Is chess a matter of understanding or memory?

[stevencarr]

Obviously the answer is 'yes'.

But which is more important? Or are the two inextricably linked?

When people train for the Asian game of Go, they memorise 100 professional games , they memorise tsumego (life and death problems).

Go is a very complicated game (deeper than chess in all honesty), and yet there is a place for rote memorisation in mastering it.

I keep reading that a top GM knows thousands and thousands of patterns of chess pieces.

The human long-term memory capacity is essentially infinite.

I have software which helps me remember huge amounts. It schedules prompts at appropriate times to prevent forgetting.

So here are two strategies for studying chess.

Strategy 1) Analyse some positions. But don't people forget their analysed variations after a week or so? Does that matter? If you learn to analyse, does it matter that you forget the results of individual analysis sessions? Just as if you learn to ride a bike, does it matter that you forget exactly where you rode it while you were learning?

Strategy 2) Memorise positions. Will that improve understanding? Or is chess such a game of recurring patterns that memory will play a useful part?

I have scraped 10,600 positions from various web sites and have programmed them basically as flash cards.

I intend to learn them by heart to the extent that I can tell you at once the answers to 90% of ones which come up. That is the standard the memory aid software can keep you at.

At the least it will be an interesting experiment.

Will my strength improve as a result of rote memorisation of chess positions?

I can keep you posted.

[Roger de Coverly]

Will my strength improve as a result of rote memorisation of chess positions?

Are you prepared to divulge what it is now? It's variable by player, but as you climb the rating ladder, the knowledge of what to do in a given position increases. So learning positions is more likely to be beneficial to an "average" player than someone higher up who should be able to solve them from scratch, possibly in an original manner.

GMs can sometimes come up with powerful over the board novelties. That isn't always because they've prepared them earlier, but sometimes because they've either misremembered the theory or never knew it in the first place.

[stevencarr]

My official grading is 157 at present, from games I played 4 years ago. It used to be in the 190's about 1997.

I played in the Heywood Major 10 days ago and made 3/4 with a grading performance of 195.

Only 4 games, but I was very pleased with my results as I hadn't played a serious game for 4 years.

The website I trawled for positions to learn was http://wtharvey.com/


At the time of the congress, I had memorised about 1050 of the combinations linked on that page.

[Roger de Coverly]

I played in the Heywood Major 10 days ago and made 3/4 with a grading performance of 195.

At that sort of level, it's whatever works best for you as a training method. Tactical sharpness is obviously good provided you can reach positions where a favourable tactic is possible.

[stevencarr]

Teichmann said chess is 99% tactics.

Of course that was a long time ago.

[Dan O'Dowd]

Interesting discussion. For my money, there's an upside down bell curve effect on this. I should first declare though that I have probably the most natural store memory of anyone on the site - I absorb languages like a sponge and remember so many of my old games in full as well as horribly mundane details about my life lol.

My intuition is that if a player at the lower levels had more understanding than memory, we'd have to presume he'd fudge things a lot. This is basically because seeing God's face up the stairs, (understanding what to do in a position) without knowing where the stairs are (how to get there), is rather difficult. You will invariably find that such players miss the finer details of the positions they are dealing with; be those tactical points within combinations, or be those positionally deep moves that either belie general understanding or are beyond the veil of human chess without some deeper digging.

On the other hand, of course, it's pretty easy to show that the same lower level player when armed with a huge dose of memoritic openings, technique in very basic endings etc, will still struggle in a fair sample across all games. I know someone who could be said to fit this pattern very well: in the two games we have shared, he bashed out so many moves of theory that I decided to take a draw because while I was able to calculate the complications, it was only giving diminishing returns versus my clock. There is another player I know who has this down to a fine art in certain openings, which gives him the 130-140 bluff skill factor that has practical returns for his money.

I took an extremely long time to get anywhere near decent at the game. Some say I still am not that decent Despite being blessed with a computerlike memory, and always as a junior looking for the positional understanding of things (in fragments, though, and I'll admit this was a lot of the time, poorly taught dogma from books, such as *Doubled or isolated pawns are bad* etc), we could say I lacked true understanding, since it was only a few years ago when I started looking at the nitty-gritty of how the positional factors meshed with tactics (we can see a relation here to having God's algorithm on the macro against the micro of understanding all the steps to get the right result from it, or with machine/assembly code, or even binary), that I began to round out as a player and be capable of good judgement in unclear 'pure' positions where some positional factors were suspended, as well as starting to be able to use that to foster a deeper understanding on a positional, Botvinnikian, grand planning level, where you still have to see the pitfalls around each step you take.

At the top level of course most players have a huge memory though some like Carlsen and Capa were famously lazy with its application. But it's not really right to use them as an example, since we can argue that most of the time, their play is so strong that they exhibit what I might term post-positional thought. To wit, while they indeed have the full understanding of their peers and history of the game, they are able to play the purer positions, and look at everything with fresh eyes and be objective.

If we look then at a rank and file Grandmaster, we very often see reams of variations, though I would argue very slightly in favour of the idea that their calculation ability is more borne of repetition and ability than core memory itself. However, at this level as opposed to the jungles of the 2200 even, you can see a marked difference in technique - that is, core knowledge of how to play out positions where a game is won but requires some manoeuvring, or in general strategical progression in endgames. This to my mind is a key point. Even the hardest-working player will struggle to absorb everything he needs, without a general ability to memorise. And after a game, all players above a certain point will be able to explain in fluent verbalisation, their thoughts and ideas (this again is one reason I excluded the top echelon, since it is probably the case that sometimes their thoughts feature pure chains of moves, albeit rarely).

There have been times though that when I'm playing as close to fluently as I can (fluency with application of that which I know), I look at a game afterwards and simply cannot remember or consciously explain the thought process I used to choose or even verify certain moves. For this reason and the above I would say there is a core linking of memory and understanding that goes on in all players. A more telling question might be to ask or investigate what level of memory is required of a player whose only task would be to learn technique. That is, scrap the game itself, give them the necessary 2000 level or whatever of pre-requisite tactical ability, and demand them only to learn to be able to beat a certain level of player (2400?) in positions that are deemed technically won that aren't too complicated. Is that only a task for working memory fueled by understanding of underlying precepts? Is memory simply a way we humanise our subconscious processing that continues to defy the understanding of the metal monsters who still can't play strategical chess properly sometimes?

[Geoff Chandler]

Hi Steven

"Teichmann said chess is 99% tactics." Of course that was a long time ago.'

It's still relevant advice for an up and coming player today.
If Teichmann said that at nine o'clock this morning you would not get many arguing.

They might lower the % marker but it is still a large part ofthe game.

Never tried the flash card method from a screen so cannot judge it.
The danger is you may become just very good at solving these 'puzzle a day' things on other sites
But that is like getting a lft to the top of a mountain with a helicopter.
You managed to do clamber up the last few feet and plant the flag, but you did not climb the mountain.

Stronlgy suspect you are not a bad player, the lay off has not done much harm.
An ex-190 player getting dropped into a pack of 150's in a major even after a four year lay off will do well.

If you have a reasonable positional sense then you never lose that only your ability to
calculate messy tactical variations suffers if you don't keep your eye and you appear to be addressing that.

Post your best win and that loss. No notes and no computer jargon either.
(give the rough ages of your opponents and your score at the time of the games + time control.)

There are a stack of good players here most of whom give very good advice.
(when they are not squabbling with each other.)
There is every chance you will pick up a freebie. A wee pearl, a gelling agent that that joins a loose end.

Is Chess a Matter of Understanding or Memory?

I often think of what my wife (who knows nothing at all about the game) said once
when I returned home with a handful of chess books from a 2nd hand stall.

"Why do you need more chess books. Surely you know by now how the pieces move."

That is all you have to remember. How the pieces move.
The rest of the game is fully understanding why you are moving them to a particuliar square.

Picked up that Levy and O'Connell book - Oxford Encylopedia of Chess Games 1485 - 1866 for £1.50 not so long ago.
(Charity shops boys - never walk past a charity shop.) Been playing over a few games at random.
Even though he was not yet born I can see where Teichmann got his quote from.

[stevencarr]

Memorising 10,000+ positions is hard work. It is something like 150 combinations a day.

[Joey Stewart]

Why would you want to spend so much time "memorising positions"?
Sounds like a good way to put you off ever playing again.

If you want to improve just follow thus simple technique:
1) pick a move you want to play
2) decide what counter moves your opponent can play in return
3) look at the options that now arrive and see whether there are favorable moves for you.

If this does not work go back to step 1 and start again.

[stevencarr]

Why would you want to spend so much time "memorising positions"?
Sounds like a good way to put you off ever playing again.

Because I want to see if I can memorise so much stuff.

I also prefer training to playing.

For one thing, when you are playing there is another guy there who is doing his best to stop you. That can be very irritating sometimes.

[Paul McKeown]

You will hear "chess is 99% tactics" from time to time, but it's just lazy sloganising. Computers are well on the way to demonstrating that it's actually 100%. You're not a computer, of course. Humans learn where the pieces go in typical positions (Smyslov called it "harmony") and try to make it work from there.

[stevencarr]

Memorising 10,000+ positions is hard work.

There is something like 180 combinations to solve each day.

Ticking off the last 20 each day is a real grind.

[David Gostelow]

I have a memory like a fish and another thing, I have a memory like a fish. I just about make an above average player. Keith Arkell says he has a bad memory. I could not replay a single one of my games. I play next to a guy who will say to me something along the lines of "Why did you play that move when last time you played a different move when playing that player (IE he remembered my game and player) while I cannot recall I have even played him before. Memory is useful , particularly in sharper lines and maybe end games , but not as important as understanding positions , pattern recognition etc

[stevencarr]

I have a memory like a fish and another thing, I have a memory like a fish. I just about make an above average player. Keith Arkell says he has a bad memory. I could not replay a single one of my games. I play next to a guy who will say to me something along the lines of "Why did you play that move when last time you played a different move when playing that player (IE he remembered my game and player) while I cannot recall I have even played him before. Memory is useful , particularly in sharper lines and maybe end games , but not as important as understanding positions , pattern recognition etc

Memory is useful, especially when you have seen a combination 4 times before, and still can't remember what the winning line is.

As I said in the opening post, people training to play Go memorise a lot. They memorise 100+ professional games and thousands of life-and-death problems.

Although Go is just as pattern based as chess is.

I want to do an experiment to see what concentrating on memorisation has on chess strength.

[Joey Stewart]

Could you provide some of your results, thus far, I would be interested to see how it is all going and how you feel the challenge is improving your chess.