Thinking of Checkmate

At what stage in the game should we start thinking of a checkmate stratergy?  From the beginning, mid game or near the end when only a few pieces left?  Or not at all and let the game progress to a natural end?  I find that once I start thinking of going for a checkmate, my ability to control the rest of the game slips sometimes quite considerably. 

Well, of course it happens some of the time that one or both of the players start an attack on the king, working with actual checkmate threats. For instance, in the Sicilian Dragon main lines where White castles long, both sides will try to open lines against the other's king and try to checkmate him.

But most games are decided in a more prosaic manner -- one of the sides wins material, trades down into an endgame, usually promotes a pawn and then checkmates the opponent. Usually the loser resigns some times before this whole process is carried out, when it's obvious that he can't do anything about it anymore. Even those Dragon games are quite likely to end with both sides avoiding checkmate, but one side still lost for another reason.

So it depends on the position, but in most chess games direct checkmate doesn't play a big role.

Like Scarblac said, it really depends on the position.  You should never just start thinking of it unless there are certain signs in the position telling you that an attack is possible.  Castling on opposite sides is often the most obvious sign that both players will now go for the throat of the other.  This is usually kicked off by throwing your pawns at the enemy's castled position.

Otherwise it's just like noticing a tactic might be available.  When a tactic is possible, first a player usually notices an undefended or loosely defended piece, or perhaps two important pieces, like the queen and king, lined up on a file or diagonal.  When this happens our tactic radar goes off and we can start to calculate.

In the same way, you shoud only really look to start an attack when one of these is present:  Basically either the king's position is loose (meaning there are weak squares around it that your pieces can coordinate to) or your attackers simply out number the oppoen't defenders on that side of the board.  If you do outnumber then you can look for sacrifices or overloading the pawn cover to force them to move -- when they move, it opens up lines which is the first condition I mentioned, a loose position.

Like you've noticed if you start an arbitrary attack, you can quickly loose control of the game.  If it looks like at some point there really is no attack available, don't be afraid to admit your idea was wrong and regroup.  Every level of player has to work on realizing when their plan isn't working and switching gears into a different idea.

Thanks for the replies, they are both very useful.  I can see that I am going to have to alter the way I play (and hopefully stop the collapse of my ranking). 

Hi Harldav,

For some time I have been training with Chess Tempo, and I now have a good instinct of when a forced mate is possible.  I would recommend the site to you, because --- among many other things --- you will learn to "see" when a forced mate is likely (or unlikely) to be present in any position.

Good luck!

One good rule, is make sure you CAN accompliish something before you attack. If all you will accomplish is making him dance a bit then turn it around easily, save both of you some time and figure out something else to do. A check that can't lead into something more is a waste of both player's times.