# Best heuristic for memorizing positions in chess

Opening variations may be memorized by recalling what the objective behind every move is. For example, the first few moves of the Ruy Lopez fight to control e5.

In the middlegame, we may find ourselves thinking about a critical move - a move that requires tactical precision. Said tactic, if performed correctly, will yield an improved position. The tactic may reach several moves, usually six or seven and occasionally ten moves.

We would like to memorize a position in the middle of our visualized tactic as a "mental checkpoint". In case we don't like a move, say our eighth move, we can return to the position after the 5th move in our visualized tactic instead of the current position (move one).

1. What is a consistent method for creating the checkpoints?
2. Next, where in the tactic do we set the checkpoint? Do we set it every fourth move, do we only create one, or do we only create it in quiet circumstances?
3. If the method is not consistent, such as it does not work for all positions, list the exceptions.
• What does "heuristic" have to do with techniques for memorizing chess positions?? What kind of trial-and-error process do you have in mind?
– bof
May 7, 2017 at 21:57

This seems a bit theoretical/artificial to me.

I'd say, just do whatever works best for you. In the end all this should happen subconsciously without you thinking about any "methods".

1. What is a consistent method for creating the checkpoints?

Do you mean, how to memorize a position 5 moves from now? For any reasonably strong player this is not a problem (could be more problematic with positions very far in the future). You could use similar methods as in blindfold chess, e.g. you could combine groups of pieces (e.g. store the information of "fianchettoed bishop" instead of the position of each piece separately). Also for tactics very often only a small part of the board/limited number of pieces are relevant; so you could ignore irrelevant pieces.

1. Next, where in the tactic do we set the checkpoint?

Could make sense to create them at points where you have a number of alternative moves. Also, could create them at positions that are easy to remember.

• I like the conglomeration of pieces into a bundle. Having to memorize f8, g8, h8, Kg8, Rf8 instead of "castled black king" would be a pain. The idea of ignoring irrelevant bits of information will also help with memory access. Lastly, creating the checkpoints during a time with lots of possibilities seems excellent to me, because it allows the most vision. May 6, 2017 at 1:30
• bundling groups of pieces is called "chunking" by psychologists. A chunk need not be localised. Functionally Bba8,Wnf3,Wkh1 is a chunk May 6, 2017 at 2:52

I agree with `user1583209` that checkpoints should be created

at points where you have a number of alternative moves

and that memorizing groups of pieces instead of each piece on it's own is probably the best method chess players know.

BUT:

I think that most players don't memorize positions (while calculating) at all because they don't need it. I can't agree with your statement:

The tactic may reach several moves, usually six or seven and occasionally ten moves.

Usually you have to calculate shorter variations (between 2-4 moves).

Imagine that you need to calculate variation that is 4 moves long and after the second move there are lots of possibilities. It wouldn't be very efficient to memorize the position, since only two moves are played and you can easily replay the variation in your mind.

On the other hand it isn't bad thing if you try to improve yourself in visualization and memorizing visualized positions, since it improves your overall chess vision.

That means I'm not completely against this idea (if you use it for example as a training method), but I think it's not so useful in practice.