How did you Improve your Strategic Positioning?

Just curious, how did you improve your strategic and positional playing (or were you unable to improve it), and whatever you did, did you notice a significant improvement in your rating? (By how much?)

Did chess books help? Looking over master games constantly? Memorizing openings? Obviously, optimally, you should do them all, but is there an easier way?

I've found this thread to be an invaluable source of information:

http://www.chess.com/forum/view/general/if-you-know

Yes, that is helpful, but my thread is different, that asks what is strategic positioning, I'm asking how you train for it, or improve it. It's hard to define strategy in a forum post.

Apart from the usual books and so on, KingsCrusher's YouTube videos have helped my understanding of positional play.

I think I got all my strategic knowlege from books and chess dvds/movies. Pachmans Modern Chess Strategy + Simple Chess by Stean are good aswell as Silmans strategy books. Also with some practice by playing more positional openings.

Interesting stuff, I'll definitely check it out. I have purchased 2 books recently, still on their way, Zurich International Chess Tournament, 1953 and Neil McDonald's Concise Chess Middlegames. Hopefully they will help.

I think you first have to study pawn structurs like crazy..

I recently came back to chess after being away from it for several years.  It has been a fun but difficult process of prying open some very rusty brain cells.  I am certainly no great player, but when I was playing before I read several of Jeremy Silman's books that were very helpful for expanding my understanding of the game.  I am working on his endgame book right now, and it is excellent.  It starts with basic forced mates and goes all the way through some fairly complex (he claims master level) engame situations.  I think his "How To Reassess Your Chess" is great for the basics of positional play.  I plan to reread that one soon as well, as I find I am in need of a review. 

First I had an intro to the "imbalances" as Silman calls it in his books How to Reasess your chess, and the Amateur's mind, how amateurs interpret them. Then, studying openings and thematic master games with them, annotating them quite heavily myself, was extremely helpful. I think it was more so the pawn structures (and how the pieces develop around it) in the openings than the openings themselves that helped, but seeing how the pawn structure (or lack of it even!) determines to a big extent what your plan is.

It improved my rating a lot, but only when I was able to back it up with tactics. I found if I wasn't blundering, then I would usually strategically crush my opponents.

Elubas, that's sort of my problem. I can develop fine, and I don't usually blunder pieces away, but when each side has developed and then begins the attack-plan phase, I try to make a plan to checkmate and or attack unprotected pieces, but this is something I do naturally, not based on some strategy.

Occasionally if I win, it's because the opponent blundered, made a mistake, or launched an attack that wasn't great---or that I launched an attack, the other player couldn't defend. But it wasn't a plan based on pawn structures, or "oh so he has a weakness here", it was simply, "what kind of checkmate can I do? Queen + bishop or Rook+ queen? Alright, let's give it a shot, let me put pieces in place" --- it's sort of like driving with blurry vision, just trying to see what works.

With that endorsement, maybe I will now read this book rather than stare at it and say "Who the hell is Znosko-Borovsky?  Boy does he stick out like a sore thumb among my early 20th century books by Capablanca, Lasker, and Tarrasch."

I eagerly await for more such secrets of strategic positioning, positional strategy, and stratositional posategy to develop my [super grand]master plan...

It just sort of gradually improved over time.  Silman's book were the first I bought so was introduced to imbalances and things very early.  Then a few hundred illustrative master games and some pawn structure learning later it's started to come together and make more sense.

Probably playing a lot and illustrative games helped the most.

I really don't know how much that increased my strength though, it's so gradual and other things are being learned at the same time.

This is why you should get a book like I or others mentioned (estragon's suggestions are great too), they introduce you to planning properly.

It just means the pawn structure is what suggests a plan, whether it's open or not, which outposts are good, etc. I would add that it's not just the central pawn structure that determines the plan, though it's probably the most important.

You can't deduce you say--- you've also said you can't infer in another thread by drawing general principles, or create abstractions. Then that means everyone is doomed to practice forever, and those super GMs are just the result of a matter of amount of games combined with luck. Then coaching is simply useless, and books only help with experience and won't really improve your game.

You can only teach chess by giving students the basic rules, and the conclusions and principles for students to remember and implement in all their games. You can't really teach experience. You can only improve a student's ability to infer or deduce ways to improve.

If what you say is true, then someone like Magnus Carlsen would never be able to reach highest rating (with the help of his country's best GM as a teacher), because it's all about experience according to you. It's that teacher's ability to communicate the lessons and principles that allowed Magnus to beat people who've had maybe 10000s of more games than him.

Am I mistaken? Am I mislead in my observation that all teaching revolves around induction and deduction?

Do you have alternate theories ? Am I being too optimistic?

I think NM Heisman somewhat agrees with me: http://www.chessville.com/instruction/instr_coach.htm

That's a funny mate, but really, I don't believe that it's a good idea to make those weird kind of openings, stronger players will rip you apart for wasting time with h4.

I think boring chess is fun...

I believe that anyone can develop a natural algorithm that can improve your ability to play chess. In fact, I know it exists. The simple fact that computers can beat grandmasters shows perfectly that there are indeed magic formulas to becoming a strong player (the difficulty is first finding the algorithm, and the second is actually being able to implement it in every game, the third being the memorization part, as most computers also have databases for openings/endgames).

Everything has a weight. Mobility, king safety, opponent's king safety, internal space, passed pawns mid-game, passed pawns end-game, pawn structure in middle and endgame... Bonuses for amount of squares your knights, bishops, and other units are attacking. Bonuses for double check, discovered check, queen check, skewers, pins, etc etc. Penalties for trapped units. Open and closed files.

Having helped in programming of chess engines, I am aware that you can break down many elements of chess and develop an understanding of position evaluation that exceeds other players. This intuition building is sometimes hard to describe for many players (which makes chess difficult to teach).

For a GM, it's pattern recognition (tactics and strategy), memorization of variations (experience), and positional and imbalance evaluation (material, positions, and potential for new positions).

Our real problem is, since we are such imperfect humans, we can't remember all our own lessons and ideas when playing a game in a timely manner. We forget things, we forget to look for opponent's best responses, etc. A chess coach would be very good at allowing you to build intuition and give you ideas on what to look for in evaluating a position.

By looking at a random chess position, with no immediate threats or tactics, a chess coach should be able to allow the student to guess what the plan should be for white, and why it would be feasible. For most average players, there would be unlimited numbers of plans, but only a few that would actually work. It's being able to tell the difference between what might work 10 moves down the road, and what won't, which tells you how strong a player someone is.

About the pawn structure stuff, sometimes the way I try to execute the "obvious" plan (due to the pawns) doesn't work against a stronger player -- not that the idea itself is unsound, but my timing is wrong, or their piece play is too good, or their "wrong" plan (considering only pawns) works. 

I guess you could argue knowing the thematic or even correct plans are as much as half the battle, but knowing how and when are most likely even harder because each position is unique and following a cookie cutter formula for plans might lead to OK play, but not good play.  (Not related to the algorithm side topic).

So I'd say that claiming pawns can give you the correct plan 90% of the time or even most times is too generous -- kind of like the advice "when you see a good move, look for a better one" you may find a good plan, but then stop and try to figure out the truth of the position.  To do this you have to consider the whole board.