Positional Advantages King Safety

It is clear that king safety is very important. If the king is not safe, then sacrifices of material can open further breaches to win material and/or produce a mate.

There is little I can add to what is found in any good book about the subject of king safety. In our computer work it has become clear that if there is only one pawn available to shelter the king, then the best pawn is the central one of the triad, the b-or g-pawn. Also, one can have a

wm ifl if



///A'.-/, '////-V/'

ép mf

ép m///: '■://////.

m, i®

HA if

All AW

ff A

rm " pi

n SI

Bishops enhancing king safety weak pawn phalanx around the king, if one has board control over that area. One subject that is seldom discussed is how pieces other than pawns can shelter the king. It is clear that an f3-knight is useful in defending a white king's castled position. However, bishops can also be used in king safety, especially when the pawn structure has been ruptured. Thus, all the positions in the above diagram show ways in which a bishop can augment king safety.

Pawn Structure

The System detests isolated pawns, and you should too. It is not so much that isolated pawns have to be defended by pieces; it is much more that the squares in front of an isolani can no longer be attacked by a pawn. The System favours a board control strategy, and when there are important squares that can no longer be controlled, the strategy has probably failed. Isolated pawns may be all right if they are passed; particularly if they are centre pawns and have advanced beyond the 4th rank. Isolated pawns are also not a serious disadvantage if they are a- or h-pawns because the squares in front of them are not very important. Side-by-side pawns on the 4th rank such as on d4 and e4, or c4 and d4 are very good. They control a significant part of the board, and as such are worth much more than they amount to individually.

Pawn structure is very nicely described in Kmoch's wonderful book Pawn Power in Chess. It does not deal very quantitatively with the value of pawns, but it does describe pawn interactions extremely well.

As already discussed, pawn structure significantly affects the value of individual pieces. In blocked positions, a knight is probably more valuable than the worse of a bishop pair. So one should manage pawn structure to suit the pieces, and vice versa. There are several examples of this in this book.

It is frequently possible in a position of average openness, to make decisions that affect whether the future pawn structure will be closed, blocked or open. Such decisions should be made with the pieces on the board in mind. If there is some possible exchange pending, it is good to go for the pawn structure that favours what will be left. There are frequently opportunities to exchange a bishop for knight, and then partially close or block the position. This is like enhancing the value of the remaining pieces, and is a distinct advantage that should be sought out when possible. The same kind of thing happens in bishop vs knight situations, where the side with the bishop should be interested in opening the position up, and exchanging centre pawns for wing pawns, while the side with the knight should try to keep the pawn structure as closed as possible and avoid pawn exchanges.

Board Control

The thing that distinguishes The System from other approaches to the chess opening is its emphasis on board control. Chess is a struggle that involves all the pieces. Historically, good players beginning with Morphy learned to get their pieces out early in the game to be prepared for the struggle. However, as sophistication set in, the best players began to realize that just getting a piece out was not enough; one should try to get it to its best location. The hypermoderns realized that sometimes a piece is extremely well placed on its original square, as for instance the c8-bishop is in the King's Indian Defence.

If one side controls most of the important squares in the centre, it will be increasingly difficult for the other side to develop his pieces to meaningful locations. Of course, there is a trade-off here. One cannot exclusively make square-controlling pawn moves without falling far behind in development. If one gets far enough behind, the other side can make sacrifices, and break the controlling bind to its advantage. So White must combine the strategy of board control with a certain amount of development to avoid getting too far behind.

Board Control is the most important advantage, assuming the king is secure. The hypermoderns built their theories on how to control and attack the centre. However, we will show much new ground to be mastered. The most important area of the board to control are the four central squares, and secondary to these are the squares c4, c5, f4, f5 that surround them. One important facet of board control is colour complexes. One should be very careful in exchanging a bishop, as this could be the primary defender of the squares of its colour. True, pawns can also defend those squares, and sometimes even other pieces. However, the bishop is primary and to exchange a bishop that is guarding important squares of its colour, can be foolhardy.

White is half a move ahead to begin with and can use this advantage either to make a controlling move which does not increase his development, or to make a developing move. Hopefully, he can do both at the same time. In any case, he should be careful to preserve his options, so that important future controlling moves will not be blocked. This is one of the most important principles in System philosophy.

The major thing that distinguishes System philosophy from previous opening theories, is the firm commitment for White to control the centre on his side of the board; i.e. the squares d4 and e4. Historically, White has been content to just get his pieces out, and then expect active play.

However, The System does not want to give up e4 to Black unless White gets something concrete in return. To this end, the move f3 frequently comes in after White has already established his grip on d4, and now wishes to control e4 also. A move such as f3 in the opening could very well be made instead of developing a piece, in the hope that the board control achieved thereby is worth the loss of time. However, as previously mentioned, White must be careful about how far behind in development this makes him. We show a number of examples of this critical concept.

Development: The Placement of the Pieces

Piece Placement is well covered in classical chess texts. The hyper-moderns discovered that the classical notions of piece placement were not completely adequate to evaluate many positions. We build on this in our theories, as we describe how to reach the maximum dynamic potential of a position. Also, classical notions of Development seem rather antiquated now. Yes, one must get the pieces out; however, not slavishly but rather in a very organized way that balances the need for getting the pieces out while at the same time asserting one's influence on the board.

Concern with development is, together with board control, the primary concern of The System. One must try to get the pieces out, but only to their best squares, and with the aim of controlling the board. So how does one improve upon the slavish development ideas of the Classical School of Chess?

There are two ways of looking at development:

a) The Classical view is: How many pieces has each side developed, counting 1 point for each one and then taking the difference between the development of each side.

b) The Dynamic view is: How many pieces still need to be developed, counting -1 point for each one and then taking the difference between the lack of development of each side.

According to Classical ideas, 1 unit of development is assigned to each piece that has left the back rank. Castling gets 1 unit, and a rook on an open or semi-open file also gets 1 unit. However, we have found that this system leaves a lot to be desired. There are situations in which a piece on the back rank is as well developed as it can possibly be. Further, there are many issues about pawns being developed (on which Classical authors differ), and whether to give more than 1 unit for (say) a knight on d4, or for castling. We use a different system, of my own design, which is both simpler and more accurate in assessing development.

In the Dynamic method, the development count is -10 for each side at the start. From this we count down:

a) 1 point for each piece that is in a good developed position (including those that have not moved yet!);

b) 1 for having the king in a safe position;

c) 1 for each of the two centre pawns advanced beyond their original squares;

d) 1 extra point for a knight in the centre, as long as it cannot be driven away, and White has already developed over half his pieces.

A rook is considered developed if the pawn of its own side in front of it is advanced at least to the 5th rank (or gone altogether). While castling gets 1 unit of development, one is well advised to heed the maxim Castle if you must, or if you want to, but not because you can! The point is that there are positions where the decision of where to put the king is best postponed until the character of the position becomes very clear. There are many examples in this book where the white king stays in or near the centre because that is the best place for it.

The question of what is the best location for a piece is frequently touched on in this text. However, it can be quite difficult to determine whether a piece is 'developed'. I have given the rules and obvious exceptions above. We have already mentioned the black c8-bishop in the King's Indian Defence that participates from here as well as from potentially any other square. However, how about a move such a sbf 1 to remove the king to what is considered the best location in the position (this move could also be forced). Is this a developing move?

The easiest way to look at these problems is simply to count What still needs to be done? If certain pieces are not participating, then they are not developed, and should get a -1. Pieces that appear to be optimally located (even if they have not moved yet) are considered developed. Also, rooks on their original square that are on an open file, or on one that is threatening to be opened are not in need of development. If a rook is 'undeveloped' but has no good location to go to, then this is still some unfinished business, but not something that must be attended to right away. The same could be true of a queen.

One other aspect of this method which is more accurate than the classical way is in dealing with sacrificed pieces. In the classical scheme a sacrificed piece is undeveloped; i.e. it does not have a development count associated with it. However, presumably it was sacrificed on purpose to gain in development of some other advantage. Therefore, it is foolish to reduce the development count because of its absence, rather than to think of how many units of development are required to bring those pieces left on the board to their optimal location.

There is one other thing about development that is important. The mere difference in development of the two sides is not always indicative of what is going on. For instance, if one side has a development count of -6, and the other has -4, then the difference is 2 units. The same would be true if one side had a count of -2, and the other a count of 0. In the latter case, one side is completely developed and ready for action, while in the former, both sides still have a way to go. Clearly, if one side has completed development he is in a position to start immediate action, and this can be very dangerous to the opponent. In the first example, the difference in development is the same, but the application of this difference is still a few moves away. So some ratio of completion of development would be useful. Since the attack usually starts once one side has completed his development, the closer one side is to completing his development, the better is his position.

When considering a pawn sacrifice for gain of development, the standard rule of "A pawn is worth three tempi" is a good one. However, be very wary of winning a pawn by giving 3 tempi. That is seldom wise in the opening.


Everyone prizes mobility (and it is important in every game I know of). However, in chess it seems to be only a good tie-breaker, all other things being equal. For instance, which is better:

a) A rook that can move to 10 different squares, or b) A rook that can only move to 8, but is attacking a backward pawn?

Clearly, the latter is better, which shows that pure mobility (the ability to move to a square) is not a good enough notion. One must consider what squares, and what is on those squares, and how central they are, etc.

All this shows that the above notions must be merged into some larger whole to get a full appreciation of a position. Since mobility is frequently tied to such other notions as whether a piece is in a defensive position, it is usually satisfactory to ignore it. That is what is done in the best chess programs.

2 The System Principles

0 0

Post a comment