FIDE rule 10.2 and juniors

[Alex Holowczak]

Tell us more!

Here goes...

This one arose out of a mistake that I made in a league game.

White had KN, black had KPP. Black didn't claim a 10.2. (It was stonewall 10.2, I wouldn't even have bothered referring it to the rules committee.) Black's flag fell. White pointed it out. Everyone looked to me to tell them what happened now. In the heat of the moment, I couldn't think of a way that white could checkmate the opponent. It turns out there is a way, via underpromotion of a pawn or two. I got it wrong and appealed to the Rules Committee about it being a draw, when it should have been a win for white. It probably wouldn't have succeeded. I withdrew my appeal when I spoke about it to Dave Welch at a 4NCL weekend, who said that the English guidance was that you can't consider underpromotion in this case. Peter Purland agreed with this. So I withdrew my appeal. I was quite happy for that to be the rule if that's what I was being told. Again, I considered this to be the norm given two Senior Arbiters were telling me, so didn't question it.

Speaking to arbiters after this just as a conversation, they seemed both oblivious and horrified in equal measure. So they suggested it'd be raised at the AGM...

Dave and Peter's logic that was the words "unskilled play" meant that underpromotion couldn't be considered, because it was less skilful to promote to a Q than to underpromote. Other arbiters pointed out that the rule doesn't say that anymore, it says "mate with any series of legal moves" (or something like that). I confess to not checking that at the time.

If Alex M could give his opinion on this, it'd be appreciated.

[Roger de Coverly]

Of course, this is distinct from the situation where one player had 20 minutes and the other 2, where the game had reached about move 20 in a G/45 game with the guy having an overwhelming position. That should have been rejected because I cold see on the scoresheet - which admittedly he didn't need to have a copy of - showed that his last few moves had been making progress. From memory, he'd won a pawn back from a gambit and had directed all his pieces at the opponent's King...

Are you saying that a player with two minutes remaining at move 20 was allowed to make a 10.2 claim? In other words the arbiter came and watched and would have awarded a draw if any of the play showed signs of uncertainly? It has to be totally wrong to interfere with a game in this manner. If the claimant hadn't been short of time, the player could play as he saw fit and if some moves appeared mysterious, then either they were just random or too deep for the arbiter.

You could perhaps accept a 10.2 claim at move 20 if the claimant had a serious material advantage. Even there, you perhaps give the decliner of the draw the chance to keep the position complex. It is after all, part of the point of a G/45 tournament that you lose if you don't play quickly enough. 10.2 mitigates this, but shouldn't abolish the notion of losing on time.

[Ian Kingston]

Ian Kingston said:

It is possible that the CAA interpretation is wrong. It is also possible that the CAA is right and that the law is an ass.

I think it's a bit of both hence my proposal to strengthen the law itself and devolve power away from the arbiter. The proposal can be found in Chess Questions.

I've read it. If I manage to come up with any coherent thoughts on it I'll post over there. I support the intent, at least!

[Peter Shaw]

Tell us more!

Here goes...

This one arose out of a mistake that I made in a league game.

White had KN, black had KPP. Black didn't claim a 10.2. (It was stonewall 10.2, I wouldn't even have bothered referring it to the rules committee.) Black's flag fell. White pointed it out. Everyone looked to me to tell them what happened now. In the heat of the moment, I couldn't think of a way that white could checkmate the opponent. It turns out there is a way, via underpromotion of a pawn or two. I got it wrong and appealed to the Rules Committee about it being a draw, when it should have been a win for white. It probably wouldn't have succeeded. I withdrew my appeal when I spoke about it to Dave Welch at a 4NCL weekend, who said that the English guidance was that you can't consider underpromotion in this case. Peter Purland agreed with this. So I withdrew my appeal. I was quite happy for that to be the rule if that's what I was being told. Again, I considered this to be the norm given two Senior Arbiters were telling me, so didn't question it.

Speaking to arbiters after this just as a conversation, they seemed both oblivious and horrified in equal measure. So they suggested it'd be raised at the AGM...

Dave and Peter's logic that was the words "unskilled play" meant that underpromotion couldn't be considered, because it was less skilful to promote to a Q than to underpromote. Other arbiters pointed out that the rule doesn't say that anymore, it says "mate with any series of legal moves" (or something like that). I confess to not checking that at the time.

If Alex M could give his opinion on this, it'd be appreciated.

There was a case very similar to this in one of the minor sections of the British a couple of years ago. That was awarded as a win, I'm sure that is the correct ruling.

[Roger de Coverly]

. Other arbiters pointed out that the rule doesn't say that anymore, it says "mate with any series of legal moves" (or something like that). I confess to not checking that at the time.

I thought Stewart Reuben and friends changed this at Dresden in 2008. You may recall a high profile women's event where the ending of NvN arose. The arbiters were unclear at the time whether a loss on time should stand. The resolution at Dresden was that if the material on the board allows a helpmate to be constructed, then a loss on time stands. I've even warned members of my club about this, that they need to remove all their opponent's pieces if trying for a win when short of time. If any pieces remain they can lose if their flag falls and they haven't offered or claimed a draw.

[Nick Thomas]

Tell us more!

Here goes...

This one arose out of a mistake that I made in a league game.

White had KN, black had KPP. Black didn't claim a 10.2. (It was stonewall 10.2, I wouldn't even have bothered referring it to the rules committee.) Black's flag fell. White pointed it out. Everyone looked to me to tell them what happened now. In the heat of the moment, I couldn't think of a way that white could checkmate the opponent. It turns out there is a way, via underpromotion of a pawn or two. I got it wrong and appealed to the Rules Committee about it being a draw, when it should have been a win for white. It probably wouldn't have succeeded. I withdrew my appeal when I spoke about it to Dave Welch at a 4NCL weekend, who said that the English guidance was that you can't consider underpromotion in this case. Peter Purland agreed with this. So I withdrew my appeal. I was quite happy for that to be the rule if that's what I was being told. Again, I considered this to be the norm given two Senior Arbiters were telling me, so didn't question it.

Speaking to arbiters after this just as a conversation, they seemed both oblivious and horrified in equal measure. So they suggested it'd be raised at the AGM...

Dave and Peter's logic that was the words "unskilled play" meant that underpromotion couldn't be considered, because it was less skilful to promote to a Q than to underpromote. Other arbiters pointed out that the rule doesn't say that anymore, it says "mate with any series of legal moves" (or something like that). I confess to not checking that at the time.

If Alex M could give his opinion on this, it'd be appreciated.

This should be a win for white. BTW are you certain that a promotion was necessary?

[Alex Holowczak]

. Other arbiters pointed out that the rule doesn't say that anymore, it says "mate with any series of legal moves" (or something like that). I confess to not checking that at the time.

I thought Stewart Reuben and friends changed this at Dresden in 2008. You may recall a high profile women's event where the ending of NvN arose. The arbiters were unclear at the time whether a loss on time should stand. The resolution at Dresden was that if the material on the board allows a helpmate to be constructed, then a loss on time stands. I've even warned members of my club about this, that they need to remove all their opponent's pieces if trying for a win when short of time. If any pieces remain they can lose if their flag falls and they haven't offered or claimed a draw.

Yes, I was aware of this. I'd forgotten that the P could become a N...

I think they changed several things because of that:
(1) Tidying up the aforementioned rule
(2) Bringing in rules about Blitz in adequate supervision; i.e. that the normal rules apply
(3) Bringing in incremental time controls for the Armageddon games FIDE run

[Alex Holowczak]

This should be a win for white. BTW are you certain that a promotion was necessary?

No. This is what I was trying to think about, and probably why I overlooked underpromotion. The pawns were on the f and g files. I thought of a situation with the black pawn on h2 and King on h1, with the white King on f1 and the Knight on f2. Transposing that one or two files to the left, black had escape squares to the g or h-file, and I couldn't see how black could be mated.

[David Shepherd]

In my opinion the rules need amending to say that with lone Knight or Bishop the game is drawn unless the player with the Knight or Bishop can force a win, allowing for some bizare sequence of help mates brings the game in to disrepute, as the chances of it happening are so remote. Rules should be logical - currently this one is not.

[Roger de Coverly]

This should be a win for white. BTW are you certain that a promotion was necessary?

There's a help-mate position in a corner with KN v KN. If a rook's pawn was available, under promotion might not be needed.

See also http://www.ecforum.org.uk/viewtopic.php?f=2&t=434

Although that thread doesn't say so, it was my impression that Stewart's change to clarify KN v KN as a win on time was accepted.

[Roger de Coverly]

. Rules should be logical - currently this one is not.

It's completely logical, if your flag falls you lose on time unless ... . It just clarifies that if your opponent has KN and you do as well, that isn't an unless. With the world championship and its qualifiers plausibly settled by blitz, these issues become important. In mitigation the "arbiter supervision" rule has also been introduced for blitz, so I believe a 10.2 is now possible for blitz with zero second increments with an arbiter for every game. Kazan had an increment from move 60 even for the blitz playoffs.

[Sean Hewitt]

The law is clear that if checkmate is possible by any series of legal moves it's a win if the flag falls.

The problem here was that two ECF Senior Arbiters (one of whom is the ECF's Chief Arbiter) were not aware of what the law says.

[E Michael White]

This is what I was trying to think about, and probably why I overlooked underpromotion. The pawns were on the f and g files.

With f and g pawns (ed. in fact any two paws connected or not) black can mate with a knight when black promotes to a Q. The rule is however as Sean says below.

I mentioned a while back that some English arbiters wanted to use this disallowing underpromotion lark. Its a disgrace if what you say is true, they should stick to the rules instead of trying to put there own flavour out there.

I guess the CAA, in its current form, has outlived its usefulness and should be replaced in a way that requires them to perform regular MOT checks on all arbiters.

The 10.2 rule should probably add "even with the most unskilled arbiting" judging by current decisions.

[Alex Holowczak]

This is what I was trying to think about, and probably why I overlooked underpromotion. The pawns were on the f and g files.

With f and g pawns black can mate with a knight when black promotes to a Q.

This became equally apparent after the event.

[Roger de Coverly]

That K v KR draw has been around for years. It's mentioned in http://www.lrca.org.uk/lrca/middlegame/26.pdf which is dated January 2007.

10.2a itself states ( my bold)

If the arbiter agrees the opponent is making no effort to win the game by normal means, or that it is not possible to win by normal means, then he shall declare the game drawn. Otherwise he shall postpone his decision or reject the claim.

Why and under what circumstances would an arbiter not reject a claim by a player with a sole king against king and rook? So I'm asking why you would even get into the scenario of postponing his decision. I could accept (as probably would everyone who has commented) that if the arbiter had been aware that forty or more moves had already been played in the rook ending, that awarding a draw was fair given the repetition. But surely not if the position is being shown to the arbiter for the first time.