at the mercy of the arbiter

[James Toon]

Two knights versus a pawn is much more challenging. Anyone know of this ending in GM practise?

I thought I'd check on a database and it's surprisingly common – over 200 games, of which more than half (57%) ended up drawn. In contrast, K+B+N v K occurs much more often and only a quarter of the games are drawn.

[James Toon]

Before anyone else points it out, the figures I gave are for all games. If you restrict the search to GM games only (both players above 2500), then there are only 12 examples of K+2N v K+P and 75% of these were drawn. Topalov is the only GM to have played this ending twice.

[Keith Arkell]

Jonathan Rogers Post subject: Re: at the mercy of the arbiterPosted: Wed Jun 10, 2009 1:19 pm



''Today, strong players resist! and if they see that their opponents are about to develop a positional bind of the sort which Keith relishes, they won't simply allow it - they will complicate the game instead!

That, of course, is not to say that we are seeing such a high level of chess that there is no joy in following it. But it is true that the audience has to work harder to understand the top games, and I wonder whether that is what Keith does not like so much. Annotators too have to work much harder than they used to in order to make the top games comprehensible. Not all are able to do this.''



I think that Karpov,Smysov,Andersson,Petrosian etc are/were able to control their games in a positional,or what I like to call an artistic manner,because they are/were so strong tactically.Simon Williams and myself have had the pleasure of analysing with Ulf Andersson and he is one of the strongest tactical players I have ever met. There can be a misconception that players who stylistically choose to play in a tactical or attacking manner are stronger tactically than players who use their analytical ability to keep careful control over their games.
Of course because defense requires greater tactical accuracy than attack,all of the above players have on rare occasions been blown away by energetic onslaughts.

You are right there Jonathan.When I am playing through games for pleasure I don't want to have to work hard.This is kind of my point.Taken to extreme perfect chess maybe almost inomprehensible to us when observed casually,because it will do away with the artificial concepts which we understand visually,enjoy artistically,and use as crutches because we can't analyse millions of variations per second.
To give you an idea of what I mean let us take the example of some endgame wins,such as 2 Bs v 1N (and there are many tougher examples than this!) which have been analysed out by computers.When we play through the main lines we don't have a feel for what is going on because the progress of the winner can only be understood by horrifyingly deep analysis.There are no visual clues.

I am not sure that annotators have to work harder though with high level computer type chess.I think it may be easier for them to simply ask Rybka for help in tactical situations,than to try to explain what Ulf or Karpov are playing at, because the latter requires of the annotator a deep understanding of chess.

[Roger de Coverly]

I thought I'd check on a database and it's surprisingly common – over 200 games, of which more than half (57%) ended up drawn.

There's a 1998 game Dearing - de Coverly from the NCL which was a fairly easy win. In fact it's possible Eddie wanted to show off his technique by deliberately provoking the ending.

By contrast there's a game Rudd - Lalic (B) where Jack finds some useful defensive ideas and holds the draw.

[Stewart Reuben]

I did have 2 knights against king and pawn against David Barasi. But the pawn was in the centre on the fifth rank, blockaded by my king. Thus it could only be blockaded on the sixth by a knight. I knew that was a theoretical draw and didn't even try.

Rook and bishop against rook is an arbiter's nightmare and very common. It should be scored 0.6 for the superior side and 0.4 for the inferior and we would all be able to go home.
In fact that should apply to all games. I would be happy in terms of score with 0.1 against Adams. Others could simply agree 0.5 to 0.5. You get the superior position, make your move and offer your opponent 0.3. It would save all that work on the game scores and reduce the need for arbiters. It would make Swiss Pairings very easy.
It would also get people experience in investing in the stock market or spread betting.
Stewart Reuben

[IM Jack Rudd]

I had rook and bishop against rook against David Eggleston recently, and managed to win it. Despite its being a theoretical draw, I think 0.6 is a trifle undergenerous to the side with the rook and bishop. Given how frequently that side manages to win, I'd say 0.8 would be more in line with the actual stats.

[Keith Arkell]

Well perhaps I'd accept 0.95,but not less. I think I have 17/17 in this ending

[Stewart Reuben]

Typical. I sensibly suggest fractions in tenths and somebody wants to complicate it and make it hundredths.
Stewart Reuben

[John Upham]

Well perhaps I'd accept 0.95,but not less. I think I have 17/17 in this ending

Keith,

Howabout R & N vs R? What are your stats with that one?

[John Hickman]

Well perhaps I'd accept 0.95,but not less. I think I have 17/17 in this ending

Keith,

Howabout R & N vs R? What are your stats with that one?

Probably around 80%, but he scores a lot better if you give him the side with the Knight

[Keith Arkell]

Not as good as Kasparov's - see Kas Polgar J.

[John Saunders]

Re K+B+N v K: I once trawled through the Mega Database to check the stats on this. One curious fact emerged - I discovered that every example prior to 1965 (or thereabouts) had been decisive but that there had been a surprisingly large number of draws after that date. It occurred to me that this would have been for two reasons: (a) only top-level games were recorded prior to recent times, whereas these days tournaments like the Albanian Under-8 Championship find their way onto the database; (b) rapidplay, blitz and quickplay finish games now appear in large numbers. Looking closer, I found just one game where a top-level player failed to win this endgame in what looked like a longplay tournament game. That game was Kempinski-Epishin, Bundesliga 2000, where they reached the position wKa5/bKc6,Nb6,Bb8 with Black (Epishin) to play. This very strong GM who was once Karpov's trainer played the unbelievable 150...Kc5 when every Russian schoolboy (and a good few British ones) would have played 150...Nd5 on auto-pilot (the W manoeuvre, as an earlier poster referred to it). I showed this game to a British GM and he was incredulous that a fellow GM simply did not know this basic bit of endgame technique. Even though the high move number indicates that he was short of time, a GM should have been expected to know his stuff and flash the move out instantly.

This is the only example I have been able to find of a high-rated player in a longplay tournament not being able to do the business. As for lower-rated players: I recall a most amusing game in the Isle of Man a few years ago where FM Andrew Smith got up from his board with K against his opponent's K+B+N. He strolled over to me and, with a grin on his face, said: "I prefer my position to his!". By that he meant that his own endgame technique wasn't up to the job of finishing the game off had their positions been reversed! He turned out to be doubly lucky because his opponent's technique was also not up to the job. After the draw was agreed, his opponent ruefully confessed to a gloating audience "I don't know my classics!". I recall thinking how embarrassing this must have been for the poor fellow and scurried off to brush up my own B+N technique just in case it happened to me. Of course it is quite easy - when you know how. Ditto, K+Q v K+R. Capablanca's 'Chess Fundamentals' covers both in easily comprehensible fashion and I saw a very nice article in CHESS a few years ago which explains B+N v K in simple terms that even the humblest club woodpusher could grasp and memorise.

K+2Ns v P is much harder, of course. Once in the mid-1990s I reached a dreadful position in a game in a weekender against a girl of about 14 and was on the brink of a most embarrassing defeat. I then spotted a way to reach an endgame where I had the pawn against the two knights. Rather smugly (and I confess, sexistly, if there is such an adverb), I thought to myself that this young girl was unlikely to have the necessary technique to win against my c.190 graded self in this difficult endgame, particularly since she only had a few minutes left of the 15-minute quickplay finish to do it. After checking the scoresheet to make sure her surname was not "Polgar", I decided to take a chance on it. But it turned out that I had seriously underestimated my opponent's skill. The young girl manoeuvred me to the brink of defeat before unluckily running out of time to finish it off. Another ten or twenty seconds and I would have been a dead man. At the end she seemed quite cross with herself for failing to win (although she was very polite to me). After this chastening experience I remember thinking "this girl is going to be a very good player". I was right. She's Jovanka Houska.

[JustinHorton]

I'm pretty sure there was a drawn "bishop and knight v king" ending in4NCL a couple of seasons ago, the lucky player being my friend Sean Terry.

[Roger de Coverly]

I'm pretty sure there was a drawn "bishop and knight v king" ending in4NCL a couple of seasons ago, the lucky player being my friend Sean Terry.

It was as long ago as 2003 at West Bromwich.

There was another drawn one in the Major Open at Douglas 2005 (and that was with a minute increment)

I've got 9 examples after that from British events all of which were wins. One really confident player chose to promote to a bishop instead of a queen.

[Stewart Reuben]

>As for lower-rated players: I recall a most amusing game in the Isle of Man a few years ago where FM Andrew Smith got up from his board with K against his opponent's K+B+N. He strolled over to me and, with a grin on his face, said: "I prefer my position to his!". By that he meant that his own endgame technique wasn't up to the job of finishing the game off had their positions been reversed! He turned out to be doubly lucky because his opponent's technique was also not up to the job. After the draw was agreed, his opponent ruefully confessed to a gloating audience "I don't know my classics!".< John Saunders

Of course the opponent did not have to know anything had the arbiter been on the case. He could have awarded the game to the opponent on the grounds that Andrew was discussing the game during play.

Craig Pitchett manages it with less than one minute left on his clock against, I think Jimmy Adams, at one of the Evening Standard Opens. It took him 25 moves. I was observing the game as arbiter and counting them off silently.

Stewart Reuben