How deep do you go in deciding a correspondence chess move?

Jose Capablanca once said when asked how many moves ahead he looked when playing an over the board chess game: – “Only one, but it’s always the right one.” Well, that cannot be entirely true, but perhaps he just knew by experience how to choose a good move. Garry Kasparov, when asked a similar question, said it depended on the position of each piece and said: – “Normally, I would calculate three to five moves. You don’t need more….but I can go much deeper if it is required. In a position involving forced moves, it’s possible to look ahead as many as twelve or thirteen moves.”

Of course, it is not just being able to look ahead three moves of one particular variation, there may be half a dozen possible moves and half a dozen possible replies to each move. So you can see that after looking at each variation for just six different moves and three moves deep you need to visualise some 46,656 positions (6x6x6x6x6x6)! That would be a brute force search and a player needs to be more selective to reduce the number of positions.  Alexander Kotov wrote a whole chapter in his book Think Like a Grandmaster entitled Do you Know how to Analyse? which concluded that three factors guaranteed finding the right move, namely, an accurate analysis of all variations that could be considered, confidence that you had taken account of all the best moves and strict economy in your thinking time. This is easier said than done, and Kotov gives a good method of how to train yourself to do this in his book. On a personal note, I did once organise a simultaneous display with Kotov at Stevenage, England in about 1979 when he visited his publisher. He played 20 boards, winning 19 and losing 1 to a schoolboy I believe. He helped me to map read when I drove him back to London!

Alexander Kotov

So this is what to do with over the board chess, but correspondence chess is a completely different experience. You are allowed to move the pieces between moves, which makes looking ahead much much easier. You are allowed to refer to books, databases and previous games as you play. In most games you are allowed to use computers to analyse positions, to find similar positions in master games and to look through endgame databases. Unfortunately, your opponent also has these at his disposal and I remember a game where my opponent suddenly announced to me that my position was lost, as he had checked it with an online six piece endgame database!! I have read that a recent World Correspondence Champion could spend several weeks on a single move to ensure it was the very best move. Even if you use a computer to analyse a position you should first decide on some candidate moves yourself without any help, then, at least, it is you driving the position rather than the computer. There is no doubt that you need to follow a plan, as well as find the best move to achieve that plan, which may change as the game evolves. Personally, I tend to move very quickly in correspondence chess. If I can see a reasonable move I usually make it and can’t see the point of dragging out games, especially if you are losing. Of course, if the position warrants it I will take longer, even looking some thirty or more moves ahead, but, unfortunately, there are then more positions than atoms in the universe…..!