Wild speculations
It turns out that careful observations and a bit of logical reasoning has brought us a new way to look at potential chess improvement by adults. We found a less rigid way to look at STM and LTM. We have to proof that it is actual viable by putting it to the test, of course. But for now we are in that pleasant state that we can make wild speculations based on the new ideas. A state that will last only until we have proven that it is NOT viable. Which hopefully will never happen, of course.
What wild speculations, or to put it more dignified, hypotheses can we build on the new ideas?
We have defined two ways of storage in LTM. Prestored memories and on the fly stored memories. I noticed for the first time the common origin of store and story, by the way.
Ok, let's fire away.
Prodigies are discovered, and coached by a coach. A coach let them explain their moves with a narrative. That's the reason why talentful youngsters make such fast progress.
When narratives of the same kind are often stored on the fly in LTM, there will come a moment that they will be prestored in the LTM. These prestored memories will dramatically speed up the process of creating new narratives.
When there are no new narratives created and stored on the fly, the stock of prestored memories will no longer grow, and the rating will stall, eventually.
The automatic pilot is the enemy of creating new narratives.
A stalling rating can get pulled raft again by creating new narratives and storing them on the fly.
What wild speculations, or to put it more dignified, hypotheses can we build on the new ideas?
We have defined two ways of storage in LTM. Prestored memories and on the fly stored memories. I noticed for the first time the common origin of store and story, by the way.
Ok, let's fire away.
Prodigies are discovered, and coached by a coach. A coach let them explain their moves with a narrative. That's the reason why talentful youngsters make such fast progress.
When narratives of the same kind are often stored on the fly in LTM, there will come a moment that they will be prestored in the LTM. These prestored memories will dramatically speed up the process of creating new narratives.
When there are no new narratives created and stored on the fly, the stock of prestored memories will no longer grow, and the rating will stall, eventually.
The automatic pilot is the enemy of creating new narratives.
A stalling rating can get pulled raft again by creating new narratives and storing them on the fly.
just solving tactics dont help ( a lot ). That we know . I put my hope in the intense aftermath and embeding in my tactics-framework.
ReplyDeleteNo worries. As usual I will be my own guinea pig.
ReplyDeleteRegarding narratives (“stories”):
ReplyDeleteWhat form do these narratives take?
Dr. Lasker provides a clue:
“In Chess, however, that number [of the different lines of play], no matter how critically one may select and how many feeble lines of play one may reject, goes into many millions. The brain cannot encompass them by a process of mere compilation. One must therefore SEARCH FOR RULES, LAWS, PRINCIPLES CAPABLE OF COMPRISING WITHIN THEIR COMPASS THE RESULT OF A THOUSAND, NAY OF TEN THOUSAND DIFFERENT VARIATIONS.”
“[Steinitz] concluded that SOME CHARACTERISTIC, A QUALITY OF THE GIVEN POSITION MUST EXIST THAT TO A DISCERNING EYE WOULD INDICATE THE SUCCESS OR FAILURE OF THE SEARCH (FOR A COMBINATION] BEFORE IT WAS UNDERTAKEN.”
“Wherever in Chess APPARENTLY the weaker force is victorious, a compensation can be detected, and THIS COMPENSATION CAN BE ACCURATELY CIRCUMSCRIBED IN TERMS OF A GENERAL RULE.”
So the form of the narratives is as “rules” or “principles,” perhaps phrased as a heuristic, a “rule of thumb.”
But. . . we already know from experience that general principles (heuristics) provide ON AVERAGE (at best) a pointer in the right direction. The higher the skill level, the LESS likely that the player is basing his lines of play on general principles.
IN CONCRETE ANALYTICAL POSITIONS, GENERAL PRINCIPLES ARE INSUFFICIENT TO DETERMINE THE CORRECT LINE OF PLAY.
What is an example narrative (JUST ONE!) that encapsulates an important TACTICAL concept into a recognizable pattern that can trigger System 1?
I'd be perfectly happy to "volunteer" as a guinea pig, IF I can get some clear idea of what to do!
I'd be perfectly happy to "volunteer" as a guinea pig, IF I can get some clear idea of what to do!
DeleteI'm working on that.
As a starting point, may I suggest some type of baseline "test" in order to accurately measure how much improvement actually occurs? Perhaps a timed test in blitz mode using Chess Tempo to establish an approximate starting rating? I don't think it really matters what mix of problems are used for this baseline test. It would really be nice if that same set of problems could be captured and then retested at the end of the designated training period, but I don't think that is crucial to demonstrate that an improvement has (or perhaps has NOT) occurred.
Deletewhen you do the same ste of puzzles again after a longer while and a big set might be ok. But how to judge the differneces in performance?, best is you do al least 1000, better 4000 CT puzzles in blitz mode ti get a baseline, with 30 puzzles per day +- "calculating" an average rating over the last 2000 attempts.
Delete@ Aox:
DeleteThanks for the concrete suggestion!
At present, I am on a "free" subscription on Chess Tempo. Do you have to be a paying member in order to calculate the statistics, like the average rating over the last 2000 problems? If so, what is the lowest level membership that gives you the necessary option(s)? As a paying member, can you do the calculations for a different member based on that member's results?
I'm a little puzzled by the extended "test" quantity and the corresponding time (weeks, if not months) to accomplish it. Wouldn't doing this number of problems over several months cause the rating to change (hopefully to improve and not to go down) as a result of the exposure to the various problems and subsequent gain in "pattern recognition" (if not also in skill)? Perhaps this effect is offset by the assumption (based on previous experience) that after 4000 problems, the tactics rating will stabilize sufficiently to make the average rating over the last 2000 attempts meaningful as a starting point for further improvement via Temposchlucker's new method of training.
Okay, not being lazy: I checked out Chess Tempo's Premium Membership options. At this point, I don't see the added value sufficient to purchase a Diamond subscription. I doubt I will ever reach the skill level where I need 8 engines running simultaneously. It looks like the Gold subscription provides the "most bang for the buck." $4/month or $35 per year seems quite reasonable for all the features.
DeleteWhat the heck, it's only money: I just signed up for a year of Gold membership at Chess Tempo!
DeleteI was doing a lot of statistics about the "non improvement" of tacticians at chesstempo.
Deletethe problem is.. you allways go up and down with your rating.. the typical new tactician at ct needs ~1000 puzzles to reach their "plateau" , to get realy use to the situation: solving tactical puzzles under time-preassure.
Some tacticians need more puzzles though..
you dont need a premium membership if you write down your start rating every day ( or every week )in a spreadsheet. But its a good thing to support richard in his work
As a gold member you can download your history and put it into a spreadsheet, there you can easily average your ratings. To do so.. you just need a gold membership at the >day< you do the download the history
a new chunk
ReplyDeletehttps://ibb.co/LrPgNfD
Black to move and win
Easy to see: RxBf3 wins
But why is Qf5 not winning? Its a double attack on the weak pieces Bf3 and Rc8?
White has a DeDoubleAttack-Method i was not that aware of: Bd1 is removing one of the targets of the double attack and an equivalent piece for the Rc8 is now under attack: the Rb3
My first inclination was to capture 1. ... RxBf3. Why? Because of two things: (1) If the White King recaptures on f3 with 2. Kxf3, then Black checkmates with 2. ... Qe4#. (2) Capturing on f3 also protects the Black Knight on f6, in the event that White decides to take advantage of the pin by the White Queen and plays 2. Rg8+. That line doesn't appear to go anywhere after 2. ... Kh7 3. Qf8 Nxg8, again because the Black Rook protects f7. After 4. Kxf3 Qf5+ seals the deal.
ReplyDeleteI didn't even consider 1. ... Qf5 because it is not as forcing as the capture.
How do you perceive this as a "chunk"?
(Not arguing with that assessment; just looking for more information.)
i did see 1...Qf5 first and made that move. I did not see the refutation partly because i thought this doubleattack is not refutable, the other methods i had in my mind to act against a double attack did not work.
ReplyDeleteThe chunk is this special method to fight a doubleattack
By stubborn calculation, moves of pieces backward are hard to find.
Is this "chunk" bound to the general aphorism "Equal or Greater Threat" combined with the specific "Backward movement removing threat (double attack) with simultaneous counter threat"?
ReplyDeleteI'm trying to figure out how to summarize this "chunk" so as to be able to remember AND to be able to apply it to other positions in the future on the basis of pattern recognition/matching.
A question: are "chunks" summarized in the form of general "rules" (principles, common "sayings" or "pearls of wisdom") which are so often imparted by masters for the benefit of us non-masters? Or is there much more to "chunks" than these shorthand "Cliff Notes"?
My previous impression of "chunks" (based on the relevant literature: Simon & Chase, etc.) has been that they are abstract visual-spatial relationships that have been generalized into a template form with "strings" attached (using the "magic" of System 1), and therefore not readily adaptable to generalized conscious "rules" (that are consciously memorizable and thus recoverable using System 2). This new line of inquiry (vis-a-vis "stories") may show my conception to be either incomplete or just factually wrong. Your thoughts?
Thanks for your help! My apology if I seem obtuse.
This is how i see it.. not that i would be a specialist in these things:
ReplyDeleteits a "chunk" of the type fork, double attack, pin, decoy..
If we have the tactical weakness that "the oponents queen and the oponents king are at the same diagonal" then the method: bring a bishop ( or the queen ) at this diagonal leads (often) to a fork, a pin or a skewer. Now there are methods to unpin, unfork and unskewer to consider.
About the term "chunk".. Its a fuzzy term for me too. wikipedia says this :)
https://en.wikipedia.org/wiki/Chunking_(psychology)
"pieces of information are bound together into a meaningful whole" , a unit which can be handled as one thing in our STM.
A "Whitequeen" at "f3" and a "Whiteking" at "d1" and a "blackbishop" at "g4" are 3 (or even 6?) chunks, Bishop forks Queen might be a single chunk.
Such "chunks" are not just rules/principles. How often some weaker player did ask me to teach them. I usually did start with endgames and/or openings. Openings are "easy" because there are the "principles of the opening" thats a "small" set of rules. I use this collection of them for explaining : https://exeterchessclub.org.uk/content/ten-rules-opening
All! of my students said: i know them, we dont need to talk about that; virtually now one was following them. In the heat of the battle they move pieces twice in the opening, make to many pawnmoves, dont pay attention to the center and so on. When you ask them why they do so they find reasons and excuses...
These rules are only known as sentence/phrase but not as played chess "chunk".
I try to show them what they do by analysing their first moves acording to the list of the exeter chessclub : how many rules are broken by this move... hard hard work with adults, easy with ( interested )kids
The standard example for chunks is a telephon number, its usually hard to memorize 6 numbers ( 4 8 6 3 1 8), but when you group them to ( 48 63 18 ) or even to ( 486 318 ) its easier for most people. This can only work if 48 and/or 486 is already an internalized chunk.
I watch a lot of chess videos at youtube, i remember one master who always asked his students to think about the continuation and he permanently insist that an answer could not be just a single move! He always did want more from his students, an idea, a line ... -> bigger chunks! To think chess just in sequences of single moves is..not good
GM Smirnovs courses are the same, he is reducing chess to a small set of principles and thinking methods and hammering them into the brain. he says it would take several weeks to internalize one.. im afraid i need a bigger hammer ;)
PART I:
ReplyDeletePerhaps some applicable thoughts from THINKING, FAST AND SLOW, Daniel Kahneman:
“. . . what we consider as “expertise” usually takes a long time to develop. The acquisition of expertise in complex tasks such as HIGH-LEVEL CHESS, professional basketball, or firefighting is intricate and slow because EXPERTISE IN A DOMAIN IS NOT A SINGLE SKILL BUT RATHER A LARGE COLLECTION OF MINISKILLS. CHESS IS A GOOD EXAMPLE. An expert player can understand a complex position at a glance, but it takes years to develop that level of ability. Studies of chess masters have shown that at least 10,000 of dedicated practice (about 6 years of playing chess 5 hours a day) are required to attain the highest levels of performance. During those hours of INTENSE CONCENTRATION, a serious chess player becomes familiar with thousands of configurations, each consisting of AN ARRANGEMENT OF RELATED PIECES THAT CAN THREATEN OR DEFEND EACH OTHER.”
“Acquiring expertise in chess is harder and slower than learning to read because there are many more letters in the “alphabet” of chess and because the “words” consist of many letters. After thousands of hours of practice, however, chess masters are able to read a chess situation at a glance. The few moves that come to their mind are almost always strong and sometimes creative. They can deal with a “word” they have never encountered, and they can find a new way to interpret a familiar one.”
“The accurate intuitions . . . are due to HIGHLY VALID CUES that the expert’s System 1 has learned to use, even if System 2 has not learned to name them.”
“… human learning is normally efficient. If a strong PREDICTIVE CUE exists, human observers will find it, given a decent opportunity to do so. . . . In the absence of VALID CUES, intuitive “hits” are due either to luck or to lies.”
“EXPERTISE IS NOT A SINGLE SKILL; IT IS A COLLECTION OF SKILLS, and the same professional may be highly expert in some of the tasks in her domain while remaining a novice in others. By the time chess players become experts, they have “seen everything” (or almost everything) but chess is an exception in this regard.”
Let’s revisit Temposchlucker’s analogy of learning to use the brakes, in Dr. Kahneman’s words:
“Some regularities in the environment are easier to discover and apply than others. Think of how you developed your style of using the brakes on your car. As you were mastering the skill of taking curves, you GRADUALLY learned when to let go of the accelerator and when and how hard to use the brakes. Curves differ, and THE VARIABILITY YOU EXPERIENCED WHILE LEARNING ensures that you are now ready to brake at the right time and strength for any curve you encounter. The conditions for learning this SKILL are ideal, because you receive IMMEDIATE AND UNAMBIGUOUS FEEDBACK every time you go around a bend: the mild reward of a comfortable turn or the mild punishment of some difficulty in handling the car if you brake either too hard or not quite hard enough. Whether professionals have a chance to develop intuitive expertise depends essentially on the QUALITY AND SPEED OF FEEDBACK, as well as on SUFFICIENT OPPORTUNITY TO PRACTICE.”
Before moving on to more relevant discussion, one more excerpt from Dr. Kahneman:
“Substituting one question for another can be a good strategy for solving difficult problems, and George Pólya included substitution in his classic How to Solve It: ‘IF YOU CAN’T SOLVE A PROBLEM, THEN THERE IS AN EASIER PROBLEM YOU CAN SOLVE: FIND IT.’ Pólya’s heuristics are strategic procedures that are deliberately implemented by System 2.”
Hmmm. . . it seems that CT-ART uses this idea of an “easier problem” with regard to the “simpler” (similar) “problem” using a 5x5 board. But I digress. . .
System II can't describe in words what to do when entering a curve. Since every narrative would turn out to be silly. System II works verbally. System I doesn't understand words. It does understand feedback though. Who is giving feedback? How does the feedback work?
DeletePART II:
ReplyDeleteI’m going to use the definition of a “chunk” provided by Gobet et al in the article Chunking mechanisms in human learning, cited in the Wikipedia article on “Chunking” that Aox referenced:
“In spite of the surface variety among descriptions, a common definition of a chunk is possible: A CHUNK IS A COLLECTION OF ELEMENTS HAVING STRONG ASSOCIATIONS WITH ONE ANOTHER, BUT WEAK ASSOCIATIONS WITH ELEMENTS WITHIN OTHER CHUNKS.”
Master Martin Weteschnik (Understanding Chess Tactics) in Chapter 5 The Double Attack defines this as the most common “motif” in chess. He gives a scheme for breaking down the double attack into three points: “There are two points, which are attacked. I will call them tactical target 1 and tactical target 2. They are marked by a box. The third point is the square from which the attacker is threatening the two enemy pieces. I will call this point the tactical base and it is marked with a circle.”
If I understand the notion of “chunking” correctly, then this “chunk” of a DOUBLE ATTACK has three strongly associated components. The difficulty in recognizing this “chunk” while playing (or solving problems) lies in the absence of one or more of the requisite components. Note that there is NO visio-spatial connotation whatsoever in the abstract definition. However, those visio-spatial aspects DO automatically come into play as soon as we designate a particular type of piece on a specific square as the tactical base. This is some of the additional “baggage” that comes “for free” with the “chunk” mechanism.
Rather than pursue this notion using tactics for examples, I switch to the endgame. There are two fundamental “chunks” associated with the endgame K+R+P versus K+R. These two “chunks” are designated the Lucena and the Philidor positions. In Silman’s Complete Endgame Course: From Beginner to Master, these two endings are introduced at the USCF Class “C” level (1400-1599). More complicated variations are introduced as the player proceeds to higher skill levels.
Let’s consider those two endings as essential “chunks” in our chess skills, since most Rook and Pawn endings resolve to one of these two situations, and Rook endings constitute approximately 50% of all endings.
I have questions:
(1) Do you know which of these two endings is a winning procedure and which is a drawing procedure?
(2) Can you describe and summarize the winning/drawing procedure?
(3) Can you play out either of these situations regardless of initial piece placement?
I submit that KNOWING the answers to the first two questions is no guarantee that we possess the SKILL to play these positions correctly in almost all situations.
So, bottom line: we THINK we “know” how to do something, usually when we grasp the concept and can place a name on it. The reality is we often only have a vague “word salad” concept that does NOT correspond to SKILL.
A CHUNK IS A COLLECTION OF ELEMENTS HAVING STRONG ASSOCIATIONS WITH ONE ANOTHER, BUT WEAK ASSOCIATIONS WITH ELEMENTS WITHIN OTHER CHUNKS.
DeleteIn order to be able to treat a bunch of elements as a whole, you need to glue them together. A narrative is such glue. Geometric pattern is such glue. Association = glue. Concepts are glue.
A narrative is a summary. In stead of working with a book, you work with the summary. But the precondition is that the summary must be correct, of course.
In summaries, you lose detail. When you need detail, you must reconstruct them from the summary.
PART III:
ReplyDeleteIt is the same idea when applied to the concept of EQUAL OR GREATER THREAT (EGT). We THINK we “know” this concept, but we don’t recognize it when it arises on the board, as in this problem. Why not? Because we are focused on other things, because we do not “see” that backward movement of a piece accomplishes multiple things at the same time ( EGT and removal of a target. In short, we have a surface-level “knowledge” of the definition, but do not possess a “gut feeling” for “seeing” it regardless of the configuration of the pieces.
This brings me to the Tree of Scenarios. I think it is an invaluable tool for organization of training and a thinking process. However, it must be TRAINED with many examples until it becomes an integral part of our System 1. Many examples must be attached to each branch of the tree through a process of conscious thought and investigation. Just because it is logical does NOT automatically make it useful! It is only through a long and involved process of bending it here and there, and seeing if it still works can we embed it into System 1. The name that we give each branch is relatively unimportant.
I think this may be at the root of why we adults seemingly cannot improve. We “learn” and “know” all the right stuff (verbiage) at the surface level, but there is nothing (or very little) concrete (specific positions; games) attached to it to flesh out the skeleton.
As I was studying the ideas recently espoused regarding narratives, I realized that my own approach to studying was to search for the explicitly stated “rules” (principles, whatever), and that I skimmed over the concrete details (in the form of variations). As a consequence, I could “talk the talk” but I doubt very seriously if I could “walk the walk” in unfamiliar situations. I need to change that aspect of my training.
Robert said : "We THINK we “know” this concept, but we don’t recognize it when it arises on the board, as in this problem. Why not? Because we are focused on other things, because we do not “see” that backward movement of a piece accomplishes multiple things at the same time ( EGT and removal of a target. "
ReplyDeleteI think its to slow to only! look at a move an think: "what is the function of the move"
i did see the double attack ( fork ) immediately ( they are quite common ;), without un-double attack for the opponent this would be the! killer move . Now i have to check.. ist there a refutation of my idea? And i know: it has to be a undouble attack. So i have now the situation: i (should) "know" the function and i am looking for the move.
move->funktion and!! function->move
thinking (only) in moves is wrong ( not sufficient ), it dont help ( enough ) to create chunks
( speculation ;)
looking (random?) at (30??) moves and judging their funktion takes to long especialy when you have to find a move backward for example; but when you seach for : remove one target and make an EGT the move pops up quick.
the set of tactical motives https://chesstempo.com/tactical-motifs.html is by far not complete, ther is only "unpinning" , undouple attack for example is missing, TitforTat and the refutations of TitforTat ( = more or less EGT and unEGT)...
the simple puzzles are : find the simple motiv
next harder puzzle : find the uncommon, not quite that simple motiv
the puzzles i get have a lot of "traps" , ihave to get more aware of the refutations of "ideas" and sometimes even the refuations of the refutations
well.. even if it might not make me stronger , its fun for me to expore the world of taktics this way
PART I:
ReplyDeleteAox said:
"I think it's too slow to only! look at a move and think: "what is the function of the move?"
"thinking (only) in moves is wrong ( not sufficient ), it don't help ( enough ) to create chunks
( speculation ;)"
I agree with you! An isolated move can NEVER be a "chunk" and will never be stored into LTM as a "chunk." However, a "chunk" CAN (and does) have associated moves. The classic Greek sacrifice on h7 (h2) springs readily to mind.
Funny: I spent a long time last night thinking about this position (again). I wanted to understand the possible reason(s) WHY I looked at 1. ... Rxf3 immediately, and (perhaps, speculate) WHY Aox looked immediately at 1. ... Qf5. Please keep in mind that the following speculations are post hoc rationalizations, but there is an experiential basis for it that may not be readily apparent.
I "think" I know WHY I looked at capturing first.
(1) The f3 square is B.A.D. (1:1), defended only by the White King. B.A.D. squares only defended by the King are especially vulnerable.
The "natural" (unforced) approach to conquering an occupied (yes, unoccupied squares can also be B.A.D.; see below) B.A.D. square is to either (1) capture on that square OR (2) add an attacker to that square OR (3) divert a defender away from that square. Option (3) is not possible. So, the choice reduces to either capturing or adding an attacker.
Adding an attacker (without first realizing that 1. … Qf5 is also a double attack on the LPDO White Rook on c8) seems like a good idea. I rejected it because of having worked very hard to hammer the “ENCIRCLING MOTIF” into my subconscious. This motif has two components: (1) An immobile “target” (prevented from moving away from the attack by either obstruction, domination or Function); and (2) Superior force ( a favorable attacking/defending ratio) against the target square. Condition (2) is met by 1. … Qf5; unfortunately, condition (1) is NOT met – the “bird” (Bf3) can “fly the coop”! Perhaps this is difficult to “see” in the initial position, because the BQb1 and the WBf3 make the (unoccupied) d1 square a B.A.D. square. Even if there is a retained image of the Black Queen on b1 preventing 2. Bd1, that does not prevent the White Bishop from taking up residence on e2.
Dr. Lasker’s “story” is great for remembering the “encircling motif” as a “chunk”:
“To name this motif, let us emphasize the TWO ideas underlying it: the idea of superior force at a given point and that of immobility. WHAT IS IMMOBILE MUST SUFFER VIOLENCE. The light-winged bird will easily escape the huge dragon, but the firmly rooted big tree must remain where it is and may have to give up its leaves, fruit, perhaps even its life. Let us NAME IT the motif of ENCIRCLING, since in this term the TWO ideas of VIOLENCE and of IMMOBILITY are blended.”
PART II:
ReplyDelete(2) Capturing is at a higher "forcing" level [(1) Checks; (2) Captures; (3) Threats] than threats (unless the threat is checkmate). This did NOT play any conscious part of my thinking when first trying to solve the problem! So, it can be eliminated as "justification" for choosing to capture first.
(3) I “think” the real reason I looked at 1. … Rxf3 first is a “chunk” that I worked very hard on: “seeing” various ways to “box” in the enemy King. After a lot of study, I realized that a Queen can checkmate with a little “help” from ONE “friendly” piece/Pawn and TWO enemy pieces/Pawns. The “trigger” I conceived was the pattern of the two enemy pieces/Pawns diagonally placed with the enemy King on the corresponding diagonal. In the problem, this configuration is met by the two Pawns (f2 and g3) and the enemy King at g2. If the attacker’s Queen can reach a square that is a Knight’s move from BOTH of the enemy Pawns (with the enemy King in between), then all that is needed for checkmate is a friendly piece to protect the Queen – the Black Knight “protects” e4. The Black Queen can reach either h1 (which is unprotected, so is immediately eliminated from further consideration) or e4 in one move. Those conditions are all met in the initial problem. Without consciously thinking it through, I “saw” that 1. … Rxf3 set up the conditions for a checkmate by 2. Kxf3 Qe4#. Any time there is a favorable resolution in the first (most forcing) variation, that is a good sign that the move(s) might work in all variations.
The next step was to apply the Popperian concept of “falsification” to “see” if there is a logical refutation available. That’s when I worked out the attempt to attack the Black King using the White Queen and Rook. Although it looks dangerous, it is rather easy to “see” that White cannot succeed without allowing Black a tempo to defend. It was during this process that I “saw” that White cannot avoid losing a piece. There is no checkmate possibility for either player.
PART III:
ReplyDeleteHaving satisfied myself that 1. … Rxf3 was winning, I “rested on my laurels” (as usual – this is BAD!) and failed to follow Dr. Lasker’s advice: “When you ‘see’ a good move, look for a better one.” I SHOULD have looked at 1. … Qf5, which constitutes a double attack against Bf3 (B.A.D.) and Rc8 (unprotected – LPDO). This move makes sense from a tactical theme/device viewpoint [NOTE: N-O-T necessarily from a Laskerian MOTIF viewpoint!]: f3 and c8 are both unprotected AND neither piece has a move which can protect the other.
I speculate that failing to “see” the “refutation” (1. … Qf5 2. Bd1) MAY be bound in some way to a limitation in how we understand the concept double attack. When we think “attack,” we often fail to consider ALL DEFENSIVE alternatives against such an attack. We do NOT (usually) think of a “double attack” when the move (1) REMOVES an attacked piece from attack, combined with (2) a COUNTERATTACK on an enemy piece. Yet, according to GM Averbakh, in the most general sense, both situations are “double attacks.”
On reflection, I don’t think there was any failure to “see” an EQUAL or GREATER THREAT. I’m not even sure that concept should be a consideration, and certainly not the central one.
That first move 1. … Qf5 sets up a superior attacking/defending ratio on f3 (1:2) for Black AND forks two under-defended pieces. That’s usually a good sign: a single move that performs TWO functions is usually a good move. Unfortunately, as previously noted, the “target” Bf3 can simply move away AND thus balance the attacking/defending ratio on f3 (2:2). Worse yet (for Black), it can counterattack a target (Rb3) which has equal weight to the attack on Rc8. I would “hope” (I cannot be certain, since it didn’t happen) that if my attention was caught by the fork of f3 and c8, when I began the falsification process, I would have thought, “Gee, this doesn’t seem to be working out quite like I hoped” and started looking at other possibilities. The advice to “GO WIDE BEFORE YOU GO DEEP” is another one of the aphorisms that I try to be aware of while analyzing. Or, in the vernacular of this blog, “Try to get a COMPLETE overall vulture’s eye view before you dive into a specific tunnel.” Often times, that “vulture’s eye view” makes the subsequent calculations easier to keep in mind because there is a “scenario” (or template or framework or “story” or, perhaps, “chunk”) which causes everything to “make sense.”
PART IV:
ReplyDeleteI think the little "chunks" (like one based on various "cues" to trigger the "box" recognition pattern) constitute what has been referred to as "miniskills". There has to be something that acts as a "trigger" to fire the pattern recognition mechanism (System 1) and a scaffolding/structure on which to hang the various possible responses. A recognition "cascade" occurs when there are sufficient "cues" and resources available to apply the pattern (or at least to apply a similar pattern). Where have we "seen" this structure before? Oh yeah: the Tree of Scenarios! Never let a good idea go to waste!
Instead of trying to memorize specific positions or moves, I think we can drive "cues" into LTM by using an apt phrase or "story" to encapsulate little bits and pieces of chess SKILL (NOT necessarily chess lore) (i.e., the encircling motif). The memory structure may be a "tree" or a "room" or a "list"; the actual structure form is irrelevant. We can't force the System 1 associative mechanism to "throw up" the exact idea needed in every situation, but we can "prime" it to get us in the ballpark with one or more reasonable alternatives. Pattern recognition (usually) only gets us an approximation of what needs to be "seen" in a concrete position. We then have to do the "hard work" of exploring the alternatives, evaluating them relative to each other, and then choosing the one we "think" will lead to the best position.
I think 1. … Rxf3 and 1. … Qf5 are both ok to start with
ReplyDeleteits just necessary to find the refutation of the wrong move quick.
Laskers phrase : "When you ‘see’ a good move, look for a better one" is only valid if there is a lot of time at the watch left for that. In Blitz-mode thats not a good idea
Here we have to be sharp to the point.
Later , in the aftermath, a wide and deep analysis is highly beneficial