Eliminating/discouraging draws

[Paul McKeown]

I think it's quite an interesting debate actually... Playing theory lines vs offbeat stuff.

That isn't how I see it. I don't think that 5... f5 is off-beat, it's just bad, a cheap trick to get a pawn to e4 and hope White will not cope. Plenty of interesting off-beat lines, many of them fairly decent.

I would mention a recent bestselling book involving a certain 3 moves as black against anything

I'm not sure that the Acc. Drag. and associated lines are that off-beat. Been around for a long time, some pretty heavyweight players have essayed them.

Here is a nice game attacking with queen and knights. More of a closed position than some of the earlier efforts. And some other pieces which get in the way... A good way to avoid a draw playing the hippo. My record in longplay : W2 D1 L4, but two great wins!

Faulkes, Nick - Hadi, Justin 2009.05.04
Round 11.4 4NCL/Div3/OXF2-WR2 Daventry

1.e4 b6 2.d4 Bb7 3.Bd3 e6 4.Nf3 d6 5.c4 Ne7 6.Nc3 g6 7.d5 e5 8.Bg5 Bg7 9.Qd2 h6 10.Bh4 Nd7 11.Bc2 a6 12.Rc1 Bc8 13.a3 Nf8 14.b4 Nh7 15.c5 g5 16.Bg3 Ng6 17.O-O h5 18.h3 g4 19.hxg4 hxg4 20.Nh2 Nf6 21.Qe2 Bh6 22.Rce1 Bf4 23.Qd3 Nh5 24.Bd1 Qg5 25.cxd6 cxd6 26.Bxf4 Ngxf4 27.Qe3 Qh4 28.g3 Qxh2+ 29.Kxh2 Nxg3+ 30.Kg1 Rh1 mate

Nice combination, but surely not a great advert for Owen's Defence? Black appears to have a King's Indian Defence with a misplaced light squared bishop after seven moves, how are you going to sacrifice it on the kingside? Think White drifts a little with stuff like 11. Bc2, but he was fine anyway until 17. 0-0, which was simply suicidal. I would have thought moving the queen, maybe 17. Qe3, allowing the normal Nf3-d2-c4 would have been in order, perhaps together with f2-f3 and re-routing the dark squared bishop.

[Justin Hadi]

I think it's quite an interesting debate actually... Playing theory lines vs offbeat stuff.

That isn't how I see it. I don't think that 5... f5 is off-beat, it's just bad, a cheap trick to get a pawn to e4 and hope White will not cope. Plenty of interesting off-beat lines, many of them fairly decent.

Well there is offbeat at 180 level and there is offbeat at 100 level. And for example the Pirc is pretty off-beat at super-gm level. I get the feeling we're never going to agree on this point so I'll leave it there.

I would mention a recent bestselling book involving a certain 3 moves as black against anything

I'm not sure that the Acc. Drag. and associated lines are that off-beat. Been around for a long time, some pretty heavyweight players have essayed them.

Have you read the book?

Here is a nice game attacking with queen and knights. More of a closed position than some of the earlier efforts. And some other pieces which get in the way... A good way to avoid a draw playing the hippo. My record in longplay : W2 D1 L4, but two great wins!

Faulkes, Nick - Hadi, Justin 2009.05.04
Round 11.4 4NCL/Div3/OXF2-WR2 Daventry

1.e4 b6 2.d4 Bb7 3.Bd3 e6 4.Nf3 d6 5.c4 Ne7 6.Nc3 g6 7.d5 e5 8.Bg5 Bg7 9.Qd2 h6 10.Bh4 Nd7 11.Bc2 a6 12.Rc1 Bc8 13.a3 Nf8 14.b4 Nh7 15.c5 g5 16.Bg3 Ng6 17.O-O h5 18.h3 g4 19.hxg4 hxg4 20.Nh2 Nf6 21.Qe2 Bh6 22.Rce1 Bf4 23.Qd3 Nh5 24.Bd1 Qg5 25.cxd6 cxd6 26.Bxf4 Ngxf4 27.Qe3 Qh4 28.g3 Qxh2+ 29.Kxh2 Nxg3+ 30.Kg1 Rh1 mate

Nice combination, but surely not a great advert for Owen's Defence? Black appears to have a King's Indian Defence with a misplaced light squared bishop after seven moves, how are you going to sacrifice it on the kingside? Think White drifts a little with stuff like 11. Bc2, but he was fine anyway until 17. 0-0, which was simply suicidal. I would have thought moving the queen, maybe 17. Qe3, allowing the normal Nf3-d2-c4 would have been in order, perhaps together with f2-f3 and re-routing the dark squared bishop.

Well I enjoyed the game

[Roger de Coverly]

Have you read the book?

The author advocates g6, Bg7 and c5 against almost anything. It can still be up to the player with the white pieces to determine what to transpose into. There are some unusual positions if white takes on c5. Provided there's a Knight on c3, it's suggested that Bxc3 be played leaving white with tripled pawns but the absence of the dark squared Bishop in compensation.

Coverage is a bit limited on the hack stuff tried by Julian Hodgson and Simon Williams, namely 1 e4/d4 g6 2 h4 or 1 e4 g6 2 d4 Bg7 3 h4. There's also a line which contrives to transpose into a main line Accelerated Dragon but allowing White to save a tempo in the process. It doesn't look like a position where the extra move has been wasted either.

[Justin Hadi]

Yep and there's also the Dzindzi-Indian (1.d4 g6 2.c4 Bg7 3.Nc3 c5 4.d5 Bxc3+). I haven't actually read the book myself but the reviews indicate where offbeat lines are recommended.

[Justin Hadi]

Faulkes, Nick - Hadi, Justin 2009.05.04
Round 11.4 4NCL/Div3/OXF2-WR2 Daventry

1.e4 b6 2.d4 Bb7 3.Bd3 e6 4.Nf3 d6 5.c4 Ne7 6.Nc3 g6 7.d5 e5 8.Bg5 Bg7 9.Qd2 h6 10.Bh4 Nd7 11.Bc2 a6 12.Rc1 Bc8 13.a3 Nf8 14.b4 Nh7 15.c5 g5 16.Bg3 Ng6 17.O-O h5 18.h3 g4 19.hxg4 hxg4 20.Nh2 Nf6 21.Qe2 Bh6 22.Rce1 Bf4 23.Qd3 Nh5 24.Bd1 Qg5 25.cxd6 cxd6 26.Bxf4 Ngxf4 27.Qe3 Qh4 28.g3 Qxh2+ 29.Kxh2 Nxg3+ 30.Kg1 Rh1 mate

Nice combination, but surely not a great advert for Owen's Defence? Black appears to have a King's Indian Defence with a misplaced light squared bishop after seven moves, how are you going to sacrifice it on the kingside? Think White drifts a little with stuff like 11. Bc2, but he was fine anyway until 17. 0-0, which was simply suicidal. I would have thought moving the queen, maybe 17. Qe3, allowing the normal Nf3-d2-c4 would have been in order, perhaps together with f2-f3 and re-routing the dark squared bishop.

Yes the bishop is miplaced on move 7, that's why it goes to c8 on move 12 Seriously though I don't think the hippo is a very good opening. It seems the main benefit is to get the position into some kind of mess, avoiding all the standard lines,so White is more likely to make mistakes. In the above game for example Nick Faulkes is a 2100 level player. It was good fun to play though, even after getting hammered by two FMs, the grizzly evidence being online somewhere.

[Justin Hadi]

It's also not the same as a King's Indian because the black knight is on e7 rather than f6. More like a modern.

[Geoff Chandler]

Hi Roger

You are right. Here



AT RHP I've just had a couple of boffins who have more computers than NASA
run over this position.

They ignored the findings for 4...d5 and concentrate on 4...Bc4 or 4...Nxe4
The computers slightly prefer 4...Nxe4 over 4...Bc4.

Whole books have been written about 4...Bc4 (the Traxler/Wilkes Barre)
4...Nxe4 gets one sentence from Estrin (20 pages on the T/WB).

Giving that these beasts are also looking at 4...d5 and all the ramifications
that brings (Fried Liver, Fritz and Ulvstead variaitions as well as the mainline)
I'm suggesting the 'Horizon Effect' is at work here.

Though I expect to be told this no longer exists on the top programs
which my lads are running.

Interesting. Cheers.

[Paul Cooksey]

er, I almost feel guilty going back on topic. Korchnoi so impressed by Kasimdzhanov's idea he suggests removing his Fomer World Champion title

http://www.whychess.org/en/node/1527

[John Foley]

A problem with Kasimdzhanov's proposal is that it is an exogenous solution, like minimum move thresholds and stretching the points ratio.

What is required is an endogenous solution whereby the draw disincentive is an integral part of the game. I describe the concept of deferred draw acceptance in the latest issue of CHESS magazine. This measure would introduce some risk into offering a draw and is therefore a deterrent. The basic idea is as follows:

If someone offers a draw, their opponent can wait for a move before deciding to accept.

[Michael Jones]

Football has drawn matches. So does cricket. So do goodness knows how many other sports - and as far as I know, no-one connected with any of those sports is proposing drastic changes to the rules to ensure a decisive result is always reached. Why are Kasimdzhanov, Shipov et al so obsessed with eliminating draws from chess? Since most of the top players are fairly closely matched, we can expect that a reasonable proportion of the games between them will result in draws; why is there any need to alter this? The problem is not draws in general - it's short, uninteresting draws. The Sofia rules go some way towards countering this, although there are still some players who try to get round that rule by taking early draws by repetition (it wouldn't surprise me if we see 1. Nf3 Nf6 2. Ng1 Ng8 3. Nf3 Nf6 4. Ng1 Ng8 1/2-1/2 soon). Clearly this isn't good for chess, and it could be argued that the obviously prearranged games between the Kosintseva sisters bring the game into disrepute, but I can't see an easy solution to it; you can't ban moves that would repeat the position without fundamentally changing the nature of the game, since in many positions a repetition is either the only legal move or the only one which doesn't lose on the spot.

The worst idea I've heard on this topic has to be the suggestion that stalemate should be counted as a loss for the player to move. It would remove the need to learn large chunks of endgame theory, since almost any material advantage (in particular K + anything vs K) would be an easy win.

[Roger de Coverly]

Football has drawn matches. So does cricket. So do goodness knows how many other sports - and as far as I know, no-one connected with any of those sports is proposing drastic changes to the rules to ensure a decisive result is always reached. Why are Kasimdzhanov, Shipov et al so obsessed with eliminating draws from chess?

They seem to be proposing an American approach, arguing that the American approach to sport is anti tie. Have a look at some of the comments on chessvibes. You may recall they (Americans) wanted to do strange things to football, like make the goals bigger, so you had more of them and thus fewer draws. Actually some Americans seem to think that penalty shoot-outs are a normal part of football instead of an exception, being used only in knock out events.

Many American attitudes are crazy by British and World standards. One year the USCF they divided their Championship into two groups so that they would have a winner in each and then a blitz play-off for the title. That's like splitting the field in Sheffield so that Adams and Jones etc. were in one group and Short and Howell etc in the other.

[John Foley]

Since most of the top players are fairly closely matched, we can expect that a reasonable proportion of the games between them will result in draws; why is there any need to alter this? The problem is not draws in general - it's short, uninteresting draws.

The draw ratio is not reasonable for the top games: the vast majority of top games are drawn. We should expect a lower ratio of draws if all games were played in a competitive spirit. What differentiates chess is that draws can occur by agreement. If draws were only agreed in technically drawn positions there would be no complaints from those who contribute to the prize funds.

[Ian Thompson]

The draw ratio is not reasonable for the top games: the vast majority of top games are drawn. We should expect a lower ratio of draws if all games were played in a competitive spirit.

So what would a reasonable draw ratio be, and why?

Is your assertion true anyway? According to my database there have been 43 games played between 2750+ players this year (excluding rapidplay and blitz) and 44% of those resulted in a win. A similar analysis for 2700+ players gives 35% wins.

You could argue that including matches in these figures distorts them because a player should be much more interested in winning the match than any one game. If you do exclude matches then numbers become 37 games between 2750+ players of which 43% resulted in a win and 216 games between 2700+ players of which 38% resulted in a win.

[Geoff Chandler]

How about something that sounds totally stupid until you think about it.

Players are only allowed 2 draws per tournament.
After that any draw is counted as loss.

I wonder. Would you still get players burning up their draws after 12 or less moves?

Two players going into the last round with the pack breathing down their neck.
Both on a draw = a loss situation....sounds juicy.

Craziest idea I've in the last 10 minutes. But do ponder on it for one minute.
If such a rule was in place would that not jazz things up a bit.

No good of course for match play (give me another 10 minutes)
but in tournaments....knowing a draw is a loss....that would be fun.

[John Foley]

So what would a reasonable draw ratio be, and why?

A starting point would be the draw ratio between top software engines which is around 37%
e.g. http://computerchess.org.uk/ccrl/4040/r ... _pure.html

Computers are more objective than leading grandmasters and thus have a lower draw ratio.