Of the 169 problems I'm studying lately, there are 72 problems that costed me rating points. Either by being too slow, or dead wrong.
From the first 3 problematic problems, I learned the following:
- I'm only partly aware of the changes that are caused by a move. I tend to focus on what a piece does on its new square, simply forgetting to look at what are the consequences of leaving the old square.
- I tend to store important stuff in Short Term Memory in stead of Long Term Memory. So that when I look at a new move, I'm totally forgotten the ponderings of the previous move.
But what about the remedy? How to fix these flaws in trial and error?
I already tried a few possible remedies.
- I tried to guide my attention by a thought process.
- I have developed the tree of scenarios, a set of standard scenarios for all kinds of positions.
- I developed the PLF (PoPLoAFun) system for guiding the attention
- Writing down all changes caused by the moves I'm pondering
- Trying to consciously store an important move or conclusion in long term memory
The diagnose is ok. Science says: we need more and bigger chunks and we should do deliberate practise, but not what and how we should practise.
ReplyDeleteThe the common improvement ideas dont work, at least not for adults.
What i try at the moment:
I have a sheme for solving/analysing tactical problems: weaknesses, methods, candidates, refutation, thinkingprocess. While i solve problems more "freely" i do my aftermath according this sheme, to find my weaknesses and to improve in patternrecognition. I learn the biggest "blunders" with spaced repetition and my aftermath-sheme and hope they become chunks somehow.
I hope that, as more i store in my ltm , as easier it get to put ( chessrelated ) and pull things (from) there.
I think i will include your "changes of a move" idea somehow in my aftermath-sheme.
A second thing i do is : moderate ( 10? min per day ) board-vision and visualisation training. For example this : http://www.chessvisualization.com/movegame_14.php ( and many other exercises ). This exercise for example should help me to store lines quicker into the ltm and i realy enjoy it.
A third idea is to learn the 1000+ 5x5 mini tactics of ct-art , they might be good chunks..?
PART I:
ReplyDeleteNM Dan Heisman, The Improving Chess Thinker:
How does one play "simple" chess?
1. Look at your opponent's move and try to DETERMINE ALL OF THE REASONS THAT MOVE WAS MADE.
[At Chess Tempo, the move prior to your first move is given, then it's your turn to find the "brilliant" solution. PAY CLOSE ATTENTION TO THIS FIRST MOVE BY THE OPPONENT! It is more important to determine ALL of the {direct and indirect} reasons this move was made than for any other moves, including your own.]
[Ask yourself this question:] WHAT ARE ALL MY OPPONENT'S THREATS?
Don't forget to look for discoveries, squares your opponent is no longer guarding, and so on. [A lot of stuff falls under that "and so on."] Often, an overlooked IDEA from one's opponent is one NOT directly involving the moved piece.
[Which seems to contradict the idea of looking FIRST at your opponent's move. BUT, you have to "see" ALL the reasons for your opponent's move, including ALL direct and indirect effects.]
2. LOOK FOR THE "SEEDS OF TACTICAL DESTRUCTION" FOR BOTH SIDES.
If you have a tactic, consider playing it. If your opponent has a tactic, strongly consider stopping it. If there is no tactic, WHAT ARE YOU TRYING TO DO? If you don't know, consider locating your least active piece and making it better. ALWAYS TRY TO USE ALL OF YOUR PIECES ALL OF THE TIME. Similarly, try to minimize the activity of your opponent's pieces.
3. START WITH CHECKS, CAPTURES, AND THREATS - FOR BOTH SIDES.
Look at what moves you might play (candidates), see what might happen after those moves, and determine which move leads to the position you like best. Always assume the best or most dangerous moves from your opponent.
4. IF YOU SEE A GOOD MOVE, LOOK FOR A BETTER ONE. [This correlates to what I previously wrote about switching your attention to a different move as soon as you "see" the first (and most obvious) move. The purpose is to avoid getting "locked on" to the first move that you "see".]
PART II:
ReplyDeleteThus we can summarize good, simple chess:
"First, see if there is a tactic for either side; if so, address it; if not, maximize the activity of your pieces and minimize your opponent's."
~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-
I used NM Heisman's numbered items, and reordered a couple of them. However, this does not have to be a step-by-step formal process. Well, perhaps initially it does, until you can drill this "process" (I would prefer to think of it as thinking "habits") into your subconscious. You will know you have succeeded when you no longer consciously think about the thinking "process"; you just do it automatically.
My suggestion is: do NOT to write "down all changes caused by the moves I'm pondering". Why not? Because the writing process is a distraction, diverting attention and focus away from the primary goal of just "seeing" and being aware of everything that is in the position. Yes, you will be able to check (after the fact) that you did not miss anything. But, it is very easy for the writing process to take center stage (since System II LOVES making lists).
The goal (as I understand it) is to learn to “see” better and faster.
I’m not sure exactly HOW to CONSCIOUSLY store information or conclusions into LTM. Memorizing the position and solution may (or may not) help future efforts. Unfortunately, we only have murky speculations as to what constitutes a chess “chunk” and virtually no information on how to selectively store “better” or “larger” chunks into LTM. That’s not surprising, given the lack of information.
When it comes to “chunks,” the usual attitude is that “I don’t know what a chunk is, but I recognize one when I ‘see’ it.”
ReplyDeleteI do not know if you already knew this stuff or not :). It is... Puzzle Rush: Chess.com's New Addictive Feature!!!
https://www.chess.com/puzzles/rush
https://www.chess.com/news/view/puzzle-rush-chess-coms-new-addictive-feature
The rules are simple:
- Solve as many puzzles as you can in five minutes.
- Each puzzle gets harder as you go.
- Three strikes and you’re out.
But right now (21st of November 2018), top players are trying to break 50. Last night, Hikaru Nakamura spent hours on a private Twitch stream trying to get there and couldn't break 49! Who will be the first to pass the barrier!?
I think we can try this out and make some conclusions? What do you think of this? :)
For me, this comes off as a joke. "How does one get better at timed tactics?" That's like, how does one get better at blitz or rapid chess. Answer: improve one's slow chess, or un-timed chess, which is what post-game analysis is, for example.
ReplyDeleteUsing "a method", to solve tactics more quickly - as opposed to how one solves tactics normally - strikes me as rather silly.
Improving in untimed tactics is easy.. just think longer. My standard rating is 2300, no problem to push it higher. A friend of mine was a 1600+ with a ct-standard raing of 2400, he did think sometimes several days about a puzzle. But i cant play a slow game where i can spend 8+ hours for each move.
ReplyDeleteAt the moment i think for improvement : Timed tactics are needed to find our weaknesses . Next step analyse deeply this weakness and i guess the next step is to memorise - repeat - make it a habit.
But.. as long as no one of us realy improved significant, we only talk silly
It's interesting you bring up the changes made by the last move as being a key factor. After similar observations, I put the question "What did the last move change about the position?" at the front of my self-generated simplified thought process. By notfocusing on this idea, I had kept overlooking important developments such as spotting key pieces or squares - my own or my opponent's - that were now left hanging or open to attack/occupation as a result.
ReplyDeleteI'm still not at 100% in remembering to mentally ask the question at the board every move. However, in the last tournament I played in, it saved me from blundering a piece as a result of a new long-distance pin on a diagonal - although I only remembered out of instinct to mentally check right before I had planned to move - and I went on to win the game as a result.
I will try to generate an improvement sheme:
ReplyDeletefor ever
-find own ( important ) weakness
-erase weakness ( for ever )
end for
procedure find own weakness
-for a while
--test skill in realistic situation ( with timer ..)
--collect situations where underperformance
-end for
end procedure
procedure erase weakness
-for each found ( highly ) problematic situation
-- deep analysis finding root of problem
-- spaced repetition on this aspect ( not just the move sequence !)
-end for
end procedure
Procedure deep analysis
-localize problem according to an ( shematic probably not even used or linear ...) thoughtproess
-find improvement within this sheme or modify thougtprocess/sheme
end procedure
at the present stage of discussion we need to modify the thinkingprocess by something like : at each move realise modifications by this move