Thursday, July 28, 2016

At the zoo

Yesterday I skimmed through my chess library, looking for an interesting book. When I have used a chess book extensively, it is worn out. I found a book that was looking almost new. Chess tactics from scratch, by FM Weteschnik. It covers pretty much what I'm busy with, so I decided to give it a try again. Since he learned chess at later age than most people, he had to work hard to improve his tactics, and he has experienced his improvement more conscious.

 "During that time, I also solved a lot of combinations to sharpen my tactical skill. I had developed my own little routine. Whenever I thought I had discovered some mechanism or characteristic of a position, I started taking notes. The work on thousands of positions grew first into a collection of unsorted tactical insights, but finally resulted in a structured overview of tactics. Over time, seemingly unconnected information, turned into a coherent concept."

Since I left the salt mines at march 26th, I have done exactly the same. I analyzed 174 positions in 124 days.

At the moment I'm looking at the different types of tempos I encounter. It turns out to be a whole zoo of different types. I look at the moves from the winning side from three perspectives:

  • What does the piece leave behind
  • What happens when the piece lands on a square
  • What threat is the piece exerting when landing
What does the piece leave behind?
In short: an empty space. Clearance that is called when there are pieces around who can make use of the emptied line or square.

What happens when the attacker lands on a square?
It might block a few pieces if the square was empty (the opposite of clearance).
If the landing square is not empty, it is a capture. We measure the effect of the landing by the piece that is captured.

Offensive:
  • What is its value?
  • Is it a defender?
  • Is it protected?
  • Is it protected by an overworked defender?
  • Is it a recapture?
  • Is the capture designed to be recaptured by another piece that will be a target?
  • Is the landing square an attacking square of a duplo attack?
Defensive:
  • Does it avoid a check?
  • Does it avoid a capture?
  • Does it avoid a threat?

What threat is the landed piece exerting?
  • This is about what the next move of the landed piece might be.
  • Does it threaten to capture?
  • Does it threaten to capture a defender?
  • Does it threaten to capture a piece that is defended by an overworked defender?
  • Does it threaten to move to an attacking square of a duplo attack? 
  • Does it give a check?
  • Does it give a double check?
  • Does it threaten mate?
  • Does it attack a pinned piece?
  • Does it attack a target that is defended by a pinned piece?
  • What is the value of the threatened piece?

As you see, it is by no means simple to tell the story of the tempos that make the combination work. Yet I belief we have to learn a few standard patterns concerning the story of tempos.



82 comments:

  1. Too many questions for me!

    On a different note:

    I've been experimenting with a specific approach to "solving" based on Antonio Gude's book Fundamental Checkmates, which I just got and started studying. Instead of "browsing" my way through it (which is what I usually do with a new book), I've been very careful to cover the solution, and then to mark the "contacts" using a highlighter. (I haven't tried to do the tests yet, which don't have a solution near the diagrams.) The purpose is to force me to do this when looking at any and all positions, in the hope that (at some point) the process will be internalized. I draw the "lines of force" from the attacking pieces all the way to the edge of the board. (I also mark squares that are reachable by either Knights or Pawns.) It's somewhat surprising that I can then "visualize" the solution much easier than otherwise would be the case. It's interesting to "see" those radiating lines of force almost "predicting" where and how the combination will unfold. Several of those training positions go on for 10-12 moves, but I don't seem to have much difficulty in visualizing the move sequences once the lines of force have been highlighted.

    Just another way to look at it, I guess. . .

    ReplyDelete
  2. My latest real rating rose to 192 ecf, which is 2140 Fide elo (with the new rating conversion formular).
    In the past I stated numbers with the old ecf/Fide conversion formular. To make my old statements comparable: with the old formular my ecf means about 2186 Fide elo.
    Again, my rating is following with some lag to my performances, because in the ecf rating, the last 3 years are taken into account, and I was weaker 3 years ago.

    At lichess.org, my rating is in the "better than 99.8%" percentil (42000 players are active there).

    Now the surprise: despite this success (at lichess at master level, and real ecf rating probably I am master level next season, too) --> my tactic ability is far worse than master level. It is at CT-Blitz around 1900. Maybe a little higher, on average.
    But other strong experts and masters are usually considerably higher rating in tactics.
    So how can I stand my ground? I believe it is a statistical thing: most tactics in games arent that deep. And I try to avoid tactics in my games, too.
    In 1-2% of all games, I miss a tactic which otherwise the ordinary 2200 fide elo player would notice (but I dont).
    However, losing 1% or 2% due to missing such tactics only costs me 8-16 fide elo.

    Well, that is my wild speculation. Maybe, if my opponents would know about my weak spot, they could score better against me? Playing a gambit hunts me down? Probably.

    My strength is my high accuracy I display with my moves. My average centipawn-loss is as low as other master players. Which means my games are often "an ocean of blue moves" (if you know chessbomb.com, where bad moves are red, and moves close to engine eval are blue.)

    My bullett and blitz ratings in online chess are not so good. Aox suggested that blitz is mainly tactics.
    So is chess 99% tactics? Maybe it is in bullet and Blitz. But in longer games, I dont think so. Otherwise I could not achieve rating like I have. My CT-tactic rating is too weak, maybe it is that of an A-class player, or at best that of a weak expert.

    Here my lichess-profile:
    https://en.lichess.org/@/Munich

    ReplyDelete
  3. I have just reviewed my list of read books. It shows I have not read any single serious book related to tactics (like Weteschnik "Tactics from the scratch" mentioned above) nor I have solved challenging puzzles (tactics rated 1800-2100). Looking at my tactics at blitz I have a strange impression that I can see more when I am short on time.

    And what is really hard to understand is my tactical skills: I cannot solve some positions (mostly at CT, but not only limited to this source of tactics) rated 1450-1600!

    I can only be proud of solving very easy tactics (rated 1000-1500). Up to now I have solved 120-150K such positions (including repeating the same ones). And when someone hangs a piece or mate (in 1 or 2 moves) I can see it instantly (it is the feeling like something "pop up" from the chess board to your eyes).

    Maybe being a very good tactician requires a lot of efficient work at the period of (at least) few years? For me it is not a problem I am not fast at tactics, but that related to the lack of ability to understand the whole concept.

    ReplyDelete
    Replies
    1. as far as i could extract from your data you have an extreme low k, you are twice as good in low rated puzzles but weak in more complex puzzles.

      So you have what we want to get: tactical vision

      But what is it you dont have? We need to guess:
      Concentration?
      stamina?
      visulaisation skill?
      ( other ) calculation abilitys
      Thinkingprocess
      ...

      I still think that an ( decisiv ) improvement in tactical vision ( dominat skill in CTS or CT-Blitz ) is not possible ( for an adult who already had some years experience of serious chessplay ) but i would be suprised if visualisation, concentration or calculation would be thaaat difficult to improve too.

      you may try to play some blindfolded games for example here : https://www.chess.com/play/computer
      an see if you can beat the engine at level 1

      Delete
    2. @Aox

      But what is it you dont have? We need to guess:
      -----------------------------------------------
      Concentration? --> TRUE
      stamina? --> TRUE
      visulaisation skill? --> TRUE
      ( other ) calculation abilitys --> TRUE
      Thinkingprocess --> TRUE

      My concentration (focus) is extremally limited up to 1-2 minutes (at best). And if you ask me about stamina... the only positive elements is my ability to play a lot of blitz (or bullet) games. Recently I have played 400 games (straight) in a period of 16 hours (with a few breaks for toilet and drinks).

      My calculation ability is only correct when there is a FORCED mate with all the moves with check - whenever there is no more than 3-4 moves mating variation.

      And finally my thinking process... has not been created yet. I try to think about the position, but it feels like I have a huge gap inside my brain.


      Whenever I check out the games I have played with the engine... I cannot believe how the hell I was able to make such blunders, oversights and mistakes. The only difference is when I solve a puzzle - the chance I will solve it correct (if it is not that difficult)... increases. However while I am playing a game - I simply cannot focus and "fire up" the move within 30-120 seconds.

      Delete
  4. As you might have noticed, I'm taking a little break. Since my come back in December I worked on chess hard every single day. I have been able to eliminate a lot of ideas, and what remains points all in the same direction: We are bad in tactics because we have a poor understanding of the mechanics. These words of Weteschnik points to the solution for this:

    "During that time, I also solved a lot of combinations to sharpen my tactical skill. I had developed my own little routine. Whenever I thought I had discovered some mechanism or characteristic of a position, I started taking notes. The work on thousands of positions grew first into a collection of unsorted tactical insights, but finally resulted in a structured overview of tactics. Over time, seemingly unconnected information, turned into a coherent concept."

    Weteschnik learned tactics at a later age, and he had to consciously work on them. Which is probably the biggest difference with learning it at a younger age.

    So I have cut out the work for me clearly: Gather a few thousand important tactical positions, and study them in depth. So far I have collected the first 200, and an in depth study confirms my poor understanding of tactics, especially the mechanics of the flow of tempos. Tools: CT, Freemind and Arena with Stockfish.

    In his book, Weteschnik describes the mechanism of the "reloader". Since I learned that mechanism, I see it a few times a day. Image that. Overlooking a mechanism that shouts you in the face every day a few times for 8 months on end. Or 18 years, for that matter.

    To take this final hurdle I need momentum. Without momentum nothing can be done. So I'm "reloading" now.

    ReplyDelete
  5. empirical rabbit assumed that we dont have huge differences in various chess skills. His hypothesis is: the various chess skills influence each other.

    But I am not sure if this is really true. If you (for instance) train only tactics, you might not become better in endgames or openings or positional play. Maybe you actually do become a little bit better, but I guess you really should tend to the other areas seperately, too.

    My tactical ability improved. No doubt about it. But it didnt improve as much as my real rating improved (from about 1800 to 2200 fide elo). And looking at other players of my rating category --> those who train tactics at chesstempo and are experts or master class players seem to be far better than me tactically. And still - I score against them sufficiently to achieve what I achieved.

    Aox tried to improve tactically, and thinks he didnt improve (much). And this is what most people do: concentrate on tactics. But what if we struggle to improve tactically? Well, then it is possible to improve, too.

    I thought that studying openings wont yield much. But the way I studied them (looking at statistically promissing opening moves and try to see some opening patterns) - I believe it helped me in achieving a better accuracy in positional play.
    Same goes for studying endgames: I looked at statistics (and thus became aware of rook endings are ~50% of all endgames), plus I started to think of the characteristics of each piece (for instance: rooks typically dont belong in the center, rooks are poor defenders/blockers but are good attackers)

    When I play blitz --> I can hardly apply general considerations like "what could the position look like later?"
    But in long OTB games - I do such considerations, and like a miracle --> what ever Munich touches - it turns into gold 10 moves later. Of course I cant calculate that far ahead. But I can sense what is promissing. And I trained exactly that.

    All this seem to compensate my comparably inferior tactical ability. Or let´s put it from the view point of my opponents: Even though they are tactically stronger --> they can not beat me. They have a 200 rating points advantage tactically - and still I am even in the outcome of results. I heard often from my opponents that I got "lucky". What do they mean? They mean that nobody could have forseen the consequences - and I agree. Still - if I am in doubt of the right move, I have the general rule "if in doubt - centralize". And "if you are not sure which of 2 different moves is better - take the one which is safer". These rules (and others) served me well - and they are the reason why I am always "lucky".

    It is a bit like having pop corn on a hot plate --> each corn jumps when it pops and you can not tell in which direction it jumps. But if you tilt the plate with a light slope towards the left side, then more corn will pop (and jump) towards the left direction.

    Long talk - little sense: chess seems NOT to be 99% tactics, but instead it is possible to compensate a tactical difference of 200 CT Blitz rating points. Tactical monsters can not beat me if I dont go into tactical positions.

    ReplyDelete
  6. @Aox: I seem to have a low k-factor indeed. But I am not sure if I had it to start with. I am afraid that is the case, so I never trained it, not improved on my k-factor.

    My standard rating is rather high, but I find it "simple" to get a high standard rating. The solution is: if you are not sure - dont move. Be patient. Only move if you think you found the right move. Or: in case you lose patience, you should have at least half an hour of patience.
    If you dont have even half an hour of patience - well, then I would suggest you stand up and come back later, with a fresh view. Just dont move. I had puzzles running for hours. Inbetween I stood up. What helped me as a motivation was this quarrel about "does standard rating mean anything?" with Richard and Co. Plus I had a high RD, so doing a wrong move would have meant a severe punishment on my rating. These two factore were enough motivation for me to keep going and searching.

    But dont get fooled: my standard rating of 2480 is really only achieved with extreme patience and a high RD.
    I do have a good K-factor, but I could imagine everybody who is an A-class player can achieve a 2500 standard rating - if he keeps being careful and is motivated like I was to prove exactly that.

    The best solver, cmuroya17, is a 2000-2100 fide elo rated player. This is my assumption of Richards indirect comment about his strength (Richard said somewhere that he is an expert player, and when somebody asked if he was a strong expert player (close to master level = 2200 fide elo), Richard denied that.)
    So Cmuroya17 is likely a little bit of a weaker player than me.

    During my real OTB games I have the rule that I try not to spend more than 10 minutes on a move. As soon as I am aware that I used more than 10 minutes, I try to move within 1 minute: I some up my findings, and try to play the move that look safest. If a safe move isnt available - then I move the move which looks "right" (for instance a move that centralizes a piece).
    One exception: I am threatend with a check mate (my opponent sac'ed his queen for an attack) and am thus a lot material up. Then it is just about survival over the next few moves, and thus I can spend more than 10 minutes on a move.

    So no matter how my K-factor is: I make a cut at about 10 minutes. Of course most of the time I move within 3 minutes, otherwise I would get into time trouble.
    I am very disciplined about my time management, which is an important factor for a good rating, too. (Time management is severely neglected: even some super-GMs are really bad in time management, like Grischuk or Ivanchuk.)
    I believe my time management gives me +100 Fide elo. Hard to prove, though.





    ReplyDelete
    Replies
    1. @Munich
      you have a high k, when you think twice as long your play gets much better, you can play blindfolded, you can solve standardpuzzles of a rating of 2200 and more.
      Tomasz dont benefit from thinking more than a very few minutes while we both can solve virtually every possible puzzle by calculating it for weeks.. if necessary

      cmuroya17 had a standardrating of ~~1850 in the year 2009 with more or less avarage thinking times. These days he spend sometimes several hours for a new/unknown puzzle, still with a phantasic high score. He simply memorised al high rated puzzles at ct, he did see them at least 20 times.
      I guess he is now an extreme good calculator but his tactical vision might be still "low" at a sub 2000 level.

      Delete
    2. I cannot make ANY benefits from looking longer at the position (puzzle) due to the lack of thinking process. It is the same if some areas of the board could have been simply INVISIBLE to me. I can stare at the board for hours, but I can't find the solution - no matter how hard I try. Why? I have not created the system yet. My thinking is extremally limited and I can see only what I saw before (yeah, pattern recognition is the key to solving puzzles by me).

      Delete
    3. I quess that such problems could be easily fixed and that you would be then at least a 2100 tactically
      I suggest to do:

      Visualisation exercises like "chess eye"
      blindfolded games
      Complex tactics
      Mate in 2 and 3 >>problems<<
      Endgamestudies
      and some training in thinking process

      Delete
    4. It is very difficult to pinpoint the specific problem(s) of a particular player and to provide helpful suggestions for improvement. Nevertheless, certainly shared solutions might be useful!

      One of the helpful things I learned from Weteschnik's book Understanding Chess Tactics (I think this is an earlier version of Chess tactics from scratch) is the idea of doing a "status examination" of the pieces - all of the pieces on the board. (I also encountered the "reloader" for the first time as a named concept in this book.) (GM Rowson refers to this as a position "survey.") If you look at each individual piece and visualize the "lines of force" radiating out from each piece (especially with the long-distance pieces) from that piece's current square all the way to the edge of the board, you will begin to "see" interrelationships occurring on squares that may be occupied or empty. It is important to "see" through both same-side and opposite-side pieces, as if they are not actually obstructions. As a means to this end (of "seeing"), I suggest taking a tactics book and a highlighter, and simply drawing the "lines of force" on each diagram as you analyze. You can either draw only the lines for the side to move, or you can draw both sides if needed to make the interactions clearer. I draw straight lines for Queen, Rook and Bishop, and curved lines for the Knight. As you draw the lines, note the squares on which there are interactions between the various pieces: attacking a common square, supporting an attack, and defending against an attack or counterattack. Only when you get several lines drawn should you gradually shift/switch your attention to potential tactical themes/devices utilizing those pieces/squares. Eventually, I think you will get to a point where you have internalized this "system" into your visualization; I know it has helped me considerably. At that point, you should stop drawing the highlighted lines, and just try to "see" without the "crutch" of highlighting. "Follow the yellow [highlighted] brick road!" to the Emerald City of better visualization!

      Another suggestion is to spend some time "seeing" the motifs that hint (or loudly announce) the possibility of tactical themes/devices and the subsequent combination of them. Neiman's book Tune your Chess Tactics Antenna: Know when (and where!) to look for winning combinations. He gives a generalized "thinking process" composed of 5 steps:

      (1) Global vision (what we have referred to as the "vulture's eye view")
      (2) Analysis of the position (generally detecting tactical possibilities and looking for "signals")
      (3) Looking for the theme(s)
      (4) Looking for candidate moves
      (5) The calculation of variations

      (BTW, Neiman introduces a tactical theme/device called the "swing door" which I have not seen elsewhere. It is a very specific tactical device, and not as generally useful as the "reloader" idea identified by Weteschnik.)

      Neiman's book is broken down into four parts.

      Part I: The Seven Signals
      1. King position
      2. Unprotected piece(s)
      3. Alignment (the geometrical motif I learned from Dr. Lasker)
      4. Knight fork distance
      5. Trapped piece(s)
      6. Critical defender/Overloaded defender (the function motif I learned from Dr. Lasker)
      7. Impotent defense/Defense too far away

      Part II: Find the Relevant Theme

      Part III: Looking for the Right Move

      Part IV: Final Test

      I have no idea if (or how much, if any) these suggestions might help you, but they did help me considerably.

      In any event, GOOD LUCK!

      Delete
  7. I will try to check some of your suggestions/recommendations. What do I mean?

    1) Complex tactics (30%)
    2) Mate in 2 and 3 >>problems<< (10%)
    3) Endgame studies (15%)
    4) some training in thinking process (45%)

    I am going to realize: Complex tactics - with the help of chess puzzles (I have about 2-3K of such to solve), mate in 2 or 3 moves as problems (as well as endgame studies) from the manuals of chess (Slavin's series) and the training in thinking process from a few books (Weteschnik, Heisman, Jusupov and my own observation and conclusions). We will see if any change takes place.

    BTW. I do not threat chess seriously anymore and that's the reason I cannot be seen as a real example (no matter if we speak about success or failure). Anyway I want to make a shot (give it a try).

    ReplyDelete
  8. Tomasz, your boardvision is much better than mine. So how is your tactics (CT Blitz rating)?
    And how is your online bullet rating compared to online blitz rating compared to longer time controls? I suggest you play on lichess.org --> it is for free and gives you easy after-analysis. Also, there arent as many cheaters like on chess.com, so you can really have longer time controls and dont get cheated often.
    lichess has some statistical analysis available, telling you for instance how many inaccuracies, mistakes and blunders you did during your game(s).
    lichess.org is fast, no lag (like it is often the case with chess.com, for instance).

    So try there, and see if your different ratings (bullet, blitz, classic) is more or less all the same.
    I am not sure if more tactics will help you a lot. Blindfold chess maybe, in case you can not properly keep a future position in your head (I would wonder, though, because your boardvision is so good - this would not fit to not being able to visualize future postions).

    I really liked smirnovs videos. The "breaking stereotypes" series is pretty good. Also watch "the most common mistake" (= "to take is a mistake") - I had really to fight the urge to capture something. It was as if it was itching me and I was not allowed to scratch. But after a while I got used to not take. Meanwhile I am so extreme in not capturing, some of my opponents suggest I am a passiv player (Steinitz type), because they consider capturing as aggressive, and not capturing as passive. Probably my opening repertoire adds to my reputation as an extreme provoking passive player: 1.e4 1...Nc6 as black, and 1...Nf3 as white (leaving often e/d pawn behind). But I am not passive. I am simply sticking to the rule.
    My "if in doubt play safe" might contribute to my passive reputation, but again, I see nothing passive in this, but just common sense. (my last 40 games --> I lost just one game to a 2250 player, and all other 39 games I won or drew).

    Anyway, clearly the way how I play changed and made me the player I am today. My tactics improvement (150 CT Blitz rating points) can only explain part of my 400 fide elo rise. The way how you go on about the game is important, too. And maybe - just maybe - it actually could influence your tactic skill at the end, too. (though this was not the case for me).

    If you find me on lichess.org - we can play some games together. I would love to see how you play and if I could find improvements in the way you play and think. There isnt one truth, though. Chess can be understood in many ways, not just my way. But if you see how I think about moves - maybe you can draw conclusions for your style, too?
    The more I know about chess, the more I have the feeling I know nothing.

    ReplyDelete
  9. Lichess stats:

    1) Bullet - 1744 (870)
    2) Blitz - 1942 (211)
    3) Classical - 1719 (2)


    There are different scenarios. Sometimes my bullet is about 80-120 points better than blitz, but sometimes it is close to the opposite.

    Let me know what rating do you want me to achieve at Bullet and Blitz. I will think about it and tell you what is possible and what is beyound my present skills.

    I have NO INTEREST at practicing (playing) blinfold. It is simply not for me.

    Unless I reach 2000 rating (at blitz and/or bullet) there is no sense of playing against you. I simply do not want to waste your time for beating me without any effort.

    "The more I know about chess, the more I have the feeling I know nothing" - it is the same to me as well. I can only say I know the rules of chess, but everything beside that is out of my imagination or skills.

    BTW. Nowadays I am working to fill the biggest gaps at my chess knowledge and create (develop) some skills. Otherwise I cannot make ANY progress.

    ANY comments or questions?

    ReplyDelete
  10. PART I:

    @ Munich:

    "There isn't one truth, though. Chess can be understood in many ways, not just my way.

    That is very true! Each of us acquires a set of skills that are unique in the combination of strengths and weaknesses.

    "The more I know about chess, the more I have the feeling I know nothing."

    I can certainly vouch for THAT sentiment!!

    @ Tomasz:

    Have you tried the highlighting method of tactical study that I suggested? I'm curious as to your thoughts about it, if you have tried it.

    Merely as a suggestion, I wouldn't focus on acquiring knowledge; I would focus on gaining skill. Knowledge does NOT equal skill! There is a certain minimum amount of knowledge required (beyond the rules of the game, such as elementary endgames, etc.) but IMHO it's more important to be able to use the knowledge you possess than to merely possess (and continually add to) that knowledge. Chess knowledge (too often!) is descriptive rather than prescriptive.

    As the old saying goes:

    Those who can, DO; those who can't do, TEACH; and those who can't do or teach, TEACH HOW TO TEACH. ;-)

    I just started on lichess. So far, it's my favorite training site! I haven't played any games yet, but my training rating is 1928 based on 107 problems (as of right now). I've momentarily gone above 2000, but can't seem to stay there very long. I'm using the "Normal" problem set (no "Easy" or "Hard" problems yet). It has always seemed weird to me that I have more difficulties with the lower rated problems (under 1600) than with the higher rated problems (above 1800); I have no idea why that should be, nor what it says about my skill (or lack thereof).

    ReplyDelete
    Replies
    1. @Robert

      I have not tried the highlighting method of tactical study. At present I test another idea. After I finish I will check it out and see how powerful it is.

      Of course I know knowledge does NOT equal skill! However my approach is simple: you have to know the basics if you dream about playing decent chess. I have been playing without possesing decent level of knowledge, but it did not give me any satisfaction (not to mention a high level of play).

      The knowledge I have already posses is used at about 90%. It means I need more knowledge - especially with the area I lack it completely.

      And to your playing skills/level at lichess.org. If you cannot solve the positions at low level - it may be connected with your attitude: if you are too focused at difficult ideas and moves, you can often miss VERY simple solution. For example: if you can win a Q for a R in 2-3 moves you may be trying to see "deeper concepts". I have the oppisite approach: I take for granted the position (solution) is simple and that's why I cannot cope with the more difficult ones.

      BTW. "Chess knowledge (too often!) is descriptive rather than prescriptive." - I do not get it. Could you explain what you mean?

      Delete
    2. @ Tomasz:

      Consider the description of various Pawns: doubled, tripled, isolated, backward. Consider the description of Bishops: good, bad, and "bad, but outside of the Pawn chain." Although these terms DESCRIBE an essentially static situation regarding those Pawns and pieces, none of them PRESCRIBE what to do with them in the given situation. What might work in one situation is essentially the WRONG thing to do in another concrete situation. Consequently, the description does not contain a prescription for what specifically to do. The Latin phrase "ceteris parabus" (everything else being equal) comes to mind. In almost all chess positions, everything else is rarely equal. It is what GM John Watson described about "general" rules or principles: in most cases, playing by such rules will get you smashed in a game of chess.

      Delete
  11. PART II:

    I learned how to play blindfold a long time ago, while I was still in "active addiction" (i.e., playing tournament chess) back in the 1970s. I saw an IM George Koltanowski blindfold demonstration of the Knight's Tour (with random pieces of information filled in from the audience on each square). Mr. Koltanowski had someone pick a starting point at random and he rattled off the Knight's moves AND the information written on each square faster than the person marking the board could keep up with him; he had to pause several times to allow the person to mark the squares he had called out. I wanted to figure out how to do blindfold chess myself. Mr. Koltanowski stated that he remembered the piece locations aurally (by the sound of the moves being called out), not visually (by sight of an internal board). I thought that quite strange, and learned how to play blindfold visually (if that is not an oxymoron). My "secret": I memorized a group of four squares (a1, b1, a2, b2), then combined those into a group of four (a1-a4-d4-d1, a5-a8-d8-d5, e1-e4-h4-h1, e5-e8-h8-h5). I never attempted to "see" the entire board at once. I would mentally place the pieces into one of the four quadrants, and keep track of them within each quadrant. Visualizing the movement across quadrants was the most difficult task, but I got used to transferring the pieces from one quadrant to another and keeping them visually in mind when needed. I didn't have to remember all of the pieces all of the time; I just remembered which pieces had been moved, and to which squares they had been moved. I was familiar with the "chunking" concept (7 +/- 2 or 3 groups of 3) and tried to keep the "chunks" to a manageable number.

    I used to give simultaneous exhibitions at different chess clubs. I once played a ten-game simul with one of the games played concurrently as a blindfold game. I won all of them, and some of the players were within 100-200 points of my rating at that time. I suspect I won because it was somewhat intimidating to be playing against someone doing that kind of exhibition, and not as a result of any real skill on my part!

    My hardest blindfold game was a consultation game involving four players. My partner and I were both playing blindfold against two other players who were looking at the board. My partner was rated about 1500; I was rated above 1700. The reason it was so hard was that my partner and I took alternate turns making moves, without any consultation or communication between us. I had to be careful in what I set up tactically, because I had to make sure my partner could "see" what I intended. We won the game, but it was grueling!

    I have no idea if blindfold chess helps OTB play. I didn't notice any significant improvement for me after I got used to playing blindfold. I can still play blindfold, but rarely do it unless asked by a friend for a demonstration of it. To me, it's more of a "parlor trick" than a serious training method, but what do I know?!?

    ReplyDelete
  12. That is pretty interesting about different ways to play blindfold. For me it works a bit like corresponding squares.
    For instance: if there is a bishop on c4, I know it can hit on f7. All I need to do is then to check if something is blocking the way inbetween. So I check if there is a piece on d5 or on e6. I know that d5 and e6 are within reach of a Bc4, too. Not because I see a diagonal, but I simply know a bishop on c4 can go to d5 of e6 or g8 (and many other squares, of course).
    It is a bit like memorizing that 8x8 = 64. I dont calculate that, I simply know that the pair 8 with 8 produces 64.
    So I am not really seeing an internal board, too, but I just know where a bishop can go if it stands on c4. I could not even tell you if it is a white or black square!

    But I guess I do see part of an internal invisible board, too. When I check a king position and want to attack it, I have a pretty good image of a king on g8, a rook on f8, pawns on f7, g6, h7, bishop on g7.
    Kind of "Chunk".

    All chunks together give me a picture. But I am not using such a picture to see what the pieces can do, but it is just to memorize where the pieces are currently standing. The calculation happens with corresponding squares only (ie a bishop on c4 is able to hit f7).

    @Thomas:
    I am not interested in beating you, but in playing a casual game with sufficient time, so we can express our thoughts why we play which piece during the game. I tell you even my plan what I intend to achieve, and vice versa. The result is absolutely un-important (and it is not rated, too).

    But we two seem to be 2 extremes of the opposite side. I assume I am a better deep calculator than you, but my boardvision and short tactic manoevers - here I am considerably worse than you.

    I would expect that your classic rating does not differ much from your blitz rating.
    At lichess, my ratings are currently:
    Bullet: 2018 (better than 91.1% of all players)
    Blitz: 2022 (better than 91%, but usually I am better than 93 or 94 % of all players)
    Classic: 2334 (better than 99.6% of all players, and usually I am better here, too, like 99.8% or 99.9%)

    You see: if I have more time at disposal, I see to make way more out of the extra time than other players.
    My explanation for this is: Bullet and Blitz is mainly tactics, whereas in standard additional knowledge is an increasingly important factor. I tend to win games not with major tactical strokes, but by accumulating little advantages and winning the endgame.

    ReplyDelete
  13. So the two of us (Munich vs Tomasz) played a couple of bullet games - and the outcome was like expected: I lost 15-6.
    Tomasz pointed out 2 possible reasons and/or improvements for me.
    a) I need to look out for counter-attacks I initiate myself (I hardly do)
    b) and I need to risk more in case I am low on time.

    While b) rather adresses "just" an improvement in my Bullet play (hey that is still great!), the other point is adressing my chess ability overall.

    In either case - rating improvement could possibly be done (in my case) without becomming better in tactics. Well, maybe. I need to try to improve on my weakness (counter attack and dynamic play forward), and then I see if it helps me. I could imagine it could do the trick, and maybe the final 100 elo and thus the FM title could be reached? Can dreams become true?

    If so, all of you could see that chess is not 99% tactics. However, it is probably true for a lot of master players, because if I really manage to cover the last 100 elo and become a FM, then I would be an exeptional FM player. A master who has not the usual tactic skills of a master, but is just about good A-class or weak expert level in tactics?

    So what do you think? Is it possible without getting better in tactics? This is important to know for many people, because it seems that improving in tactics is really difficult, if not impossible at some point.


    ReplyDelete
    Replies
    1. @ Munich

      "Is it possible without getting better in tactics?"

      I'll answer that in more depth when I have more time; my short answer is: I think it is POSSIBLE (as in "not IMPOSSIBLE"), but not very likely statistically. If you can always choose the best move in every position without tactics, then you can beat anyone. Aye, picking the "best" move is the problem!

      Did you and Tomasz play your bullet games on lichess? If so, I could not find the games. I would be very interested to see your games, if that is acceptable to you both. I think the games could be instructive to more than just the two of you.

      Delete
    2. It is easier to find such called "the best move" (if can agree such a move exists!) when you know what is your plan and what are the elements of the position.

      It is IMPOSSIBLE to find the best move in every position without tactics. At least with the exception of "queit positions" it may not be done without the support of tactics (and combinations as they come from tactics).

      Our bullet and blitz games were just for fun and to test some hypothesis. I am not sure if they are (can be) instructive to anybody. They can only tell if I had faster mouse and body reactions than my opponent (especially when I used my time tricks to flag my opponent most of the time).

      BTW. It is up to Munich if he agrees to publish (give access) to the games we have already played. I have nothing against, but imho - they are worthless as they were play at extreme conditions. It is the same if a student was forced to write his final maths exam in only 10 minutes (normally it takes 180 miuntes). What conclusions could we make from such an extreme approach?!

      Delete
    3. @ Munich

      PART I:

      And now a longer reply. . .

      Richard Reti, the great Czech grandmaster of the first half of the 20th century might be thought to have proposed an answer in his response to a question as to how many moves he prepared ahead in a chess game. "Only one," he replied, "but it is always the best move."

      Reti was joking, of course. [WAS HE?!?] Grandmasters typically [WAS RETI A TYPICAL GRANDMASTER?!?] look about five moves ahead in a given position, and sometimes many more. But Reti's quip captures something at the heart of human chess-playing ability: intuition. And in so far as humans rely on intuition rather than calculation for the decisions of everyday life, his statement captures what is at the heart of human intelligence more generally.


      I ask the questions I inserted because there is an ASSUMPTION that Reti was being somewhat of a "smart ass." I don't think he was. Allow me a short digression into martial arts training.

      As an instructor, I was asked many times what it would take to GET a black belt. My stock reply was, "Do you have $10.00? If so, I'll sell you the one I'm wearing or you can go buy one at Junk Fu Martial Arts Supply for $10.00." The prospective student (usually, but not always) respond with a second question: "What would it take to BECOME a black belt?" My truthful answer was "A lifetime of hard training." There is a considerable difference between HAVING a black belt and BEING a black belt.

      Delete
    4. PART II:

      I personally think that the "Chess is 99% tactics" percentage is mistaken. In actuality, it is 100% tactics. Or, to quote Tigran Petrosian, former World Champion: "In general I consider that in chess everything rests on tactics. If one thinks of strategy as a block of marble, then tactics are the chisel with which a master operates, in creating works of chess art."

      Think about that for a few moments before rejecting it as nonsense. Assume that it is true. Now consider how you are able to reach very high skill (almost to the FM level of play), yet (supposedly) without very high skill in tactics. If chess is 100% tactics, then that seems impossible. The general conclusion must be that either (a) you have much higher tactical skill than you think you have, or (b) there must be more than one way to "see" (be aware of) tactics. Maybe Reti was pointing out that he "saw" in a different way from the usual "I go here, he goes there, then I go there, and he goes here, etc." In short, "thinking" like a computer is NOT how a human thinks. We don't build trees containing millions of moves, and then use the alpha-beta algorithm to "prune" (based on a relatively primitive evaluation function). We use "rules of thumb" (general principles) and intuition to guide the move search.

      Consider an extremely complicated tactical position. If calculation of tactics was the "sine qua non" of chess expertise, then in the most complicated positions, calculation would be basis of every master's approach to "solving" such positions. But that is exactly what does NOT happen! The master relies upon his "intuition" to guide his move search in the positions most likely to "require" tactics.

      I infer from this that it is POSSIBLE to reach a pretty high level of overall skill without being at the highest level of tactical skill. Other factors (positional skill, knowledge of general principles, knowledge of typical endgames, etc.) MIGHT compensate for a lack of high tactical skill.

      No less an authority on tactics than Dr. Emanuel Lasker opined that World Champion Wilhelm Steinitz was (relatively) poor at tactics and endgames (something that Lasker excelled in), but was a superb and deep thinker at strategy. After seeing Steinitz's combination against von Bardeleben, I seriously question how "poor" Steinitz's combinational vision actually was!

      Link: Wilhelm Steinitz vs Curt von Bardeleben

      I think it would be EASIER to reach FM level with at least an expert's grasp of tactics.

      More importantly, do whatever you find enjoyable about chess! If studying tactics until your eyes bleed does not appeal to you, then study other things, or, just play games. In the long run, it's all about a GAME to be enjoyed.

      When it stops being fun, I'm looking for something else to waste my leisure time.

      Delete
  14. Yeah, I can confirm Munich's conclusions.

    The bullet outcome proves only one (at least to me): I am better at playing on time and "flag out my opponents". I can even go as far as having 0.01s and my opponent had about 12 seconds and I managed to win on time. Yes, it is a "time trick". The same as I used against Munich. If we had played the games at 3+0 the score would be at least 14:6 to Munich's favour (considering applying the hint of playing the last 10-15 seconds with looking at his opponent's time).

    The need to learn dynamic play and counter-attack is a MUST have to any master. I am quite surpised Munich was able to reach 2200 ELO rating without implementing this element is his games.

    And in your case my dear friend - you can achieve a (fide) master level without working very hard on tactics. Anyway if you improve your counterplay (counterattack) you will be FORCED to be better at tactics - not much stronger in tactical puzzles, but in assesing dynamic possibilities of your and your opponents potential.

    I am not sure if JUST improving dynamic play and counterattack give you a master level. It may be needed to improve the endings as well (if you do not have endings at master level).

    I do not think you will be an exceptional FM player. It is a matter of different traits of a master. If your positional play is extremally good one and you can compete against masters at equal ground - there is no problem to reach master level by you. I could assess you as an expert at tactics, but master at positional and endgame play. It is possible, but reaching IM level may not be possible (unless you improve your tactics to the next level).

    Summing up: it is possible to reach a FM chess level without being master at tactics. However you have to play very well positionally and probably IMs and GMs would be able to beat you at forcing dynamic play. Anyway if you manage to be better at dynamic play you will be better at tactics - not just in solving random positions, but to asses dynamic possibility for both side - much better and deeper.

    I hope my observations can help you to improve your level. I cannot imagine any player rated 2200+ without the skills of dynamic play and counterattacks. There are some players who prefer to play positionally, but when you force the dynamic play - they have to know how to react for such an unbalance.

    Any questions or comments? :)

    ReplyDelete
  15. About the phrase: Chess = 99% tactics
    The phrase that chess would be 90% or some even say 99% Tactics was told centuries ago ( by Tartakover or Teichman? ) in Europe where the term tactic was used different to the common use of "tactic" in chess in the US of A these days.
    Some did use "Tactic" like we use "Strategy" today. Lasker was using the term "tactic" similar to "a more or less short sequence of actions" with gaining some benefit but maybe just an small positional one. What is commonly called tactics these days would be a combination with Lasker ( a sequence of forced! moves with a clear gain of material or checkmate..) but a combination is today a sequence of forced moves with a sacrifice

    So using the old phrase of chess beeing 90% tactics makes no real sense these days

    If we want to know how important tactics ( = CT- like thinking ) is then we need to know how many forced moves sequences gaining wood or checkmate are necessary to calculate in a game.. and in a closed game that might be not that much

    GM Smirnov say in his "Calculate till mate" that tactics would be completly unimportant because a game would not be won by a combination. He asks : how many games did you win because of a "fork"? But he might think of a higher rated player who already spot simple tactic like i spot a red spot at a white wall

    ReplyDelete
  16. About "tactical skill" and "tactical vision"

    "Tactical vision" is the skill to spot easy tactics without much thinking. The "tactical skill" is the ability to solve tactical puzzles ( in a given time ).

    Munich*s tactical vision is "low" but his tactical skill is not low with complex tactical puzzles As longer munich is thinking= as more time munich spend on a tactical problem as better he gets. Empirical rabbit did call the growth factor: "k"
    k is the gain of elopoints per doubling the thinking time. While munichs k is exceptional high ( say 250 ? ) tomascs k is exceptional low ( say 50? ). ( I did not "calculate" these k`s but i did analyse their ct-performance before intense )
    Tomasc tactical vision is ~~ 2100 and the tactical vision of munich is ~~1900 ( this i did calculate ) So the tactical vision of Tomasc and the k of munich would make an IM (in tactics) ;)

    Somehow munich is gaining more and more information from a position and is able to use this information to generate better and better moves.
    While there is a "cut of" in the results of then thinking of tomasz

    So the question remains: what skill is k ( mainly )?
    Munich claims he would have a superior positional vision , meaning he spots where the pieces should be positionally. But such a skill dont grow with thinking time ( without calculation ). So i think k needs to be ~~calculation. By saying "calculation" i think of the thoughts like : "if i do this he does this then i do that ..." and not necessarily "CCT" or "Candidate moves". I cant think of no other method to gain more and more information as more time you spend.

    ReplyDelete
  17. Tactics are a forced way to gain wood. Most tactics in a game are not played but avoided. They limit the possibilities of both players though. So you must know enough of tactics to be able to avoid to fall prone to them.

    I quitted chess, and I will only come back when I gained at least 300 points at CT. There is a simple reason for this: without a thorough knowledge of tactics, my games are literally comparable to a game of chance. I have no idea what I'm doing, and I'm constantly surprised by the positions on the board. I don't see the position as a consequence of my own bad play. It is as if I see the positions "as new", continuously. Exactly what the brain scans of amateurs show. If it was not for my opponents to play horrible as well, I could not win a single game. I think it is ridiculous to play a game in this way.

    The upside of my approach, is that IF I manage to gain at least 300 points at CT, and restart playing chess, we will soon know how it contributes to my OTB game rating. It will settle finally a few questions.

    The difference of descriptive and prescriptive knowledge, is a matter of scale. Like a doctor, you must first diagnose the disease based on the symptoms (descriptive). If every illness has several possible cures, you need to know a manifold of facts in comparison to the amount of symptoms and diseases. What cure must you prescribe?.

    For tactics, a few thousand positions where you know the symptoms, the diseases and the cures, are enough according to FM Weteschnik. I'm going to find out whether he is right. IF it works, I would be very surprised if the same method wouldn't work for positional play as well.

    ReplyDelete
    Replies
    1. Its is true that tactics are involved in a game even if they where not actually played. But then they where part of the calculation and had to be considered. Some masters said lately that strategic rules or principles like: "if you are attacked at a wing you should start a counterattack in the center" are nonsense because in virtually every gm game that would not happen. But these masters know that rule an take care that if they start a wing attack there is no danger of a central counterattack. So of course Smirnovs argument is not very strict ;)

      Its a russian proverb that it would be enough to understand "300" positions realy deep to become a strong chessplayer. Books like "GM-Ram" and "Chess Training Pocket Book: 300 Most Important Positions and Ideas" are based on this proverb. But i think thats wrong.

      About the difference of descriptive and prescriptive knowledge: Its not enought to recognise the weakness , you need to know the method to make use of the weakness to. Say there is a weak backrank = weakness, then you have to bring some heavy pieces at the backrank or threaten this = method. Without rooks and queens a weak backrank is no real weakness. Its important to have all the knowledge and not just parts of it ;)

      And Weteschnik... he was not able to improve his play in germany, he always stayed below 2100 DWZ ( =German Elo ) even after! getting his title ( in a poor country with a lot of titled players without income within a few months ) his performance in germany was bad. Fide did change the rules for getting a title later and the request is now to have at least a DWZ of 2200 to get a FM. To many strange sudden increase of elo in eastern countrys happend in the 90`s : http://www.schachfeld.de/threads/9945-fm-martin-weteschnik-stimmt-da-die-elo-zahl

      Delete
    2. PART I:

      Both Temposchlucker and AoxomoxoA make a point regarding the quantity to be learned: whether it is 300 or 3,000 (or some other magic number) positions, it is not "how many" and "how fast" but the skill that is acquired from the studied positions that is important. I hypothesize that it is not actually the specific content of the individual positions that is important. In other words, the set of studied positions does NOT have to include every possible tactical nuance that could ever occur in order for skill to be developed. If one problem can convey the essence of an idea, then piling more similar positions on top of that idea will not add any new ideas. A significant chunk of skill is evolved through exercise of the non-chess aspects: the drive to understand the position, the tenacity to keep studying until the position yields its "secrets," etc. Those are personal traits, not chess skill per se.

      I have stated before that I think the notion that "more is better" widely misses the mark as far as chess improvement is concerned. My personal ambition (given my work history as a "lazy" programmer) is to maximally reduce the amount of work required to learn to the minimum extent possible. After "solving" a specific type of problem, the "secret" of that type of problem should be absorbed as skill. If it is not, then the student is not approaching it in the necessary frame of mind required to embed it permanently in LTM. For absorption of long-term benefit into LTM, speeding through a large number (BILLIONS AND BILLIONS, to quote Carl Sagan) of typical exercises (especially in the "salt mines" doing thousands of problems as fast as possible) is not likely to leave anything of substance retained. "A rolling stone gathers not moss." One of the things I admire about the chess improvement "quest" is the willingness to consider new ideas of training, followed by rigorous experimentation to see what benefits (if any) can be derived from them. I refer to it as the "left corner nibble" method: bite off the leftmost corner tidbit, chew it up thoroughly and then digest it. Lather, rinse, repeat until your overall knowledge base has expanded to incorporate most of the required skills. There will always remain regions of "Here there be TIGERS!" but (eventually), you will have sufficient overall skill to accomplish the job. [I can hear the anguished screams now: "But it will take too long!" Nothing worth learning is ever accomplished instantaneously.]

      Extraction and absorption of general principles and patterns through personal hard work (dare I use the term "deliberate practice"?) is what is needed for SKILL development. I think this is the value of the so-called Kotov or Stoyko exercises. If you cannot remain mentally focused and disciplined during training sessions (regardless of the specific aspect of chess under study), why would you expect to be focused and disciplined during a game? YOU WILL PLAY LIKE YOU TRAIN!

      Delete
    3. PART II:

      I know from my own experience that superficial study habits lead to superficial "knowledge" and not long-term skill. Only by delving deep into the recognition "signals" and the generalization of typical methods am I able to improve. I have had the desire to rapidly improve and have tried many superficial methods, only to experience - no significant improvement. It has only been when I have given up the "easy way" and accepted the "hard way' that I see any real improvement. That applies to everything I've ever tried to master, not just to chess. How many martial artists do you know who start training at age 40 and go on to reach 4th-degree black belt (earned with sweat, not bought at the local "Take My Dough" through a buy-a-belt program)? (Maybe there's an encouraging example for ADULT "improvers"!)

      Knowing what a fork or pin is DEFINED to be (the description) is NECESSARY for recognition, but is INSUFFICIENT for deciding what can be done with it. In a given position, should you set up and then execute a fork or a pin? IT ALL DEPENDS ON THE SPECIFIC POSITION! This becomes a matter of judgment, which can only be honed and refined by making ever more nuanced critical judgments in many similar situations.

      Whether Master Weteschnik "gamed" the FIDE system or not to gain the coveted "Master" title, there is still value in his idea of collecting instructional material that yields specific ideas, and the process of refining those ideas into coherent generalized methods of recognition and play. I quote him:

      "During this time [training under IM Tibor Karolyi] I also solved a lot of combinations to sharpen my tactical skills. I had developed my own little routine. Whenever I thought I had discovered some mechanism or characteristic of a position, I started taking notes. The work on thousands of positions grew first into a collection of unsorted tactical insights, but finally resulted in a structural overview of tactics. Over time seemingly unconnected information turned into a coherent concept."

      Do you take notes on what you learn from a given position? Do you regularly review your notes, looking for common unifying concepts, resulting in generalized principles and "rules"? If you are not doing things like this, then I suggest that you might be wasting a significant amount of your training time and effort.

      You cannot acquire lasting skill based on someone else's hard work.

      Delete
  18. About improvement in tactics

    we did see in the saltmines: while extreme simple tactical vision ( = simple boardvision ) is seemingly improvable more complex board vision is hard to improve and tactical vision ( starting with problems of the complexity of m1) is not improvable.
    That is of course said for adult players with several years of serious chess experience
    In my eyes there are only 2 possibilitys for improvement in tactics:

    i) either we make complex board vison improvable making them easier by improving at the relevant subtasks.
    This is ehhhh very very very dry... but very interesting too;)

    ii) we improve in k.. somehow.. But i have to admit that i dont know precisly what k is and how to improve it. I guess its thinkingprocess + calculation + logical reasoning + ..

    Tempos analysis of his "blunders" is a method to improve in thinkingprocess ( and logical reasoning? )

    ReplyDelete
  19. OFF The TOPIC:

    I am wondering what is the influence of identifying the coordinates (square) to correct vision or calculation. Or as least what is the quickness of finding (searching) the squares. I think It may have some connection (relationship) to the saltime "#1" we had been solving for some time.

    I would like you to ask to practice a bit this excercise and let me know what AVERAGE score are you able to maintain (after at least 40-50 repetitions/attempts).

    Average score as white: 37.10
    Average score as black: 35.00

    https://pl.lichess.org/training/coordinate

    For quite a long time I could not break 30 playing as Black. Nowadays it is much better, but my goal is to reach Average score 40 as white and as black.

    ReplyDelete
    Replies
    1. I have just did it. It turned out practicing about 8-10 hours is enough for me to reach 40 (corrected squares) as Black and as White.

      The highest score of any single attempt (solution) was to score 47 or 48! As for the time of 30 seconds it means 1,5 puzzles (squares) per second!

      Does it make any sense to our conclustions or theory to work untill I score 45 at both of these puzzles? What do you think about it?

      BTW. Dear Tempo - I miss your (new) posts. Could you estimate when the next one is going to appear?

      Delete
    2. i think you waste your time. At an old laptop with a bad internet connection and a touchpad i did score 16 and i think that is already enough

      Delete
  20. thats a typical "visualisation" exercise. You can find many of this type of exercises with some type of rankings here: http://chesseye.alexander-fleischer.de/o/

    There aew many masters supporting this type of exercise , for example : https://www.google.be/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=0ahUKEwjz08SNndPOAhXE1hoKHTTHBmkQFggiMAE&url=https%3A%2F%2Fwww.chess.com%2Fvideo%2Fplayer%2Fachieving-full-board-awareness&usg=AFQjCNHy7bPzw3sHQLkDYN9DyDkiqy0PYA&sig2=_tPC84tKLow8Vk0hojW_vA

    I think such exercises will help you to play blindfolded and the ability to play blindfolded will help with calculation.
    I dont think that this special exercise will help directly to improve the play. But it will help to read chessbooks for example.

    ReplyDelete
  21. OFF The TOPIC:

    I haven't done nearly enough repetitions (obviously): White = 17.5, Black = 16. I just started with this coordinate training, so I'm not too concerned with the score.

    Different Subject:

    The position given below is an example of the kind of position that defies my ability to "see" a clean solution.

    FEN: [r3qb1r/1b1nk2p/p3pp2/6BQ/Np1p4/3B4/PP3PPP/R3R1K1 w - - 0 1]

    It is from the game Yuri Yakovich (2540) - Giorgi Giorgadze (2590), Yerevan open (5) 1996 (found in Fritz 11).

    [Event "Yerevan op"]
    [Site "Yerevan"]
    [Date "1996.??.??"]
    [Round "5"]
    [White "Yakovich, Yuri"]
    [Black "Giorgadze, Giorgi"]
    [Result "1-0"]
    [ECO "D47"]
    [WhiteElo "2540"]
    [BlackElo "2590"]
    [PlyCount "55"]
    [EventDate "1996.??.??"]
    [EventType "swiss"]
    [EventRounds "11"]
    [EventCountry "ARM"]
    [Source "ChessBase"]
    [SourceDate "2004.01.01"]

    1. d4 d5 2. c4 c6 3. Nc3 Nf6 4. e3 e6 5. Nf3 Nbd7 6. Bd3 dxc4 7. Bxc4 b5 8. Bd3
    Bb7 9. e4 b4 10. Na4 c5 11. e5 Nd5 12. O-O a6 13. Ng5 cxd4 14. Nxe6 fxe6 15.
    Qh5+ Ke7 16. Bg5+ N5f6 17. Rfe1 Qe8 18. exf6+ gxf6 19. Rxe6+ Kxe6 20. Qg4+ Kd6
    21. Qxd4+ Kc7 22. Rc1+ Kb8 23. Bf4+ Ne5 24. Nb6 Ra7 25. Nd7+ Ka8 26. Nb6+ Kb8
    27. Nd7+ Ka8 28. Be3 1-0

    The tactical variations are mind-boggling to ME!

    ReplyDelete
  22. PART I:

    I'm nearly through Antonio Gude's book Fundamental Checkmates, excluding the "test" positions. (I'm saving those for a review/test of what [if anything] I have actually learned from this study.) I want to (again) tackle Paata Gaprindashvili's book Imagination in Chess: How to Think Creatively and Avoid Foolish Mistakes, using my "new" process, to "see" if I have really learned anything useful. Previously, the positions in that book seemed WAY above my chess skill level.

    As a control check on what I think I am learning, I went back to Tim Brennan's Tactics Time!: 1001 Chess Tactics from the Games of Everyday Chess Players to apply the same process I have been using with Gude's book. The difference in complexity is startling. Whereas in Gude's positions, there are always many lines of force to be drawn and resolved, in Brennan's book there are usually only one or two or three at most. I think this is one of the sources of the complexity (and thus difficulty) that we experience as players. I have an idea that it would be useful to rate problems/positions by the complexity of the interactions of the lines of force, but that's a topic for another day.

    I've observed something rather interesting about myself as I worked through the book: I have to be very careful about my mental attitude while solving in order to be successful. I tend to shy away from the discipline required to work through ALL possible lines to resolve a position. THAT one "failing" is more responsible for my lack of skill than anything else, IMHO. I have the knowledge and capability to solve even complicated positions with variations running out to 10-12 moves BUT I often settle for a more superficial (partial) "vulture's eye view" (and get the solution wrong, or miss the really significant variations) UNLESS I keep focused on the process I have devised. This is not a skill that is unique to chess; it is purely mental discipline.

    ReplyDelete
  23. PART II:

    My process starts with the "vulture's eye view," an examination and assessment of the whole position. As part of that process, I mentally "draw" lines of force emanating from each piece (and sometimes Pawns) interacting immediately or potentially (within a few moves). (I use a highlighter as a training aid to force the saccades to the proper focal points of interrelationships/points of conflict. This is a training aid that can be discarded after the process is internalized.) As I draw those lines of force, I try to be conscious of the available motifs (not necessarily tactical themes/devices). (I'm not trying to be pedantic about the differentiation between motifs and tactical themes/devices; I'm merely trying to accurately describe what I'm doing.) Perhaps there is a loose piece; an immobile piece; interference between two pieces; a weakness in the King's field; I try to see what motif(s) are applicable and what (if any) additional force(s) can be brought to bear. And so forth, until I have surveyed/examined the entire position. I make no attempt to calculate "I go there, he replies there, I go here, he replies here" types of moves and counter-moves. After covering everything on the board, I (usually) have a pretty good intuitive "feel" for the critical components/pieces/region(s) of the board. If I do NOT do this process completely, almost inevitably, there will be one crucial line of force that I have overlooked, and THAT omission will be why I do not "see" the correct solution! Surprisingly, if I do this process thoroughly, then the solution usually can be "seen" without much difficulty. There are positions (such as the one I gave above) that elude any easy solution, but those are exceptions, not the rule.

    After identifying the available motifs (which give the reason for "what to do" (strategy), I then shift to looking for tactical themes/devices that take advantage of what I have "seen" "how to do it" (tactics). Intertwined with the identification of those themes/devices are possible potential moves that enable those themes/devices to be activated. Somewhat surprisingly, this allows me to "see" several "moves" ahead, even though I am not actually calculating move sequences or variations. It is somewhat more holistic and at a higher level of abstraction than the individual move level. It is more by "feel" than by rigorous thinking about move variations.

    At this point, I shift to calculating the actual move sequences/variations that I will use, guided by the accumulated information from the preceding steps. I don't try to enumerate everything using Kotov's method (calculating like a computer) of the tree of analysis. (As GM Lein complained to GM Tisdall: "I don't think like a tree; do you think like a tree?") Often, the move sequences seem relatively simple to figure out, because it is a matter of priorities (which move has to be done first, then the next, etc.).

    For me, the critical "missing link" is maintaining my mental discipline to "follow the evidence wherever it leads," which means I have to FIRST collect the evidence and make a "map" before plunging into the briar patch. If I stumble off into the briars with no map to guide me, I will almost invariably have no clue how to solve the position and end up "scratched."

    ReplyDelete
  24. I think there are 2 elements of tactical : tactical vision = the ability to spot "simple" tactics quick and calculation/thinkingprocess = to find the solution to a tactic puzzle by thinking more or less long.

    I think it is possible to measure both skills.
    - Tactical vision should be measurable with "fast" puzzles like the puzzles at the chess tactics server
    - Thinking process/calculation should be measurable with slow puzzles like chesstempo in the mixed mode

    for the last months i try to improve at these slow processes, but the sucess is not overwhelming. Visulalisation is a weakness of mine i cant see pattern in the future well enough. Now i am working at a tool to improve here a "Play It Forward". Dan Heisman describes the exercise here https://chesscafe.com/dan/heisman164.pdf : In this exercise
    you are asked to solve normal tactical problems, but not before visualizing the moves (usually two moves for both sides) leading up to them. In other words, you are given a position and two moves for each side, and then, after you visualize those moves, you are
    asked to solve a tactical problem like White to play and win
    .
    Here https://www.youtube.com/watch?v=hCYEAQnfX-Q is a video about a related tool called visualwise. Visualwise does show the move in a different way though. I hope i can generate a few thousand puzzles and calculate the loss of performance of the tactician per move to visualise. This will help me to measure my improvement in visualisation. If some of you would be interested we might share our experience with this tool.

    ReplyDelete
    Replies
    1. I'm going to be writing an article about this on Better Chess Training sometime soon, but I was thinking of this issue as well. I've seen the Visualwise problems and they look interesting, but a simple solution I have found to be helpful is solving 3 and 4-move checkmate problems. I did this extensively when I was around 1500 USCF and it helped me immensely improve this skill.

      I think part of the reason it was so effective was that it isolated the calculation ability because I didn't have to spend as much time on "evaluation" of the positions because it was either checkmate or it wasn't. Also, I used Polgar's Massive tome "Chess" where the mates in 3 were very well selected - e.g. only one correct answer, sometimes deceptive "2nd" answers.

      Delete
    2. There are several subskills of the tactical skill:

      Vision = thats seeing "things" without thinking ( like a simple backrankmate or just a backrankweakness )
      Visualisation = seeing "things" in the future ( after a given sequence of moves : find a fork )
      Calculation = the ability to step through the tree of variations systematically and without too many "loops"
      Thinking method = looking for the candidate moves to be calculated , logical reasoning and so on

      There are several suggestions to improve calculation, for example to solve study's and/or mate in 2 or mate in 3 "problems ! "
      My experience with extra checkmate training was: i did get better in checkmate problems but worse in "gaining material problems" and "pawn promotion problems". My result was just a shift in attention not a gain in "over all performance".

      With the new exercises i am generating is hope! i can adress not only the visualisation skill but the tactical vision skill too.

      The problem with visualwise is, that the puzzles are not adjusting to the skills of the user and there are by far not enough puzzles. I guestimate that someone will need 1000++ puzzles to reach the maximum performance. The last days i did collect material for ~~68000 puzzles and now i need process them to something usable

      I hope i will be able to measure the tactical visualisation abilitys of the user by presenting the puzzles with different "depth". That way i can calculate the loss of performance per visualised move.

      In my opinion a perfect training would start to measure the different skills first and only then creat a trainigsplan according to that. Example is the "Chess Exam of Igor Khmelnitsky"

      Delete
  25. Life is picking up speed, so chess will be on the backburner for a while. As you might have noticed already. But I'll be back. . .

    ReplyDelete
  26. Recently I have played a few serious games with special focus at tactical possibilities. It seems most often I miss correct tactical solutions due to the lack of ideas (themes) and very weak visualisation. When I see all the tactical ideas (themes) clearly - I hardly ever miss tactical shots (except quite deep ones with various variations to consider).

    At some level just solving 3-4 checkmate puzzles is not sufficient to see all the tactical possibilities at the board (especially against stronger opposition).

    A simple explanation. If you learn DEEPLY 20 games that show how to finish off mating attack - your level of "mating attacks" should improve. However if you ONLY learn some skills from solving forcing mate variations (all with checks) for 200-300 checkmate puzzles... you will learn the forcing mating finals. And playing against strong opponent you will not be able to reach the position that says: "White to play and mate in 3 (or 4) moves" (especially if all the moves are checks).

    Do you know my friend what I mean?

    ReplyDelete
    Replies
    1. I agree with you. I was just reflecting on how it was very helpful for me at that point in my development. It wasn't the only type of training I did, but it helped me very well in preparation for a couple tournaments at the time (class tournaments to be sure, but still competitive for me). I don't do them as often now, except for fun, but they were very helpful at that level.

      Delete
  27. Recently I observed my chess tactical understanding lacks a solid fundations. It is directly connected with a better understanding of tactical themes (motifs), but sometimes the lack of vision stops me before executing the winning shot. It is called "cutting the variations too early". I think it may be adressed at serapate topic (thread) as well. We sometimes simply forget (or we are blind) to look closer and see if our "bad/incorrect variation" is really not beneficial.

    ReplyDelete
  28. Like Aox, I dont really know what skill is the driver for a good k-factor.

    I tried Tomasz suggestion to play more for counter-attack. So far the result had only been that my Blitz rating goes down. This isnt unusual when we change the way we think. However, after some weeks I feel this is just not "my" way of improving. In general (for me): counter-attacks create tactics. I am not very good in tactics (compared to my peer group).
    And: counter-attacks (ignoring the opponents attack, and instead attacking something myself) usually backfires badly. It does more harm than good (at least that is my impression).

    Trying more counter-attacks has been an interesting experiment, but it is a really dangerous way to play chess. For me it is like losing control and throw randomly the dice, with having no clue what it will all lead to.

    ReplyDelete
  29. Munich. I want to comment your attempt to change things (improve your chess).

    1. The ability to play counterattack is a MUST. You cannot escape before that as long as you are a perfect positional player (solid defender). However even at some case you simply HAVE TO accept the unsound sacrifice or start your own aggresive actions - otherwise you are lost. It does not occur very often if you are playing against players of your own level (and not very dynamic/aggresive style).

    2. If your blitz goes down (after realizing my suggestion) it is OK. You are not that fast as me and you should not think about being significantly better at blitz because it requires a lot more practice and faster reactions (not to mention being ready for taking much more risk: for example - to win on time you have to move faster not playing a good moves, but fast ones).

    3. Yes, DYNAMIC play and counter-attacks create dynamic positions. They are much more difficult to evaluate, but they are not THAT hard to understand and play. Of course if you have never tried playing such style - you may be shocked at first how difficult it is and how hard it is to understand the positions when they appear on the board.

    4. If you are not good at tactics, try to improve this area. Let me be very precise: you do not have to be VERY fast at tactics, but you have to spot 2-3 movers (tactical shots) without much effort. Unless you learn how to recognize such positions - your progress may be very limited or extremally slow. And of course you have to know how and what to count when the tactical position arise.

    5. I think you did not understand my intention correctly (or I have expressed it in a not precise way. You should TEST tactical skills and dynamic (aggresive) play at blitz, but when playing serious games - play your own style WITH THE attention (focus) at the dynamic play you have been testing while playing blitzes. Otherwise (quote) "It does more harm than good (at least that is my impression)". You do not have to be next Kasparow or Tal who sees all the tactics for 8-12 moves ahead and can evaluate it correct. You simply should FORCE yourself to always look for tactical solutions unless your time and position allow to do so. In a period of 4-6 months you can see the difference. The only requirement is the decision to learn it and use it as additional weapon to beat your opponents. As far as I remember you have NO PROBLEMS at holding most of your opponents, but you seldom beat these guys, do you?

    to be continued...

    ReplyDelete
  30. 6. [Quote] "Trying more counter-attacks has been an interesting experiment, but it is a really dangerous way to play chess". Sorry, but it is simply not true. At least it was NOT my intention to suggest you a way to improve your chess. I pointed out what and how should be done and at first you HAVE to accept failure as your mind has to learn how to rebuild your system of playing to a new level. I had THE SAME experience over and over again. For a short period you take back 2 steps, but after some time (no more than 3-5 months) you "magically" go forward 3-4 steps!

    7. [Quote] "For me it is like losing control and throw randomly the dice, with having no clue what it will all lead to". Playing random chess is a disaster - there is no doubts about it. What I advise is to try playing MORE dynamically and TEST more positions while playing blitz. When you play a SERIOUS game try playing your own, but ALWAYS look for some ways to add the "new weapon" and see how it changes your position.

    It is the same if I had given you an advice: "when you see the possibility to sacrifice your Queen - ALWAYS do it". No! It is ridiculous! What I advise is to TAKE into account the possibility to sac your queen or exchange it for R+B (R+N) or two light pieces, but ONLY at the positions when you have some additional compensation (most often a pssed pawn or controlling the lines or entrance to your opponent's camp).

    To give you some hope and a tiny proof: I have been playing more dynamic (and brave) for the last 2-3 months and recently I beat some players who I could not even draw in most cases. And the more I seek a dynamic hidden at the position, the more often I can find "the only" moves and "magically" some of my strong opponents finally collapse (they make stupid 1-2 move mistakes AFTER the tension goes down).

    ReplyDelete
  31. By the way: my blitz rating goes down for about 120-150 points, but in return my standard rating has increased for about 120-150 points UP! From a mere 1850 up to 2050 (anyway nowadays my rating is about 50-70 points inflated).

    And I made a decision: I COMPLETELY quit playing bullet chess. I can play some blitzes, but bullet chess is VERY detrimental to my playing level. It is simply too random for me. The same may be in your case my friend. If you have been playing (fast) blitzes you could feel the same way as I felt when I have been playing bullets!

    ReplyDelete
  32. Two months has already passed away... is there any hope to find a new article at your blog my friend? I am really addicted (!) to your discoveries, testing and exchanging ideas... that I cannot wait any longer ;) :). Help! Please just let me/us know what is the most probable (likely) schedule to expect a new article! I miss these very much :(.

    ReplyDelete
    Replies
    1. I'm in the process of getting a new job, so priority lies elsewhere at the moment. I can't predict when I will be back, only that I will be back. A lot will depend on how easy the transition will be.

      Delete
    2. Thanks for your reply! I keep fingers crossed on my and your success. In the meantime I will update my chess blog. See you next year! :) Good luck my friend!

      Delete
  33. Curious: I noticed that recently (last couple of days) Blogger (or Temposchlucker) has removed the links to other blogs. Is that an automatic result of prolonged inactivity? I shudder to think that all of the hard investigative work over such a long time time could just disappear into the ether.

    ReplyDelete
    Replies
    1. I don't know, in the past my links continued to exist. Maybe it is a new policy of blogger. They are removed from my template.

      Delete
  34. Good luck to you as you transition into your new job. The Mall where I work in Security is being bought; finalized contract perhaps in November. I have no idea what my situation will be, job-wise.

    ReplyDelete
  35. Best of luck in your new job! Looking forward to your writing in the future.

    ReplyDelete
  36. I keep fingers crossed on you as well my friend!

    Back to chess - You may not believe, but I have started to see the BENEFITS of your chess tips shared in these great articles. I do not waste that much time when I have a bit more complex position. And at some of tactical positions I simply see the best move in 2-3 seconds (!) instead of looking for it... for a minute!

    And I have already strated reading (studying) chess books and I will solve a few hundreds of puzzles (from a workbook) rated 1700-2100. If I find any interesting positions and ideas... I will write them down. When you come back to your blog I will share my findings.

    Looking forward to your writing in the future... as your articles are were inspiring and thought provoking! :) Just keep it in your mind my friend!

    ReplyDelete
  37. Bizarro world: the "My BLog List" just reappeared. I hope that is a result of something you did, rather than just BLogger doing it randomly.

    I hope all is going good for you on the job front. Still no info on the Mall buyout here.

    ReplyDelete
    Replies
    1. The board vision exercises reappeared too. It's just blogger pulling the strings. I did nothing. Or maybe the Russians are trying to make us nervous.

      Delete
    2. I notice that "magic" at the other blogs as well. You did nothing and neither Russians nor Americans had to call FBI or CIA to come back the links magically ;) :).

      BTW. There was some virus that linked to Blogspot links and made some damages - that's why blogspot removed all the links and after some time recover these. That's the version I heard of.

      Delete
    3. The NSA used to copy my writings in silence. Hence I thought the Russians must be involved. But maybe it is just a hacker in need to be a nuisance indeed. I assume we will never know.

      Delete
    4. I told you many time Tempo NOT TO share your chess views as they are really controversial ones! And now you will be forced to explain what it means to decypher chess notation, ideas, motifs and "all the stuff we do not understand and cannot crack at FBI or CIA" ;) :).

      BTW. NSA is helping Karjakin to stay alive at World Chess Championship against Mozart of Chess - Magnus Carlsenovich Carlsenus. Now you know why I have been praising your chess ideas so much, don't you? :D

      Delete
  38. A few days ago I made the decision to solve about 16-18K chess puzzles ONLY from (paper) workbooks or manuals. The complexity (degree) of these tasks are at the range of 1600 up to 2200. My goal is to solve these at the ratio of 90% and see which ones are/were really too hard to solve and what was the reason I could not find the solution.

    If anyone is interested at this chess experiment/test just let me know. You can help me by sharing your questions at the form of list. The more questions the better. You can make suggestions, recommendations or any other form of comments.

    I am going to solve these puzzles to the end of 2019. It means there are just 3 years (and 2 months) left.

    Looking forward to your replies my chss friends.

    BTW. I have already solved about 20K puzzles, but the majority of these were no harder (difficult) then 1600.

    ReplyDelete
  39. @Tomasz:

    I am definitely interested in your results!

    1. What are the source materials that you are using?

    2. How were the puzzles rated?

    I would love to have a program that objectively, consistently and accurately rated puzzles similarly to Chess Tempo and other online tactics servers, without having to have a large group of users solve them, and then retroactively applying the Glicko 2 rating system to the solvers, and thus indirectly to the problems.I am unaware of any such rating program.

    My recent experience with tactical problem solving has convinced me that the idea of "surveying" the entire position, identifying areas of interest (specific pieces/relationships that seem to be "obvious" in some sense) and then visualizing the attack/defense trajectories of all involved pieces (not the same as surveying all legal moves) works best for me. I have to be careful to look at all relevant pieces and all relevant attack/defense trajectories.

    It must be working (to at least some extent) because I have been able to keep my tactics rating on Chess Academy and lichess in the 1800-1900 range without much difficulty. I'm getting a lot more of the 1900-2100+ puzzles to solve, and (often) solving them correctly. Prior to changing the way I visualize, I was never able to get above 1800 on either site.

    As an integral part of the training process, as I previously suggested, I use a highlighter to mark the attack/defense trajectories from each piece to the edge of the board (for long-range pieces Queen, Rook, Bishop) and at least two "hops" with the Knight. The Pawn gets a trajectory to the promotion square or a forking square or a discovered attack if there is any possibility of advancing the Pawn. The King I leave out of consideration unless I sense a mating attack possibility for either player. If you want to differentiate more clearly the attack versus defense possibilities, I suggest using two different colors of highlighters, one for attack and a different color for defense. It is rather amazing how the interactions jump into vision using this training technique. Keep in mind that the goal is to train the vision, so whenever you can "see" the contours without doing the highlighting, do so. Eventually, I found that as long as I focused on rigorously doing this process, I didn't really need to actually highlight the information.

    At first it will seems rather mechanical and you will overlook certain key trajectories, causing you to miss the solution. Careful reconstruction of your thought process will reveal why the mistake was made. In my experience, the most common problem is the thought process, not a lack of knowledge. Take careful note of every problem, especially those that you didn't see all the opportunities. The long-term goal is (as it has been discussed here extensively) to "see" quickly and accurately what resources are available yet hidden in a given situation.

    As you get habituated to the process, you will find that more and more possibilities will appear almost by osmosis. (I almost wrote "by magic" because it fells magical to me.) I've completed nearly 1,000 problems in the Tactics Time book by Tim Brennan and Anthea Carson so far. I already have the Tactics Time 2 book for follow-on training. The follow-on to that is up for grabs at this point.

    I suggest that it will take at least 200-500 problems to get comfortable using this process. If you do try this idea, please post on how it worked (or not) for you.

    Good luck, my friend!

    ReplyDelete
  40. @Robert

    1) The materials I am using are the paper books I have been buying for 20 years. There are 80%-90 of these are Russian (publishing house), and the rest are English. They are workbooks and manuals. Most of these are only puzzles, but some puzzles (for solving) are inside the book related to tactics or combinations (f.e. "1000 chess mating combinations").

    2) The puzzles are rated according to the levels of complexity. For example some of these are recommended to B-A class players (1800-2000), and some with the minimum level you can start solving (1600 and above). They are most often tactics and combinations, but some of these (let's say 20-25%) are related to endings.

    3) What you are doing with the markers ("I use a highlighter to mark the attack/defense trajectories from each piece to the edge of the board") I am doing in my mind. I do not do it in a purpose way, but rather unconciously (with little effort). And I am rather interested at finding holes in my knowledge and skills... to solving all of these puzzles perfectly (with 100% score of correctness).

    4) Your recommended method is a very good way to learn about the interactions and relationships with all the pieces (especially involved in the action). I will use it when I could not solve the puzzle and see what happens. I want to quote you: "Careful reconstruction of your thought process will reveal why the mistake was made". That's what I want to use this method. By the way - I have known this method for about 10-12 years, but I have not used it too often (as I did not need it).

    And yes, I agree with you: "...it will take at least 200-500 problems to get comfortable using this process".

    5) "If you do try this idea, please post on how it worked (or not) for you". As I said before... if I need it I will try it on. However I want to do/test this (and many other) ideas at huge number of puzzles because I want to see which one is best for the specific type of puzzles. It is the same as mentioned by Tempo, Munich or Aox hint: "mate, material gain or pawn promotion".

    And what interest me the most is the connecting (merging) the tactical skills with the knowledge of the particular position and its elements. I estimate my level may rise to 2100-2200 at tactics... after I will have solved 16-18K of harder puzzles. Take notice that 5-10% of these are at the level of 2200 or even (some) 2300! Some people can say: 'damn hard, but they are a master level and that's why they cannot be easy'.

    All the puzzles number is: 43432 (I have a scoresheet at Excel and it adds me all the workbooks puzzles). I have solved about 20-22K of easy puzzles (up to 1600) and the "second part" has to be solved - otherwise I cannot achieve A-level (2000 rating).

    Oh, I could forget one thing! The idea or highlighting the pieces was tested by me at some puzzles (less then 50). It turned out it is not beneficial unless I can understand the ideas/motifs inside the puzzle(s). That's why I want to focus at the motifs first and after I cannot solve the position - use the marker to see what are the sections and areas the pieces are intersectioned and what important relationship I simply missed.

    ReplyDelete
  41. Today I thought again about "what is the driver of a good k-factor". I guess a high k-factor is achieved when you cut down on thinking (calculating) because you lose rule-by-thumb knowledge (guidance - temposchlucker identified them as kind of "patterns", too.
    a) I am a Blitz player of average strength - for an ~1800 fide elo (A-Class) player.

    b) trouble is: I am not a 1800 player but rather a 2200 fide elo player.

    For instance I got beaten a lot in Bullet and blitz games when I played Tomasz (cant remember the outcome, but I lost decisive like 5 versus 20 or so.)

    I dont feel much special when I play longer games.
    I tried to observe my thinking process during longer games. I see all the threats that are there. I try here, and try there. I am not happy with here nor with there. So I try again and again. And suddenly I find a move that keeps it all alive. I check a little longer, but usually, I can move within 20 seconds of checking. In a long game I take 1 minute. But anyway, I am then ready to play the move "that is just right".

    Having said that, it reminds me suddenly of Rubinstein (it was him?) who said something like: "I usually look just 1 move deep - but always the right move!"

    This fits much more to my thinking. It isnt that deep, nor can I imagine people calculate much less than I do - because I really dont calculate that much.

    Often I let myself guide by common sense:

    "knight on the rim looks dim"
    "lose pieces drop off"
    "pushing f2-f4 often is dangerous for the king, even though might not happen something now - it just leaves my king vulnerable, so I follow the recommendation of GM John Rowan who said that the dangers of this move is often underestimated, and if there is no need - better dont move it"
    "My experience is that I lost many games in the past when I played O-O-O, so if not needed - I better go O-O"
    "the h2-h4-h5 attack is not that strong against the g7-bishop fianchetto (dragon-set-up)"
    "knights are the best defenders for the king"
    "rooks dont belong in the center"
    "in the openings - the defenses which move ...Ng8-f6 early have often inferior statistics. Yes, that includes most "indians", too!"
    "if you are white - symmetry is rather good. If you are black - assymetry is rather good."
    "if you have the knight in an endgame: keep your pawns behind. If you have the bishop - advance them and gain space"
    "bishops are far-distance-weapons, and only need a little manoeverability (to increase their activity). So place them on the 2nd or 7th rank, but no more than on the 6th/3rd rank."
    "the pawn constelation c4/d4 and missing e-pawn means that the d4-pawn is notorious weak, almost like an isolated pawn (it doesn matter if b3/a2 pawns are also their, so the c4/d4 pawn is not a hanging duo). To fight this pawn play d6-d5 - cause that isolates the d4-pawn. if you have no d-pawn (as black) put pressure against the d4-pawn on a half-open d-file."
    "The best way of fighting against the Hippo is to do nothing and keep a manoeverable center (and "do nothing for a long time"). With more room at disposal you shuffle your pieces from left to right or right to left - and black has trouble to keep up."
    "in pur pawn endgames you are most often doing well if you stick to "activate king first - pawns later".
    "pure pawn endgames are notorious difficult - if no need to do so, keep the last piece on the board. Dont go into an "even looking" pure pawn endgame if you have no advantage."

    etc.
    I have many, many guidlines - and I feel I have much more guidlines than many other players. If you read through my guidlines: did you know about them (besides of the first 2 which everybody seem to know)? These guidance-rules are maybe one of the drivers of a high k-factor? They help me to cut down on thinking - I dont calculate more, but I actually calculate less!

    ReplyDelete
  42. We have played one blitz game and chatted a bit. The present score at blitz is 16:6 for my favour.

    And I think we made some nice conclusions (discoveries) based on our (various) approach to tactics.

    Munich - the player who use set of rules and filter out the weak moves due to the 'super set pack rules' (sspr). This way he can limit the moves to 1-2 best ones (or good enough/sufficient to stay in the game). He does not need too much energy for this process, but sometimes he may be confused which move is better due to the "necessity for analysis".

    Tomasz (me) - the player who use the experience and the intuition based on hundreds of puzzles analysed and some knowledge of tactics. He rather calculate better positions with the use of tactics, but when he is playing blitz game he STRONGLY rely on intuition.


    This way Munich needs more tactical experience together with deeper understanding of tactics (motifs, refutations and differences between the specific moves at the positions in front of him). If he would be able to add this factor to his chess strenght he would be able to see simple tactics much faster. And it will be enough to be a master level at chess (and in blitz the performance should improve significantly).

    Tomasz needs the set of rules - the same ones as Munich is constantly using. This way he can feel the harmony of the pieces and "cut off" the inferior moves and positions... without the necessity of counting some variations. And the next factor Tomasz has to fix is to understand the dynamic of position better with much better positional understanding (especially plans, counterplay and realising the advantage).

    I hope this short description may trigger a longer discussion. It can help to explain why Munich needs that much time at blitz and cannot beat Tomasz. And in reverse - Tomasz plays much faster even at the cost of weakning the position. He does not hesitate that much and plays the moves "which looks nice".

    ReplyDelete
  43. There are lots of rules I use which substitute or guide tactics.
    I can not write them all down, but I give you hints where to look at:

    a) rook endgames happen often (50% of all endgames), but seem to be the least known. Look a rook endgame book up.

    b) "to take is a mistake" (look up: "the most common error" by GM Smirnov)

    c1) "the threat is more powerful than its execution" --> you have a discovery or a check possibility? If it does not yield anything - keep it!

    c2) (my own rule): you are threatened by an "empty thread" - you dont need to avoid it. You have a good move instead? Dont get impressed by threads that dont yield anything (see point c1). It is itching to react to a threat that is empty - but you dont need to. In a long OTB, you have the time to consider to leave things as they are and follow your own plan. Often you go out of the way eventually 1 or 2 moves later, but you dont need to react too soon. I know it costs time. I know it is itching. But your opponent will think a lot, too, in order to try to make the empty threat work.

    d) dont spend more than 10 minutes on a move. As soon as you are aware you used more than 10 minutes - promiss yourself to move within 1 minute. You need to accept that things are so complicated that you just cant judge it all. Sum your findings up (10 minutes will have given you some insight), think of a general rule (which move is doing more for your centralization? Or which move is safer and wont be dangerous?)
    The feeling not to know which move is better is awful - but you need to let go at some point. More time of thinking only will get you into time trouble later. Also - you will gain a lot of insight in 20 minutes, but when you then speed up you feel that you dont have the same insight that you had when you were thinking for 20 minutes. This is an awful feeling and the main reason why people speed up too late (when they are in time trouble, they speed up, but not as much as neccessary). The gain of insight (the gain in accuracy) you achieve if you use more than 10 minutes is almost always not worth it.

    ReplyDelete
  44. e)opening statistics. I had wild discussions about that issue. Most people will find excuses to why these statistics are not reliable. And the reason are good (for instance rating differences are not taken into account). Nevertheless - I play statistically promissing openings, and guess what? --> I win more games with them!
    I also found strange rules (guidance) about what a good opening looks like. One very odd rule is: if you play a black opening where you move ...Nf6 on move 1 or 2 - it usually is bad. Yes, all the indians are bad. Yes, 1...e5, 2...Nf6 is bad. A lot of openings are bad where ...Nf6 is the third move, too. Yes, 1...e5, 2...Nc6, 3...Nf6 is not great. Yes, 1...c5, 2...d6, 3...Nf6 is not as good as others, and so on....
    You can make up your own rules, but the thing is: try to make sense of the moves and the chaos. There are guidlines of good openings. f2-f4 attacks are not so powerful as people might believe.
    f)I did synergies between my openings. I found 1.Nf3 cuts down most possibilities, and I get either a KID, a Maroczy-bind, a Reti-gambit, or a catalan position most of the time. This enables me to know theory up to move 10-15, with various move orders, but at the end with the same/similar positions. I know them well.
    g) if you have an advantage, the game is "sliding" towards the side who has an advantage. Which means: when you got out of the opening and stand well - you will likely win. It is like as if there was a football match with this special rule: if one team gets a penalty - it is a red card only. And the guy who got the red card is not leaving the ground, but is actually changing the sides! so a 11 vs 11 players football match is then a 12 vs 10 players match. If you start standing bad in chess - you likely lose. That is why GMs concentrate so much on openings. The common believe that people learn too much openings and that this was bad - this is wrong! Openings are very imporant. The trouble is: people dont look at statistics, but try to make a certain opening "work". Maybe because there is sometimes a nice trap. Forget about such seduction: play statistically good openings and you will win more games!
    g)bishops are (like rooks) far distance weapons, too. A white bishop on c4, b5, f4, g5, a5, h5, g3, b3 is prone to attacks cause it is too close to the center. Better statistics yield openings where you place the bishop on g2, b2, e2, d2, or d3, e3 (in case these bishops can not be attacked with Nf6-g5 or Nc6-b4). Only knights feel really well in the center.
    h)in endgames the rule "king first - pawn moves later" is true for most endgames. If there is an exception, than it is: if you have a king and a bishop. Then advance your pawn, and sac your bishop later to create a far advanced passer (break-through tactics).

    ReplyDelete
  45. There is one other "skill" I use, though: I try to avoid sharp tactical positions. This isnt always possible, of course, but we all know that we can have some influence on the character of a game: Do we play a gambit or do we play (like I do) 1.Nf3.
    At the end I can not prevent an aggressive player to sac something against me, but I dont need to pour oil into the fire and start playing gambits myself.

    And another thing worth to know: When I have played a game at lichess.org - afterwards I run the analyzis. This analyzis tells me the amount of inaccuracies (>0.3 pu), mistakes (> 1.0 pu) and blunders (>1.0 pu). It also tells me the average centipawn loss per move. Which is (in my longer time control games) pretty low.
    Now, I dont really know if this high accuracies (= low average centipawn loss per move) is only the result of my guidance rules (the "sspr", see Tomasz comment above). What speaks against these rules is: in most games I apply these rules, but in each game I assume I use only 1 or 2 rules which my opponent does not use. This is hardly enough to get the amount of inaccuracies, mistakes, blunders down from 12 to 6, plus the average centipawn loss shouldnt be too much influenced by 2 moves (influenced by knowing two guidance/sspr rules). On the other hand, I can not rule that out. Just feel that I rather doubt that. Unless I apply even more rules than I am aware of (unconciously). Could be.
    I am going to "test" this and look now for the inaccuracies my opponents did, and try to find out what it was that they missed and why I wouldnt have played have of their inaccurate, or blunderous moves. Was it due to a guidance rule?

    ReplyDelete
  46. P.S. I checked now with one game which I won against a similar rated opponent with

    georgy123:
    6 Inaccuracies
    3 Mistakes
    1 Blunders
    51 Average centipawn loss

    Munich:
    1 Inaccuracies
    1 Mistakes
    0 Blunders
    22 Average centipawn loss

    I checked: I would not have done 3 of his inaccuracies due to general rules.
    I would not have done 2 mistakes, but probably that is a "follow up double mistake" which is one error due to the same rule.
    And I would not have done the 1 blunder, because I identified the checkmate-in-1 pattern.

    So if I had been him, I had done only 3 inaccuracies, 1 mistake, 0 blunders. And mostly because of general guidance rules that would not have even considered the wrong moves he did. (1 blunder and 1 inaccuracies I had avoided due to pattern recognition).

    Here the game:
    https://en.lichess.org/NrYMX7oI/black#69

    I would not have played

    9.Ng5 (rule: develop first, and the knight is not secured but is later just prone to attack. Also: I would like to keep the option to attack the bishop - until it yields something. I would instead be convinced that I follow the rule of development and had played 9.Be2 instead.)

    13.O-O-O?! (it loses a pawn, it is simply hanging. Pattern recognition? Board vision? Probably)


    32.Rg3 (makes the rook less active. I would have liked to keep the option to give check from the side, so I could win the h6 pawn in some variations. Rule: in rook endgames, play active with your rooks. Rooks are poor defenders, but good attackers.)

    33.Kb4? (I know that in rook endgames you can get into a mating attack, similarly like in queen endgames. But a lot of people are carelessly activating their king, not knowing about the dangers. I had defenitely known about this as general rule about rook endgames.)

    34.g5? (this is a follow up error, somehow. White didnt see any danger with his previous move, so he did not see 34...c5+! comming. So this error is either a tactically one, or (in my case) it had not been moved due to the same previous rook endgame rule about king safety).

    Fazit: yes, guidance rules had been a big driver in winning this game and explains my high accuracy.
    The involved tactics were not that deep, so with my CT Blitz rating of 1900 they were mostly under control).

    ReplyDelete
  47. PART I:

    In general I consider that in chess everything rests on tactics. If one thinks of strategy as a block of marble, then tactics are the chisel with which a master operates, in creating works of chess art.” – Tigran Petrosian

    I'm am slowly becoming convinced that chess is more than 99% tactics.

    There is a spiraling cycle back and forth between attack and defense as we gain skill. At first, we "attack" using obvious moves, then we learn how to "defend" against those one-move attacks. We then extend our attacking capabilities another step, perhaps coordinating two pieces in an attack against a specific target, then investigate the corresponding defensive tactics. And so it goes, ever increasing (hopefully) the complexity of what we can "see" until at some point, we reach the "master" level of playing strength.

    I think (but cannot confirm, since I am nowhere near that "master" level) that this process continues with various subtleties as skills improve. In essence, strategy becomes whatever possibilities remain after taking into consideration what the opponent can do to you tactically.

    The problem with using general principles to guide move selection is that often there is a conflict between two or more general principles in a given concrete position. It is also the case (more often than not) that in the specific concrete position, the general principle is NOT applicable (the so-called "exception to the rule"). For fast time controls, this is perhaps a good approach for producing good ideas of what to do. But the consequence is that it trains the mind to use that "general principles" approach when it is (perhaps) not appropriate, like in a long time control game.

    As has been demonstrated, playing by general principles may allow a player to reach a fairly high level of skill. General principles are a distillation of thousands, perhaps millions of specific situations in chess.

    Unfortunately, IMHO, general principles also can be a limiting factor as well. As I learned in martial arts training, YOU WILL FIGHT LIKE YOU TRAIN. If you "train" your mind to utilize general principles to guide you, I believe it will ultimately limit your highest skill level. I certainly am NOT the originator of that notion.

    ReplyDelete
  48. PART II:

    From E. A. Znosko-Borovsky's The Middle Game in Chess, pg. 60:

    "The strength of the position as a whole is decisive here, and this is not a question of an automatic adding up of the various separate elements. Once we have made an analysis of a position we must know how to complete it by synthesis."

    "Two points here are particularly noticeable. The first is the concentrated action of the forces. A piece may occupy a strong square and yet have no prospects and lack all effective force. Worse still, it may have power and yet be of no assistance whatever to the other pieces. It is then sheer dead weight. . . ."

    "The second point, the logical complement of the first, is that, in the middle game, when a certain plan is under consideration, the general principles (occupation of the centre, open lines, strong and weak squares) are of less account than the selection of an object of attack, against which all available forces are to be launched. On this point Alekhine goes so far as to say "all general considerations must be entirely forgotten" and "only that which contributes to the execution of the plan is of any avail."


    John Watson's book Secrets of Modern Chess Strategy is written on the thesis that modern chess essentially has ceased to be driven by general principles or rules, and is instead all about concrete variations from beginning to end. Mihai Suba's Dynamic Chess Strategy is also based on that same thesis.

    Yes, I know that an "appeal to authority" is a logical error. I just think that the more we move away from playing by general principles and focus on the unique specifics of each and every position, the better we will become as players.

    ReplyDelete
  49. But you are well aware that I found out it works pretty much the opposite with me?
    My general principals - I can hardly think of them during blitz games. Blitz is probably mainly "tactics".
    But in my longer OTB games, I think more often about guidance rules, which are not at all tactics.
    For instance: "rooks are poor defenders, but strong attackers", or the Yoda-like witness: "if a time advantage you have - an object of attack you must create"
    Or: "if I have an advantage in development, I need to make the position sharper to profit from this advantage"

    Tactics are for me concrete moves like "if I go here, he has to go there".
    I found out that in long OTB games, mostly shallow tactics are prevalent, but many of the weak moves my opponents did are not weak tactically - they are just wrong according to guidance rules. You might need the ability to see most "easy" tactics. However, my CT Blitz rating of just around 1900-1950 seems to be enough to reach 2200 fide elo. This is not typical for a 2200 player, but in my case it is that way. I dont win my games due to tactics that often, but mostly because I got into better positions.

    ReplyDelete
  50. @ Munich:

    Yes. I am aware of what works for you. Please note that I acknowledged this above (although I did not address you by name):

    "As has been demonstrated, playing by general principles may allow a player to reach a fairly high level of skill. General principles are a distillation of thousands, perhaps millions of specific situations in chess."

    My point is, that although I am conversant with most of the general principles, that (by itself) is NOT sufficient for ME.

    For the rationale regarding the abandonment of general principles and the (almost exclusive) pursuit of the specific concrete considerations in each position, I refer anyone interested to Watson's book Secrets of Modern Chess Strategy. He is infinitely more eloquent in his reasoning than me.

    General caveat applies: no personal or financial connection, blah, blah, blah. . . .

    ReplyDelete
  51. There is an interesting article in the links on the right side of the blog that is somewhat related to this discussion: Hugh Patterson's The Unexplored Territory. His thesis is that given the Shannon number [named after Claude Shannon, which is a conservative lower bound (not an estimate) of the game-tree complexity of chess of 10^120, based on an average of about 10^3 possibilities for a pair of moves consisting of a move for White followed by one for Black, and a typical game lasting about 40 such pairs of moves], there is an unbelievably large game space still available to be explored. It is his opinion that we do not search in this space because we rely on general principles to guide us toward victory, and, although that approach is successful for winning more often than losing, the very success of using general principles inhibits further growth in the realm of the possible.

    This sounds very similar to Watson's thesis in Secrets of Modern Chess Strategy.

    My opinion? I'm fairly certain (probability = 1) that I don't possess sufficient grasp of general principles to go off the beaten path looking for the exceptions. I will probably continue to follow the road MORE traveled, leaving the road LESS traveled to those intrepid explorers called Masters seeking to advance the art of the possible in chess.

    ReplyDelete
  52. I've made some progress (I think) on my tactical vision. I finished going through Tim Brennan/Anthea Carson's two books Tactics Time ! and Tactics Time 2 for the second time, using my highlighter. I'm currently at 1950 on lichess and 2048 on Chess Academy tactics trainers. For the first time in a long time, I've cracked the 1600 (at 1601 - ha! ha!) barrier on Chess Tactics Server without much difficulty.

    I'm still working on tactics. Instead of highlighting from key pieces all the way across the board, I'm just highlighting the key "weak" square(s) that "jump out" while surveying the position. I then try to visualize the tactical themes that combine the solution into a coherent sequence of moves. In some (rare) cases, I find continuations that seem to offer a more complicated defense than the given solution. I'll then explore those kinds of positions in depth, using Stockfish for analysis AFTER I've decided which direction to go. In almost all cases where I "miss" the reason for the solution given, it is because I didn't consider ALL possible "rays" of a key piece. I'm focusing very hard on "seeing" the potential interrelationships between pieces and squares FIRST, before looking for tactical continuations.

    Living in interesting times!

    ReplyDelete
  53. As far as i see these books are just a collection of puzzles which could be puzzles of chesstempo. CTS would be a great instrument to measure the tactical vision if they would have more puzzles.
    Presently i do some : https://shop.chessbase.com/de/products/the_path_to_tactical_strength
    and "Strategy 3.0" http://chessok.com/?p=21207 , which is mainly tactics too

    ReplyDelete