
• bishops transposing diagonals
• Pieces jumping over pieces. (In the last Hastings a slowplay game went 1 d4 Nf6, 2 c4 d5 3 Nc3 Bb4 4 e3. A spectator noticed, pointed it out to the arbiter and this was corrected.)
• a knight moving to its blind spot, for example Nf3 to Nd5

Of course where a player makes an illegal move, his opponent can claim a win. We are referring to those situation where he does not do so.
An arbiter who did not like the Law could always state that the supervision is not adequate. Indeed that would be true unless the arbiter is quite strong.

The great problem is wording the Law.

[Phil Neatherway]

I agree that the K&N v K&N decision brings the game into disrepute. It is completely ridiculous.

[David Shepherd]

I doubt white could claim that black declined a draw and was going for a win as my understanding of the rules of the play off were that a draw for black was counted as a win. I think clearly she just did not see the draw in the time scramble.

I am not sure if it would be relevant anyway - either there was enough material left to checkmate or there was not (depending on whether help mates should be allowed). Maybe the arbiter should have stopped the game when K+N v K+N was reached.

[Stewart Reuben]

John Upham wrote: In the same vein, can a King vs King ending be won by flagging ones opponent?

In one of our local leagues a hapless weak player was browbeaten into resigning with K vs K & 2 x Knights by a spectator from the opposing club.

King v king you cannot win, or lose. The game is drawn.

The team captain of the hapless player missed a trick. The opposing club had clearly cheated. Thus the hapless player should get his draw (he could not win as he could not deliver mate). The captain could have claimed that the opponent should lose. That would have been my position as a captain and then agreed it should be drawn as a compromise, which would have been my decision as an arbiter.

Matthew said The rules are open to interpretation, but I have always believed that the intention of the rules was that the least skilled player should mean one who gives all their pieces away and not one who cleverly positions them to allow checkmate. Therefore, K + N vs K + N should be a draw.

I cannot speak for my associates in the committee. But I definitely meant it to mean provided a mate could be constructed with the remaining pieces. It is being amended in Dresden to simplify the rule. The Appeal Committee was right about king and knight v king and knight in the Women's World Championahip. Just as I was correct to give a win for a bare bishop against rook and two pawns in Rumania this year. But that is why I want to change the Laws, now having had some experience.

Try the following, because you may not be aware:

Black king a1. Pawn a2. Queen f4. White king d1. Pawn e2. Queen a3. White plays 1 Qc1ch One or other flag falls. The game is a drawn as white is forced to stalemate black.
You can amuse yourself by posting black queens on e3, f4, g5, h6 and white ones on c8, c7, c6. It is still a draw.

Stewart

[Matthew Turner]

By this logic, if White had a pawn on d2 and a king on d3 against a Black Bishop on f8 and King on f7, then White could lose on time, because it would be possible for them to promote the pawn to a knight and then allow checkmate. If this is the intention of the laws then I think this is completely wrong.

[Stewart Reuben]

Matthew, it is not completely wrong. It is simplistic. The whole concept has come a long way since 1984 before which it was possible to win on time with a bare king. The reason for the debate is the rise of knockout events and the need then for Armageddon finishes.
My 'commonsense' suggestion fits your requirements, I think. But it is very woolly. We have to remember the Laws are applied by arbiters all over the world, with varying understanding of English, or indeed none at all.
Think of a better way of expressing the concept. It will be discussed in Dresden.
We have had a great number of problems with 10.2 cannot win by normal means in slowplay games - which I introduced in Islington in 1973 with great trepidation.
Somebody has said why not have a delay mode of perhaps 2 seconds to overcome this problem in Armageddon games? Fair enough, but as a wrinkly I think 3 seconds are necessary. But what time should white get against black's five then to overcome his advantage of draw odds. I suspect ten minutes. I am sure 6 for white against 5 for black with no adding time favours black from my experience.
Stewart

[Sean Hewitt]

Somebody has said why not have a delay mode of perhaps 2 seconds to overcome this problem in Armageddon games? Fair enough, but as a wrinkly I think 3 seconds are necessary. But what time should white get against black's five then to overcome his advantage of draw odds. I suspect ten minutes. I am sure 6 for white against 5 for black with no adding time favours black from my experience.
Stewart

I read somewhere about an idea where player A determines the time for white and black, and then player B gets to choose which colour they would rather be under those time controls. What do you think of that?

[David Shepherd]

A few test tournaments/matches could be run with just Armageddon games to test the different time limits with time added to find the best combinations of time in order to give equal average results to black and white.

[IM Jack Rudd]

Somebody has said why not have a delay mode of perhaps 2 seconds to overcome this problem in Armageddon games? Fair enough, but as a wrinkly I think 3 seconds are necessary. But what time should white get against black's five then to overcome his advantage of draw odds. I suspect ten minutes. I am sure 6 for white against 5 for black with no adding time favours black from my experience.
Stewart

I read somewhere about an idea where player A determines the time for white and black, and then player B gets to choose which colour they would rather be under those time controls. What do you think of that?

I like it. An extension of the "you cut, I'll choose" method for dividing a cake.

[Greg Breed]

Somebody has said why not have a delay mode of perhaps 2 seconds to overcome this problem in Armageddon games? Fair enough, but as a wrinkly I think 3 seconds are necessary. But what time should white get against black's five then to overcome his advantage of draw odds. I suspect ten minutes. I am sure 6 for white against 5 for black with no adding time favours black from my experience.
Stewart

I read somewhere about an idea where player A determines the time for white and black, and then player B gets to choose which colour they would rather be under those time controls. What do you think of that?

I like it. An extension of the "you cut, I'll choose" method for dividing a cake.

Hah! You beat me to it Jack!
Sounds great to me.