Pawn defence

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Pawn defence

Post by Lyudmil Tsvetkov »

Well, this is a very important rule.

It is always well when your pieces are defended, primarily because of tactical considerations, but engines usually do not consider this, as it is expensive. And indeed, defence among pieces themselves might be skipped, leaving to the search to do its job instead, however, I think pawn defence is an absolute must with a big added value.

Recently I saw such rule being successfully implemented in Texel, so I might be accused of stealing other people's ideas, but that is not so important.

A reasonable implementation might run as follows:

- not make a distinction between the specific rank upon which the piece is defended, as this would probably already be covered by some aspects of pawn psqt, but necessarily distinguish between the different pieces being defended
- I would give some 10cps bonus for minor piece defended by pawn (or better 11cps for the knight and 9cps for the bishop), 5cps for rook being defended by pawn, and just 3cps for queen defended by pawn, both for the middlegame and the endgame

Of course, as mentioned, the bonus is due out of purely tactical considerations, as this usually saves you time and gains valuable tempos, for example when a piece is attacked and not defended by a pawn, it will have to either retreat, or seek own defence, which loses a move, while if the respective piece is already defended by a pawn, you might simply neglect the threat, as capturing will result in recapturing and not losing material, and calmly proceed with other relevant moves able to improve your game.

So this rule is very important and very true at the same time, I almost see no exceptions to it, but strangely, as far as I know, not too many engines make use of it.

In case someone argues that would be redundant with outposts defended by pawns, please rest assured that the redundancy rate would be small, as minor outposts defended by pawns constitute not more than 20%, and more likely somewhere 15% of all possible pawn defences.

Of course, the implementation would be, you find a pawn somewhere, and on the rank in front of it to the left or to the right you find an own piece, then you give the bonus for the respective piece.

Pawn defence is possible on ranks 3 through 8 and is important everywhere.

Anyone else apart from Mr. Oesterlund having succeeded with a similar idea?
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Pawn defence

Post by Lyudmil Tsvetkov »

To post one or 2 diagrams to get it going.

[d]6k1/1pN2ppp/1P1b4/r3n1q1/R7/1P1B2Q1/2P2P1P/6K1 w - - 0 1
This is just an illustration, and not a position to evaluate.
No matter whose turn it is to move, white does not care about any possible black captures, as all of its pieces are defended by pawns, Bd3 by the pawn on c2, Ra4 by the pawn on b3, Nc7 by the pawn on b6, and Qg3 by the pawns on h2 and f2.

Black's pieces, on the other hand, are all not defended by own pawns, so white is much better here in terms of purely tactical considerations.

[d]1r4k1/1r5p/3p2p1/1RpPp3/p1P1P3/P5P1/7P/1R4K1 w - - 0 1
The fight for the open b file is won by white, precisely because the rook on b5 is defended by an own pawn, c4, while the black rook on b7 not.
This might decide the outcome of the game.

[d]2Rr1rk1/1P4p1/7p/4b3/3p2P1/7P/6B1/2R3K1 w - - 0 1
Pawn defence is important also on the 8th rank.
If it were not for the b7 pawn defending Rc8, this might end in a draw. Now white wins.
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Pawn defence

Post by Lyudmil Tsvetkov »

[d]4r1k1/pp1nrpp1/1qpb2p1/3p4/3P4/1QP1RN1P/PP1N1PP1/4R1K1 b - - 0 1

Quite often, pieces defended by pawns exhibit another valuable tactical asset, namely that it is usually not advantageous for an enemy piece attacking such a piece defended by a pawn to capture it, as in this case the pawn structure of the side with the defended piece will improve.

Statistically, I think a similar scenario happens in around 70% of all cases, for the reason that the pawns defending the pieces are usually placed on less central files, as the central pawns quickly go forward, and most of the pawns remaining to defend own pieces are less central. Also, pieces generally strive to occupy more central locations, so this again makes them likely to be defended by less central pawns.

Above, neither white nor black could cature on b6 or b3, as in this case the pawn structure of the side with the captured piece will improve, with a more central pawn, from a to b file, while the weakness of the doubled pawn here will be unnoticeable given the location of the remaining pieces and that it would be connected to other own pawns.

For the same reason, black rook capturing on e3 would severely improve white's pawn structure, as the f2 pawn goes to e3, becoming more central, while at the same time the relative pawn symmetry on the king side, would be broken, that would additionally threaten black's chances to achieve a draw. So the white rook on e3 being defended by a pawn is a significant asset, and black must avoid capturing and look for ways of defending its own rook on e7, which shrinks the choice of reasonable black moves.

For comparison, black can not place a rook on e6, as this would damage strongly its pawn structure, but that is purely tactically related.

[d]1r2r1k1/2p2p2/1p1pb2p/5np1/8/1BPPN1P1/2P2P1P/R3R1K1 w - - 0 1

Same here, neither side should capture either on b3, e3 or e6, as in this case more central pawns are created, or weaknesses disappear.
Still, white has better chances to gain something from the complications, as it has 2 defended pieces, Bb3 and Ne3, while black a single such piece, Be6. Meaning that, white can worry less about enemy attacking moves and concentrate on moves improving its position instead.

So I really would not skip a factor like this in eval.
User avatar
velmarin
Posts: 1600
Joined: Mon Feb 21, 2011 9:48 am

Re: Pawn defence

Post by velmarin »

With all the respect we all deserve.
This is for beginner player. look for something with more substance or more flavor.
There is no motor that does not see a pawn on seventh.
Except for Skipper.
Henk
Posts: 7218
Joined: Mon May 27, 2013 10:31 am

Re: Pawn defence

Post by Henk »

velmarin wrote:With all the respect we all deserve.
This is for beginner player. look for something with more substance or more flavor.
There is no motor that does not see a pawn on seventh.
Except for Skipper.
At least he is right about Skipper.
Masta
Posts: 28
Joined: Fri Jul 26, 2013 6:24 am

Re: Pawn defence

Post by Masta »

velmarin wrote:With all the respect we all deserve.
This is for beginner player. look for something with more substance or more flavor.
There is no motor that does not see a pawn on seventh.
Except for Skipper.
The rule isn´t about a pawn at the seventh. You are just ignorant. It is a general rule that can occour any place at the board.

Manners when arguing against someone that already gave many successful contributions to chess engine projects.
User avatar
velmarin
Posts: 1600
Joined: Mon Feb 21, 2011 9:48 am

Re: Pawn defence

Post by velmarin »

Masta wrote: You are just ignorant.
I am an ignorant and have no manners.?
It was a joke.:P :P

But that does not give you reasons, all engines have know what is worth a promotion, or a pawn candidate. Nearly all have support for other pieces.

You would say that you are idiot or stupid,:evil::evil:
but I will not say it is, you are very ready Yes Sir, congratulations
Masta
Posts: 28
Joined: Fri Jul 26, 2013 6:24 am

Re: Pawn defence

Post by Masta »

velmarin wrote:
Masta wrote: You are just ignorant.
I am an ignorant and have no manners.?
It was a joke.:P :P

But that does not give you reasons, all engines have know what is worth a promotion, or a pawn candidate. Nearly all have support for other pieces.

You would say that you are idiot or stupid,:evil::evil:
but I will not say it is, you are very ready Yes Sir, congratulations
I didn´t insult you...ignorant means what it is...you ignores some knowledge.
That´s what I wanted to say...and yeah, again you are being ignorant.

I´m not insulting you. :?
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Pawn defence

Post by Lyudmil Tsvetkov »

Masta wrote:
velmarin wrote:
Masta wrote: You are just ignorant.
I am an ignorant and have no manners.?
It was a joke.:P :P

But that does not give you reasons, all engines have know what is worth a promotion, or a pawn candidate. Nearly all have support for other pieces.

You would say that you are idiot or stupid,:evil::evil:
but I will not say it is, you are very ready Yes Sir, congratulations
I didn´t insult you...ignorant means what it is...you ignores some knowledge.
That´s what I wanted to say...and yeah, again you are being ignorant.

I´m not insulting you. :?
Thanks for the moral support, Alexandre!

Are you Brazilian or Portuguese?

There are 20 times better chances that you are Brasilian, but still... :)
Lyudmil Tsvetkov
Posts: 6052
Joined: Tue Jun 12, 2012 12:41 pm

Re: Pawn defence

Post by Lyudmil Tsvetkov »

velmarin wrote:With all the respect we all deserve.
This is for beginner player. look for something with more substance or more flavor.
There is no motor that does not see a pawn on seventh.
Except for Skipper.
Hi Jose.

I bet Bouquet could gain 10 elo with such parameter added, if it is well tuned. :shock:

Think about it, most engines have a term for pawns attacking enemy pieces, and derive measurable elo gain from that.
As far as I know, most engines also do not have a specific term for pawns defending own pieces, but the 2 are closely related and strictly non-redundant.

If the first term gives you some added value, chances are the second one could also do so. I do not know if it will work and where, but it is a reasonable eval term.

Eval helps to pick the right moves, and on occasion it matters if a piece is defended by a pawn or not. Couple of centipawns more or less could distinguish between different positions. It might not tell immediately, there might be not tactical resolution in sight, but the difference in eval will certainly tell at some point.