Sunday, June 28, 2020

Answers to Robert


Robert said:
I have seen numerous assertions (such as from Garry Kasparov) that planning involves (at its best) considering what will happen within (approximately) the next 5 moves, sometimes (often?) fewer than that.

Isn't planning essentially looking at the general situation and determining (through the process of elimination via calculation) which sequence of moves provides the best probable outcome? If that is so, then PopLoaFun seems to provide an excellent basis for determining a (shorter term) plan. A plan for the entire game seems to be too general and abstract for practical use move-by-move.


Planning starts with WHAT should I try to accomplish.

Followed by HOW can I do that.

It is too early to say how big the role of PoPLoAFun is going to be in the endgame. About the middle game, I am inclined to say that 25% of the moves are tactical by nature, meaning that all three parts of PoPLoAFun are directly concrete involved. The remaining  75% focusses on creating lines of attack, closing them for the enemy and wrestling for domination on the existing lines of attack. How you exactly are going to use the lines of attack is of later concern.

Are endgames particular (concrete; tactical) or general (strategic) in nature?

If I'm not over ambitious, read I don't throw the kitchen sink at move 3, I will get a middle game. I put a lot of effort in investigating the middle game. After a few months I reached the final formulation of middle game planning: Get the most piece activity against the least activity for your enemy. With this simple rule, I can judge any middle game move.


With my new approach, I am starting to get endgames on the board. With so many years of attacking and gambits under the belt, I'm relatively under experienced in endgames. I simply am not used to get them on the board. But now I do get them, I smell a possibility for improvement.

10% of the endgame is about specific endgame knowledge. Those 10% tends to be 98% of the content of any endgame book. The remaining 90% is simply not treated in endgame books.

This 90% is about endgame planning. Start at the highest level. There are two possible plans in the endgame:
  • Attack the weak pawns
  • Create a passer
There are a lot of assisting rules for endgame planning. I will devote several posts to that.

Robert said
My experience with various openings (admittedly, I rarely look at specific opening lines unless I'm upset at how I played a particular line) leads me to believe that the openings (for Class players up to approximately Expert level) are not critical for winning chess. My experience is that usually I win or lose because of an oversight (usually tactical) that has nothing to do with the outcome of the opening. In short, I don't win or lose the game in the opening.

Would you please elaborate on why you consider the Caro-Kann defense to be insufficient for a Class player in tournament play?


It is a matter of taste. And of confidence. I trust the London system. I trust the Dutch defense with black reasonably well. There are a few lines in the Caro Kann that I don't like to play. Not because they are bad, but because it is not according to my taste. What I'm searching for, is to play with black against 1.e4 in a way which gives me the same feeling as the London system with white.
I used to mess up the Caro Kann since I hadn't the slightest idea what I was doing. But with my new middle game approach, I fare fairly well with it, lately. But there are a lot of variations that are so commonly well known, that I still don't know after move 12 whether I face a grandmaster or a beginner. Meaning that I denied my opponent the chance of 10 moves to go astray.

11 comments:

  1. The first chess book Carlsen read was a booklet named Find the Plan by Bent Larsen

    ReplyDelete
  2. @ Aox:

    I couldn't find a source for Larsen's Find the Plan. Is the following a fair summary of Larsen's method, or is it something else?

    Link:

    KW explains Bent Larsen's 8-Point Method for Assessing Moves

    Bent Larsen's 8-Point Method for Assessing Moves

    1. What type of pawn structure is it?
    2. What is good and what is bad about my position?
    3. Which pieces do I want to exchange, and which do I want to keep?
    4. Which side of the board should I play on?
    5. What is my dream position?
    6. What does my opponent want to do?
    7. Can I take a step in the right direction?
    8. Which moves are worth taking a look at?


    KW provides the following general advice (taken from Botvinnik, IIRC):

    One principle I have always used in games, that I read somewhere [Botvinnik!], that made a lot of sense, was to:

    1. On your opponent's move, concentrate on strategy, i.e. look at the point count, strengths and weaknesses on both sides, and think of where you want your pieces ideally;

    2. On your move, concentrate on tactics. Ask what did your opponent threaten or accomplish with the last move. Then concentrate on every check and capture you can make, and follow the lines for a couple of moves. (This is how Tal played - he was a master at sacrifice.)

    Following that, concentrate on what your opponent can do in the future to improve his position, and what you can do now before he does it to prevent his moves (Petrosian was a master at defending, and virtually stifling his opponent's chances).

    ReplyDelete
  3. FWIW, here's August Livshitz's suggested approach to solving problems:

    0. Which side is to move?

    [This is NOT part of Livshitz's 'thinking process' - I added this step simply because I - too often - forget which side is supposed to be 'winning' when considering the position from both sides.]

    1. What is the material situation?

    (If I am a rook down, I will have to find something pretty drastic, such as mate or the win of the opposing queen!)

    2. Are there any (permanent) strategic factors in my favor?

    Do I control the center, do I have a superiority of forces around the opponent's king, is his king exposed, does he have any weak pawns or squares, etc.?

    3. Are there any (temporary) tactical factors that I can exploit?

    Is the opponent's back rank weak, are any of his pieces undefended, are there any geometrical relationships between the pieces, etc.?

    4. Having determined the advantageous features of my position, can I find a sequence of moves which will exploit them, using the stated theme as a guide?

    It's fairly obvious (at least, to ME) how PoPLoAFun fits into that schema, like a hand into a glove.

    ReplyDelete
  4. the booklet "Find the plan" is now part of "Bent Larsen’s good move guide by Larsen, Bent".
    I think this book ist not thaaat interesting for the method but it gives many examples of plans to see what is a (ideal) plan, what positions call for a plan...

    Position
    1k2r3/2pr2pp/ppp2p2/5R2/4P1P1/1P1P3P/P1P5/5RK1 w - - 0 0
    White to move

    Answer:
    2. (Capablanca-Janowski, New York 1913)
    White stands better! Whether the advantage is decisive with the best defence is hard to say, and the continuation of the game is of no help, as Janowski did not defend himself very well. But what is the advantage? We can mention the doubled black pawns and the fact that White has good centre pawns. Then we can indicate the plan, which is to carry out a minority attack on the king's wing, and so, the push g4-g5. This will lead either to Black having a very weak pawn on f6, or else (after fxg5)the e4-pawn will suddenly become a very strong passed pawn. Such a passed pawn in the centre is better than the black pawn majority on the king's wing.
    This is all very well, but first we shall deprive Black of his counterplay. Of what does that consist?
    Something like c6-c5, Kb7-c6, b6-b5 and at a good (for White bad) moment, c5-c4! In the diagrammed position no master would be in doubt that 28 b4! is the right move. The doubled black pawn is kept in its place or constricted by this move. There followed: 28 b4! Kb7(?) (There is nothing to do on this wing and the black king should rush to the king's wing) 29 Kf2 b5(?) (With the plan a5,
    but White gets there before him) 30 a4! Rd4 (After 30...bxa4 31 Ra1 or 31 Ra5 the black pawn
    position is lamentable) 31 Rb1 Re5 (According to Capablanca, Kc8 is better) 32 Ke3 Rd7 33 a5 Re6
    (33 ...Rxe5 34 gxf5 would increase White's majority in the centre) 34 Rbf1 Rde7 35 g5 (Had black
    played h6, and White h4, Black would have incurred an isolated pawn at once) 35...fxg5 36 Rxg5
    Rh6 37 Rg3 Rhe6 (Or else d4 and d5) 38 h4 g6 39 Rg5 h6 40 Rg4 Reg7 41 d4 Kc8 42 Rf8+ Kb7 43
    e5 g5 4 Ke4 Ree7 45 hxg5 hxg5 46 Rf5 Kc8 47 Rgxg5 Rh7 48 Rh5 Kd7 9 Rxh7 Rxh7 50 Rf8 Rh4+
    51 Kd3 Rh3+ 52 Kd2 c5 (Desperation, White threatened Ra8) 53 bxc5 Ra3 54 d5 Black resigned.
    Was the position already won after 28 b4!? I am not sure, but I would have thought it extraordinary
    had Capablanca not won it against Janowski. With a good position and the better endgame technique he ought to win early every time. The move is rather like Nimzowitsch's famous rule for handling pawn weaknesses: restraint, blockade, destruction! 28 b4 together with the rook position on the 5th
    rank restrained the doubled black pawn; after 29...b5? it was in fact blockaded, but White had no
    need for destruction because, funnily enough, Black took care of that himself.

    ReplyDelete


  5. Position
    8/pp2kppp/4p3/8/1P6/P3PP2/5P1P/2K5 b - - 0 0
    Black to move

    Answer:
    1. (Cohn-Rubinstein, St. Petersburg 1909)
    The position was reached after an error of judgement by White: 24 Ra1-c1?? Rc4xc1 25 Kd2xc1.
    Rubinstein had foreseen that this pawn endgame was won. In this position the winning procedure can be worked out exactly. It starts with 25...Kf6!!, and the plan is to attack the isolated h-pawn and thereby force the white king to g1. After that some pawns are played forward for exchange and then the board is 'cleaned' so that the black king can go 'west' from h3, and take a pawn, perhaps on e3 or e4 or perhaps right over on the queen's wing.
    So, this is really quite an obvious plan, considering that the need for White to defend his h-pawn is foreseeable. The question is whether White can manage to take the pawns on a7 and b7 and to queen on b8 just as fast as Black can get to h1. However this is a very simple calculation. We continue with the game: 25...Kf6! 26 Kd2 Kg5 27 Ke2 (A simple analysis of the white counter attack is: 27 Kd3 Kh4 28 Kd4 Kh3 29 Kc5 Kxh2 30 b5 Kg2 31 Kd6 h5 32 Kc7 h4 33 Kxb7 h3 34 Kxa7 h2 35 b6 h1=Q 36 b7 Qa1 with the exchange of the new queens and an easy win. Later one might discover that Black can win yet another tempo with 32...b6). 27...Kh4 28 Kf1 Kh3 29 Kg1 e5. (The first part of the plan has been carried out, and now the second part is naturally introduced by a 'positional' move, which 'restrains' the doubled white pawns. 30 e4 leads to a loss after 30...g5 31 Kh1 h5 32 Kg1 h4 33 Kh1 g4 34 fxg4 Kxg4 35 Kg2 h3+ 36 Kf1 Kf3) 30 Kh1 b5 31 Kg1 f5 32 Kh1 g5 33 Kg1 h5 34 Kh1 g4 35 e4. (Or 35 fxg4 fxg4 36 Kg1 e4 37 Kh1 h4 38 Kg1 g3 39 hxg3 hxg3 40 Kf1 Kh2 (40...g2+ 41 Kg1 a6! also wins, but I have shown the other solution which could have been used if White had played 30 b5) 41 fxg3 Kxg3 42 Ke2 Kg2 with zugzwang, forcing the white king to move away from the e3 pawn) 35...fxe4 36 fxe4 h4 37 Kg1 g3 38 hxg3 hxg3. White resigned because of 39 f4 exf4 40 e5 g2 41 e6 Kg3 42 e7 f3 43 e8=Q f2 mate!

    ReplyDelete
  6. there are tactical combinations.. when there are ( many ) contacts between the pieces
    there are "to do lists" = Guidelines for to find the right move
    and there are plans like the ones obove = when the opponent has no serious threats to keep you busy,when the initiative is yours

    The thinking process to get to the right move starts of course with the question : what is
    You need to be aware whats going on ( thats something we should do mainly while our opponent is thinking )
    be aware of what to do, what the weaknesses are= ( HTRYC + Pawnstructure + Tactical weaknesses..), make plan

    after opponent makes move

    update of "what is"
    if move = exchange or pawnmove update "to do list", plan ( only then there is an influence of these )

    then you need to find the move
    check tactics (calculate forcing moves)
    find candidate moves
    select

    ReplyDelete
  7. FWIW, I found a (poor) photocopy of Larsen's book online. Fortunately, the Find the Plan section with Larsen's questions is readable.

    Excerpt from the Find the Plan section

    This book of course can be read in many ways. But you must not read it quickly. Bear in mind that a master has an average of four minutes for each move, and the positions here represent critical moments where there might well be good reason to think for much longer. And, in this connection, you should ask yourself the following questions:

    Is there any material advantage?
    Who has the better pawn position?
    What sort of pawn offensive must I aim at?
    Is there an exchange of pieces that is especially preferable?
    Is there any piece working only at half speed just now?
    Are there any holes in my or the opponent’s positions?
    Who has the majority in the centre?
    Are some squares, in the centre or elsewhere, of special importance?


    Funny: I keep finding support for the idea that rushing through the "solution" of myriad "problems" is NOT the pathway to significant chess improvement. Quality seems to be much more important than quantity.

    If that is the case, how is it that GMs "know" 100,000 or 300,000 or millions of positions? Or is it the case that they actually do NOT "know" that many positions?

    Perhaps their skill lies somewhere else.

    Don't ask me: I'm not an Expert.

    ReplyDelete
  8. every of these questiosn is a good board vision exercise

    ReplyDelete
  9. Drawing the essentials of PoPLoAFun in a paint like program, seems to me the only way to train the essential skills. There simply is no time enough during a game to follow a structured thought process, and I don't think we are anywhere close to a method to automate such thought process, which is complicated by nature and contains redundant steps by default.

    Meaning, that we must deeply understand a few standard plans and know them by heart.

    Furthermore, since middlegame moves are all about domination of the lines of attack, drawing the lines of attack of a position might provide us an essential skill which goes beyond tactics.

    ReplyDelete
  10. In order to improve my writing of the score during a game, I started with writing exercises. SCID has a reasonably timed auto replay pgn viewer which shows me the moves that I must write down at a good speed.

    Apparently I don't know the names of the squares by heart, and I'm addicted to think about the move in stead of writing it down.

    The training is way out of my comfort zone, which some people find an indication that it is a good training.

    We'll see.

    ReplyDelete
  11. I used to be bad in bullet. But I could improve. I believe it helped me to become a little bit better, not only in bullet. I would need to play real OTB games, which I dont, but I think what bullet did for me was to safe a little bit time in decision making. I play with more confidence, or I get a feeling for danger and less promissing moves. Tough to say what it really did in my thinking system. Kinda better pre-filter of candidate moves, I suppose? I still have the same amount of candidate moves (usually 2 or 3), but there is this "danger" knowledge attached to them, too.
    At the end of the day, I think it reduces my decision time to make a move. Or maybe it improved my calculation time, not sure. But I know a bit faster that I want to play the move. I safe more time in my rapid games, without playing worse. Which means the extra time I safe early on can come in quite handy in complicated positions, not only in Bullet but in any time control.

    Probably it all boils down to this: eliminating your weaknesses will make you a better chess player.
    I used to be comparibly bad in bullet, now I am not that bad anymore.

    With openings, if caro Kan or not - I play statistically promissing openings, and since I did that, I really got hold of more promissing positions for me. I like positions where I have the better game, so my tip is: play an opening that is statistically promissing, and you will see - the stats will work for you, too.

    ReplyDelete