Chess language
The past 21 months I absorbed 2571 patterns. What have I learned?
Tactical themes form a language. There are only about 60 themes (30 mates, 30 motifs like pin etc.), so that are not an awful lot of words to learn.
The problem lies in recognizing (seeing) the themes in all circumstances. At average I have absorbed 2571 / 60 = 43 patterns per theme.
And slowly a new factor starts to emerge. You can only visualize what you have absorbed. In order to visualize a combination that exists of 3 themes, you need to be able to see the 60 x 60 x 60 patterns.
There are 60 x 60 x 60 = 216,000 combinations of 3 themes. You cannot calculate this. But if you can recognize the 60 themes UNDER ALL CIRCUMSTANCES, than you can visualize any combination.
If I remember well, Dan Heisman told that when he was 18, he often had opponents who stopped calculating halfway a combination. He used that to concoct a theory about calculating until quiescence. Which is a typical description by system 2 of a system 1 phenomenon. It is totally useless, because it distracts you from the main problem and asks you to do what can't be done by system 2.
What does this caveman math tell us?
If you know what you are after, it is easier to see whether you are on the right track. Then it doesn't bother you to be busy with only one theme for weeks on end, since there are only 60 themes. There are a lot of subtleties, especially concerning the intersection between themes.
White to move |
3r3r/1pk2p1p/2p1p1p1/b2n1q2/3P1P2/P1PQ3P/2BB2P1/3R1R1K w - - 0 1
Can you can describe all themes and the intricacies concerning the intersection between the themes?
UPDATE
White to move |
There are two clear duplo attacks:
- Blue: Discovered attack of bishop and pawn versus bishop and knight
- Green: Discovered attack of bishop and queen versus queen and own bishop
Temposchlucker asks:
ReplyDeleteCan you can describe all themes and the intricacies concerning the intersection between the themes?
Presuming that was not a rhetorical question...
I’ve been working for some time with GM Aagaard’s three questions as the orientation and starting point for examining a position (or problem). The first question is:
Where are [what are] the weaknesses?
The first thing that caught my “eye” (almost instantly) was the BBa5; it is LPDO. That caused a cascade of other, more specific questions. Is there any White piece that can attack it? Yes, WBd2. What hinders that attack? The WPc3. Can it be moved out of the way with gain of tempo? Yes, 1. c4, simultaneously attacking BBa5 and BNd5. Black now faces the usual dilemma: how to respond with a duplo move so as to not lose a piece.
The first and most obvious rejoinder is to simply capture WBd2. Recapturing with the WRd1 allows Black to exchange the queens, followed by moving the BNd5 to safety. The solution to that problem is to recapture with WQd3, retaining the attack on BNd5 and uncovering an attack on BQf5, another duplo move. Back must save the queen, and BNd5 is lost.
The second rejoinder is to utilize an Equal or Greater Threat (EGT). Since there is no useful check on the White king, a capture or threat to the White queen comes to mind by capturing it. That works EXCEPT that White can capture BBa5 WITH CHECK, retaining the capture potential on BNd5 and BQd3, again a duplo move. Check takes precedence over captures and other threats. After b6, White can capture the BQd3 and retain the option to move away the WBa5 (if Black moves the BNd5 to safety) or capturing BNd5 if Black captures the WBa5. In either case, White has won a piece.
I’m not sure of “all themes” in play. There are discovered attacks, double attacks, intermezzos, and a certain amount of applied logic used to order the questions and to eliminate alternative variations. The questions are triggered sequentially while moving forward logically along a main variation [variation processing].
I did not SEE a complete set of themes nor even think about identifying everything in the position based on a theoretical set of themes and patterns. Instead, it was more like Kline’s RPD [Recognition-Primed Decision] model, although not consciously applied. Patzer “SEES” something “interesting” to System 1 and then investigates it using pattern recognition to guide the search.
The Socratic Method emphasizes following the evidence wherever it leads, rather than arriving at a final answer. It involves clarifying questions, eliminating inconsistent reasoning, and exploring inquiries logically. Maybe it’s more like that.
Yeah, I know: I flunked the question. So, what else is new?!?
UPDATE in blue.
ReplyDeleteThe tactical themes provide the static salient cues. The intersections between the themes and each other or the rest are the dynamic cues.
The rest of the board seems to sleep.
The tactical language with examples/exercises:
ReplyDeletehttps://neoneuro.com/downloads/chuzhakinssystem.pdf Page 32 ff
The Chuzhakins system seems to lean heavily on words and system 2. Have you made progress with it?
DeleteThe interesting point of this position is that one of the targets is your own piece currently (Bd2)
DeleteA picture (with arrows) worth at least 450 of my words. Perhaps I should have waxed more loquacious in my description in order to reach 1,000 words. Just kidding!
ReplyDeleteI’m confused: what is the “pattern” that is associated with these two (I assume) duplo themes?
We have not defined what a pattern is, so I'm flexible on that one. I identified 60 themes. Depending on your definition, that means somewhere between 10 and 100 patterns. If you define a mate as a trap then 30 mates can be defined as 1 theme.
ReplyDeleteI noticed that the static themes emerge when I fiddle around with these positions. It would be a mighty skill when the relevant static themes start to pop up when looking at a position.
The dynamic features are more difficult to spot. Yet their amount is definitely finite.
The dynamic features define the move order.
We find the tempi here (with additional punch), the target exchange (one target replaced by another one, etc.)
I try to get away from the big numbers by focussing on the relevant elements only.
So I try to SEE the pictures and the arrows in my mind and to stay clear from the word salads. And to stay clear from the sleeping part of the board.
I also prefer SEEing to SAYing in my training. I mostly accomplish that by focusing attention on what I SEE, while making a concerted effort to NOT translate what I SEE into words. Unfortunately, Blogger does not allow pictures in comments, so I'm stuck writing word salads in comments. In spite of appearances to the contrary, I do try to minimize the word count.
ReplyDeleteUse as many words as you need. Or like. As long as you don't think you are finished when you verbalize matters. (and you don't think that).
DeleteIt is all about connecting system 1 and system 2 via attention. Since both systems look over the shoulder of attention. Don't get convulsive over avoiding words.
PART I:
ReplyDeleteWhat is the distinguishing features of the example position? In its broadest (most abstract) sense, it is a prime example of the Principle of Two Weaknesses. Weaknesses are so much more than just weak pawn structures.
The first weakness is obviously on the surface: the “weak” BBa5 which is LPDO—with the additional complication that the LoA from WBd2 to BBa5 is blocked at the moment; however, diagonals can always be cleared by a simple pawn move. It takes mere milliseconds to SEE and classify (by looking through intervening obstacles), thereby focusing attention on the attack/defense relationship. When a first weakness is detected, it is natural to SEE if there is a way to take immediate advantage of it. If the WPc3 did not exist, the immediate response would be to capture the loose piece. That this capture occurs WITH CHECK is unimportant for exploiting this first weakness, but it might (and WILL) become important in the larger scheme of things.
How the first weakness came to be is of no importance. It matters not whether it was from an oversight, a miscalculation or as a result of your brilliant play. The important thing is to recognize it.
In a situation where you have identified a first weakness, you have to decide how to exploit it. If it is a situation of adding attackers and/or removing defenders (leaving aside the many ways to accomplish the goal of gaining superiority over the weakness), then focus attention on that one weakness.
On the other hand, if everything seems to be in balance, then keep it in conscious view but do not try to exploit it immediately. This is where the idea of a second weakness comes into play. The second weakness will often be in a different local area of tension. However, there will most likely be some dynamic connection between the two weaknesses.
The second weakness is also on the surface, but it is not recognized by a formal name, and is therefore not as readily apparent because we train ourselves on named tactical themes/devices. It is the “weakness” of any piece when “attacked” by a pawn or piece of a lower value. We consider the pieces to be more powerful than the pawns (reflected in the point count values). However, when attacked by a lower valued piece (or pawn), the higher valued piece must retreat if the threat cannot be neutralized by capturing it or by executing an Equal or Greater Threat [EGT] as an intermezzo. Even in that event, eventually the higher valued piece must give way (presuming the game has not ended). It might seem trivially obvious, but the “attack” on BNd5 can be uncovered by c4, forcing BNd5 to move elsewhere on pain of capture. Such an attack (in isolation) is not considered a “tactic” because it is too simple.
PART II:
ReplyDeleteThe Principle of Two Weaknesses is applicable to every part of chess. It is embodied in the idea of the duplo move. It is embodied in G Averbakh’s generalization and dematerialization of the double attack concept to include more than direct double attacks (for example, a fork of two enemy pieces); threats (of varying orders) can be combined with direct attacks and other threats to create double attacks in the broader sense. It is also applicable to positional weaknesses and to endgames.
As an example of a first weakness, observe the following position from the game Petrosian-Bronstein, Amsterdam 1956.
FEN - 1rb2r1k/1p1n2q1/p2Q2p1/P2Npn1p/2P1N2P/6P1/1R3P2/1R6 w - - 0 36
Black's last move was 35... Nf5. The greatest defensive player in the history of chess did not observe his own first weakness, and played a horrendous move: 36. Ng5. He was immediately punished for his failure to SEE what should have been intuitively obvious as a weakness.
As an example of waiting for a second weakness, observe the following position from the game Christiansen-Karpov, Wijk aan Zee, 1993.
FEN: r2qkb1r/p2p1ppp/1pb1p3/7n/2P1P3/P1N1B3/1PQ2PPP/R3KB1R b KQkq - 3 11
The World Champion ignored a surface-level clue regarding the first weakness (the LPDO BNh5) and failed to consider THAT when he made his next move creating a second weakness. White had just played 11. Be3 and Black replied 11… Bd6. White ended the game with his 12th move. He did not have to create the second weakness; Black did the work for him.
Obviously, usually things are somewhat more complicated. Often the complications arise because the two weaknesses have to be connected by a series of tactical moves. Regardless, the same concept applies.
The devil resides in the details!
An aside:
ReplyDeleteWhile looking for games with the Greek Gift Sacrifice, I “discovered” an interesting feature of Chess Tempo that helped tremendously. I was looking at the following game:
Kamsky, Gata (2709) vs Shankland, Samuel L (2611)
Date: 2014-03-16
Event: 23rd Eastern Class 2014, Sturbridge USA
Round: 4
Result: 1-0
Opening: Indian Game, General (A45)
1. d4 Nf6 2. Bf4 d5 3. e3 e6 4. Nd2 c5 5. c3 Nc6 6. Ngf3 Bd6 7. Bg3 O-O 8. Bd3 Qe7 9. Ne5 Nd7 10. Nxd7 Bxd7 11. Bxd6 Qxd6 12. dxc5 Qxc5 13. Bxh7+ Kxh7 14. Qh5+ Kg8 15. Ne4 Qc4 16. Ng5 Rfd8 17. Qxf7+ Kh8 18. Qh5+ Kg8 19. Rd1 e5 20. Qf7+ Kh8 21. e4 Ne7 22. Qxe7 Bb5 23. Rd2 Qxa2 24. Qf7 Qa1+ 25. Rd1 Qxb2 26. Qh5+ Kg8 27. Qh7+ Kf8 28. Qh8+ Ke7 29. Qxg7+ Kd6 30. Rxd5+ Kc6 31. Qf6+
The GGS occurs after Black’s 12th move. White from that point on is playing for one of two results.
I set the position immediately prior to the GGS and selected the Search Using Current Position button. There were 82 games listed that had reached this position, yet via different opening move sequences. It is interesting that White won 68.8%, drew 17.5% and Black won 13.8% of the games. I didn’t expect Black to do that well.
I’m slowly working my way through the games. What is fascinating is to determine what led up to the GGS position. Did one player create the precondition weakness(es) by forceful play or did it arise merely by accident or oversight?
Now we are talking about caveman mathematics, I did some educated guesswork about the three main areas where I fail. How many rating points will mastery of these areas give me?
ReplyDeleteEndgames 200 points
Openings 100 points
King assault 100 points
This means that tactics can give me an additional 400 points. This might come a bit out of the blue, but I compare my 1700 points with a grandmaster of 2500. 2500 - (200 + 100 +100) = 400
I have had a close look at the commenters of the Tata steel chess tournament. They were commenting with the aid of the evaluation bar of Stockfish. They had IM and GM level themselves.
I got the following impression:
Woking on openings, endgames and king assault give finite results. Working on tactics has not such boundary.
So there is at least some truth in "chess is 99% tactics"
What does that leave as points for positional and strategical concepts/patterns? Or is it simply that gaining all of the available points from the other 4 areas will carry along positional/strategical concepts/patterns for "free"?
DeleteTINSTAAFL might apply.
Positional and strategical points are gathered in the THREE areas of the WHAT.
DeleteThe what (endgame, opening king assault) is finite. The how (tactics) is infinite.
ReplyDeleteProbably all 4 areas (endgame, opening, king assault) suffer from the same finiteness: the law of diminishing returns.
But since tactics are governed by combinations, the amount is huge. Unlimited from a practical point of view. The only limitation is the frequency of occurrence.
The Chuzhakin system seems to be a sophisticated way to count for targets in the position. Maybe it is a clever way to deem whether it makes sense to look for tactics. Less than 5 targets? => make a positional move.
ReplyDeletePlaying the best openings moves known to man was the standard, back in the days of Botvinnik. But in this last Tata tournament, people played the 5th opening choice of Stockfish that was still good enough AND most likely to escape the opening preparation of the opponent.
ReplyDeleteThe COW has not only an option to find novelties, but can show statistics of the usage of specific opening moves at the same time.
The move of Carlson towards freestyle chess might be an attempt to get rid of this part of the game. For me it is a warning not to invest too much time in openings.
I have a clear plan how to attack the opening, which I will explain in due time.
ReplyDeleteI have a clear plan how to approach the ending. I will start with pawn endings.
ReplyDeleteAnd I have a clear idea how to approach the assault of the king. I'm going to use my tactical problem sets for that (389 mates and 350 tactics). It will take me only a few weeks to absorb them fully. After that, I'm ready to start with the assault. Those problem sets contain enough king assaults to work with. I know the games where the positions were abstracted from, so I can see how the assaults came about.
ReplyDeleteLINK: https://www.youtube.com/watch?v=PRSN3vrh5uM
ReplyDeleteVideo: Chess Amateur vs Professional: The Difference
Andrea Botez [Blitz – 1946] vs WGM Dina Belenkaya [Blitz – 2209]
“In this video we show the difference between her and Andrea's approaches when it comes to analyzing the same chess position.”
FEN: r1b1r1k1/1pq2pbp/2P2np1/p1p1p3/2P1P3/2N2NPP/PP3PB1/1R1QR1K1 b - - 0 15
The position is in the late opening/early middlegame stage but with no king attack in sight. Can tactics provide the “clues” for selecting the next move?
How should Black continue — and WHY?
There is a reference to the book Techniques of Positional Play: 45 Practical Methods to Gain the Upper Hand in Chess by Anatoli Terekhin, Harald Keilhack, and Valeri Bronznik [New in Chess, 2023]. Every Russian schoolboy supposedly is very knowledgeable of the patterns in this book (in the original Russian edition).
Game:
Portisch, Lajos (2620) vs Fischer, Robert J (2690)
Date: 1967-10
Event: Sousse Interzonal+, Sousse
Round: 2
Result: ½-½
Opening: King's Indian Defense, Fianchetto Variation, Classical Main Line (E69)
Problems: 122947, 122948, 122949
1. Nf3 Nf6 2. g3 g6 3. c4 Bg7 4. d4 O-O 5. Bg2 d6 6. Nc3 Nbd7 7. O-O e5 8. e4 c6 9. h3 Qb6 10. Re1 Re8 11. d5 Nc5 12. Rb1 a5 13. Be3 Qc7 14. Bxc5 dxc5 15. dxc6 bxc6 16. Na4 Bf8 17. Qb3 Nh5 18. Qe3 Qa7 19. h4 Ng7 20. Kh2 f6 21. Bh3 Bxh3 22. Kxh3 Ne6 23. h5 gxh5 24. Rh1 Rad8 25. Kg2 Qg7 26. Kf1 Qg4 27. Rh4 Qg6 28. Qe2 Bh6 29. b3 Rd7 30. Rd1 Rxd1+ 31. Qxd1 Rd8 32. Qe2 Bg5 33. Nxg5 fxg5 34. Rxh5 Rd2 35. Qg4 h6 36. Rh2 Kg7 37. Nc3 Rd3 38. Nd1 Qf7 39. Kg2 Qd7 40. Qf5 Rxd1 41. Qxe5+ Kg8 42. Rxh6 Ng7 43. Rg6 g4 44. Rxg7+ Qxg7 45. Qe8+ Kh7 46. Qh5+ Kg8
Of course you can't deduce the WHAT from the HOW. But the WHAT isn't really rocket science. Every Russian kid knows it. And since the WHAT is rather finite, it is difficult to make the difference with the WHAT. It is the HOW where we can make the difference.
DeleteNeedless to say that we need to master the WHAT first.
I had a closer look at 30 tactical themes of the caveman approach. 5 of them are absolutely critical:
ReplyDelete- double attack
- discovered attack
- pin
- skewer
- annihilation of the defender
If you are able to SEE all of these automatically, than this will boost your tactical acumen ginormously. There is hardly a combination without one or more of these themes. So after solving a tactical puzzle, make that you identify the relevant themes!
I tried to do the same with the 30 mates, but you need them all.
The mentioned 5 tactical themes have a limited (finite) amount of scenarios coupled to them. After absorbing the scenarios, the relevant scenarios pop up automatically.
ReplyDeleteI'm confused (so what else is new?!?):
Delete"After absorbing the scenarios, the relevant scenarios pop up automatically."
Aren't the scenarios coupled to themes also just the scenarios that are relevant, or is there a difference that I'm missing?
After absorption, I presume (perhaps erroneously) that the relevant scenarios associated with a particular theme WILL pop up automatically. Isn't that the point of absorption?
A theme has a lot of scenarios attached (1:m).
DeletePin:
- a pinned piece is a bad defender
- attacking the pinned piece
- ways to unpin
- absolute and relative pins
- etc.
Some scenarios are theme independent (n:m).
Pin:
- add a defender
- add an attacker
- clearance
- annihilation of the defender
- etc.
In different positions with pins, different scenarios are relevant.
The magic of skill is that the relevant scenarios pop up.
Thanks for the clarification!
DeleteI have selected 350 puzzles. I.e. the first five chapters of the 1001 tactics for club players.
ReplyDelete- double attack
- discovered attack
- pin
- skewer
- annihilation of the defender
I'm studying the scenarios until they are absorbed.
In 100 Tactical Patterns You Must Know: Learn to Recognize Winning Chess Moves, Frank Erwich had this to say about his approach:
ReplyDelete[EXCERPT]
In 100 Tactical Patterns You Must Know, I offer a unique perspective. While presenting a diverse array of tactical themes, I focus not only on general positional features like undefended pieces or an exposed king, but also on the specific arrangement of chess pieces, a crucial factor in executing successful tactics. Focusing on typical piece configurations allows you to evaluate and decide more quickly and efficiently in certain positions.
{He provides two examples; I have left out the verbiage.}
Marcin Dziuba [2310] vs Frank Erwich [2196], Litohoro European Championship U16 1999:
FEN: q4rk1/rb2p1b1/1n1p2Q1/2pP4/8/6P1/PP2PPB1/RN2K2R w KQ – 0 19
Wesley So [2656] vs Anish Giri [2588], Wijk aan Zee 2010:
FEN: 7k/2p3pp/p7/1p1p4/PP2pr2/B1P3qP/4N1B1/R1Qn2K1 b - - 0 37
{This game is on chessgames.com.}
Piece placement is the key to identifying what chess patterns are available in a particular position. The key information [scenario?] here is that the bishop is pinned by the queen and the rook on the open file can directly attack the king. The checkmate itself differs from my game, but the pattern is almost identical.
{IMHO, there is literally NOTHING that is “almost identical” between the two positions, EXCEPT, of course, the abstracted checkmate sequence requiring the rook to be “sacrificed” while attacking the king, distracting and forcing the king to step into the mate position.}
Indeed, a player of So’s calibre should have seen this, but unfamiliarity with this specific pattern led him, like me, to have a blind spot for this possibility. Undoubtedly, So had solved many tactical puzzles in his life, but that was not enough to prevent the disaster in the game.
If you haven’t seen something before (or have seen too little of it), it turns out to be very difficult to find it over the board. And this is the point I want to make. Having more specific knowledge about patterns facilitates the process of finding the right moves. Instead of struggling with different candidate moves, the specific positions of the pieces can immediately tell you what you can and cannot do in a given position.
[END EXCERPT]
There is a companion Workbook that has over 500 puzzles, based on the same concept.
A question:
Do these two examples belong to the same scenario?
Yes. The mate combination is prepared by the preparatory move "the magnet". A piece is sacrificed with check to "draw" the king to a square where he can be mated. It is quite common. For instance in combination with Lolli's mate.
Delete"If you haven’t seen something before (or have seen too little of it), it turns out to be very difficult to find it over the board."
DeleteThis is absolutely critical.
Calculation gets another meaning. Instead of a search for a new variation, it becomes checking the existing scenarios that pop up.
ReplyDeleteWhen no scenario pops up, there is no tactic. Then it is time to look for a positional move.
ReplyDeleteThis can only be reliable when you are sure that you have absorbed all or at least the most frequent occurring tactical scenarios.
The transition to the endgame and the endgame of both these games is instructive.
ReplyDeleteWhat should be the plans for both sides, based on the advantages and disadvantages of each side?
Game 1:
FEN: 8/8/PR2p3/4kp1P/5p2/r5p1/6P1/5K2 b - - 0 45
Black won (after White missed a mate-in-1 and blundered his queen).
1. d4 d5 2. c4 dxc4 3. e4 Nf6 4. e5 Nd5 5. Bxc4 Nb6 6. Bd3 Nc6 7. Nf3 Bg4 8. O-O Nxd4 9. Nbd2 Nxf3+ 10. Nxf3 c6 11. h3 Bxf3 12. Qxf3 e6 13. Rd1 Qc7 14. Qg3 g6 15. Bg5 Bg7 16. f4 Nd5 17. Be4 O-O 18. Rac1 Qb6+ 19. Qf2 Qxf2+ 20. Kxf2 h6 21. Bxd5 hxg5 22. Be4 gxf4 23. Rd7 Rab8 24. Rc5 Rfd8 25. Rxd8+ Rxd8 26. Kf3 g5 27. Ra5 a6 28. Ra3 Bxe5 29. Rb3 Rd7 30. Bd3 Kg7 31. a4 f5 32. Bc4 Kf6 33. a5 Bb8 34. Bxa6 bxa6 35. Rxb8 Rd5 36. b4 Rb5 37. Rb6 Ke7 38. Rxc6 Rxb4 39. Rxa6 Rb3+ 40. Kf2 g4 41. Rb6 Ra3 42. a6 g3+ 43. Kf1 Kf6 44. h4 Ke5 45. h5 Ra1+ 46. Ke2 Ra2+ 47. Kd3 Rxg2 48. Rb5+ Kf6 49. Ra5 Rh2 50. a7 g2 51. a8=Q Rh3+ 52. Kd2 Rg3 53. Qd8+ Kg7 54. Ra7+ Kh6 55. Qf6+ Kxh5 56. Rh7+ Kg4 57. Rh4+ Kf3 58. Qxe6 Rg8 59. Qe2+ Kg3 60. Qe1+ Kf3 61. Rh3+ Rg3 62. Qxg3+?? White resigns
Game 2:
FEN: 2r3k1/pb2rpp1/5n1p/q1pp4/P7/2PBPN2/1Q3PPP/3RR1K1 w - - 2 20
White won.
1. d4 e6 2. c4 d5 3. Nc3 Nf6 4. Bg5 Be7 5. Nf3 O-O 6. e3 b6 7. cxd5 exd5 8. Bd3 Bb7 9. Qc2 Nbd7 10. O-O c5 11. Rfe1 Rc8 12. a4 h6 13. Bh4 Re8 14. Qe2 Ne4 15. Bxe7 Rxe7 16. Rad1 Ndf6 17. dxc5 Nxc3 18. bxc3 bxc5 19. Qb2 Qa5 20. c4 d4 21. Re2 Bc6 22. exd4 Bxf3 23. gxf3 Rxe2 24. Bxe2 cxd4 25. Qd2 Qxa4 26. Qxd4 a5 27. Ra1 Qb4 28. Bd3 Re8 29. Bc2 Re1+ 30. Rxe1 Qxe1+ 31. Kg2 Qe7 32. c5 Nh5 33. c6 Qg5+ 34. Kf1 Nf4 35. c7 Black resigns