REDUX Developing a sense for the initiative
Originally posted at Dec 13, 2017. REDux added in RED.
Now the ToS is simplified, it is time to get a sense for this way of looking to the position. To describe it in those terms.
r3kb1r/2qn1pp1/p4n2/4pNB1/1p3P2/6N1/bPPQ3P/2KR1B1R b kq - 0 1
[solution]
In terms of the tree of scenarios:
Diagnosis:
I devoted a lot of time in the past with describing the advantages of the flight of the vulture. But is it really so beneficial as I have always advocated? Practice has shown no significant advantageous result. Since I have become allergic for everything that causes the overwhelming of my mind, we maybe must reconsider this. The initiative offers enough handles to prune the branches of the tree of scenarios. Do you really need to know that there is a rook at h1 when you are mating the king? I noticed in practice that I can do often very well without knowing where every piece is. Only when I get stuck it might be necessary to look any further.
I already concocted the idea of looking at the position in the following order:
REDUX II
In order to prevent my mind from being overwhelmed, I must get rid of all those pesky lists. No list of targets, but just one target at the time. The most logical target is the white king. I focus on the king.
What is the most logical point of pressure? Without further thinking, c2 pops to mind. So I focus on c2. No list of points of pressure, just one point of pressure at the time.
What do I want to know about a point of pressure?:
Bf8 can attack the king in two moves 1. ... Bc5 2. ... Be3
But again, e3 is not on the brink of being outnumbered.
Next point of pressure is b3. 1. ... Nc5 2. ... Nb3+
This would put you on the right track.
Notice how many branches are pruned with no risk of pruning too much.
Plan: add attacker to b3
Cashing in: it becomes apparent that you not only have to look for counter attacks when you want to cash in, but at any moment that you don't make a double function move. When you potentially give the initiative away. Due to the multiplex immobility you can permit a move that hasn't a double function. That isn't a two headed monster. (1. ... Nc5)
Now the ToS is simplified, it is time to get a sense for this way of looking to the position. To describe it in those terms.
 |
| Black to move |
r3kb1r/2qn1pp1/p4n2/4pNB1/1p3P2/6N1/bPPQ3P/2KR1B1R b kq - 0 1
[solution]
In terms of the tree of scenarios:
Diagnosis:
- Target: The king and queen look juicy
- point of pressure: c2; b3
- line of attack: c-file
- function: c2 defends king and b3 (overworked)
- immobility: King has Lack of Space; c2 is pinned; K and Q are at a knightforks distance
I devoted a lot of time in the past with describing the advantages of the flight of the vulture. But is it really so beneficial as I have always advocated? Practice has shown no significant advantageous result. Since I have become allergic for everything that causes the overwhelming of my mind, we maybe must reconsider this. The initiative offers enough handles to prune the branches of the tree of scenarios. Do you really need to know that there is a rook at h1 when you are mating the king? I noticed in practice that I can do often very well without knowing where every piece is. Only when I get stuck it might be necessary to look any further.
I already concocted the idea of looking at the position in the following order:
- targets
- points of pressure
- lines of attack
- functions
- King
- Queen
- Promotion
- Loose piece
- Rook
- Minor piece
REDUX II
In order to prevent my mind from being overwhelmed, I must get rid of all those pesky lists. No list of targets, but just one target at the time. The most logical target is the white king. I focus on the king.
What is the most logical point of pressure? Without further thinking, c2 pops to mind. So I focus on c2. No list of points of pressure, just one point of pressure at the time.
What do I want to know about a point of pressure?:
- it must be in contact with the target
- the target must be immobile OR
- the point of pressure must be in contact with two targets
- the attacker must be able to occupy it in 2 moves at maximum
- it must be on the brink of being outnumbered
Bf8 can attack the king in two moves 1. ... Bc5 2. ... Be3
But again, e3 is not on the brink of being outnumbered.
Next point of pressure is b3. 1. ... Nc5 2. ... Nb3+
This would put you on the right track.
Notice how many branches are pruned with no risk of pruning too much.
Plan: add attacker to b3
Cashing in: it becomes apparent that you not only have to look for counter attacks when you want to cash in, but at any moment that you don't make a double function move. When you potentially give the initiative away. Due to the multiplex immobility you can permit a move that hasn't a double function. That isn't a two headed monster. (1. ... Nc5)
ReplyDeleteIts one thing to describe a solution, an other thing to find it. I think it might be more important to find the reason why this puzzle is so hard:
Some possible reasons are
Target: Rh1 ( and Kc1 ?)
line of attack: diagonal h1-a8 and then eventually a-file for the queen with chackmate at the backrank
and
you dont want to move the Nd7 because it looks like its dangerous..
and
Nc5 is a silent move but the position seems to need a stronger move because of black possible Nd6+
...
Its necessary to get ridd of such wrong ideas quick to be able to focus on the right ideas
"Its one thing to describe a solution, an other thing to find it."
DeleteUsually I try to describe how the find it. I want to do it different this time. In stead of to put my already overly trained slow logic brain at work and make it bulkier, I try to educate my unconscious mind by just showing it the essence.
It looks like there are three possibilities on the ToS that might apply.
ReplyDelete(1) diagnosis -> immobility -> Lack of space
(2) plan -> lack of space -> Squeeze the box
(3) methods -> exploit immobility
The "cue" (for me) is the immobility of the White King - it has NO legal moves, i.e., it is already in the "box." That immediately brings to mind - how to attack the White King as fast as possible. The Black Queen is already engaged in pinning the c2-Pawn.
There are two available Black pieces for adding additional attackers to the "box." The Black Rook (a8) can add to the pressure on c2, but that still does nothing to attack the White King with a check. That leaves the Black Knight at d7. It just "happens" that there is a juicy square (b3) from which the Black Knight CAN give check. So, 1. ... Nc5 rapidly comes into focus as a very significant second-order threat.
What can White do in the interim? Nothing, really. It will take at least one move to open up an escape from the "box." Intermezzo moves (such as 2. Nd6+ Bxd6) do not change the "box"; it is still locked shut around the White King. Because of the BBa2, capturing the b4 Pawn (2. Qxb4) does nothing to open the "box"; Black can still play 2. ... Nb3+ (reinstituting the pin by the Black Queen AND zapping the White King inside the "box." White has no choice but 3. Qxb3 Bxb3, losing the Queen for a Knight.
So, it looks like scenario (2) is the operative one. After considering 1. ... Nc5, it begins to be apparent that there is a fork on the White King and White Queen on b3. Although the White Queen CAN move (somewhere), that Queen move will not relieve the "box" problem: the White King still has NO moves after the Black Knight lands on b3 with check, simultaneously closing off the "escape square" (d2) opened by moving the White Queen.
I didn't even consider trying to "attack" the WRh1 along the a8-h1 diagonal (with, perhaps, 1. ... Qc6). Why not? Because it loses the initiative to 2. Bg2, skewering the Black Queen and Black Rook. Lack of King mobility is a much more important "signal" than attacking an isolated unprotected piece which can be protected (or moved) without adverse impact.
When there is an immobile King with the "box" already formed around it AND there are at least two pieces "attacking" the "box", my approach is to look very long and hard at how to bring more pressure to bear against that King. Whatever piece(s) can be brought to bear must be thrown at it.
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 selected is of any avail." - E. A. Znosko-Borovsky, The Middle Game in Chess, pg. 60.
"ALWAYS CHECK: IT MIGHT BE MATE!"
(I love pithy sayings that have double meanings!)
I experienced the fact that the Q and K are on a knightsfork distance as a form of immobility. 1. ... Nc5 was a move without tempo. But if it had been, there was no way to prevent the fork without the loss of wood.
ReplyDeleteI'll explain how I looked at this problem.
ReplyDeleteMinute 1: Seriously, another tactics problem? (Not you, my sentiment at the moment about every one seemingly suggesting that everything in chess is a tactics problem, and now here is a difficult one to solve).
Minute 2-3: 1...b3, 2.c3 Qxc3, 3.bxc3 Bh3+, 4.Kb2 Rc1. Oh, wait, 3.Qxc3 prunes that branch. This is perhaps the biggest problem I have when calculating (and it just cost me my last OTB game) is applying the process of elimination. I will go unconstrained on some variation, only to figure out after much time spent that the whole line can be discounted at move one or two. It's as if I see long lines, and then maybe decide to blunder-check them - extremely haphazard methodology of analysis.
Minute 3: Well, this is a tactics problem, and Black is in favor. hmm, no real immediate threats against Black, so what is this free ...Nd7 doing? Nothing, supposed to be attacking since it's a tactics problem and Black has been given the green light. So, 1...Nc5 and yes unstoppable fork, and 2.Qxb4 fails as well.
Actually, I solved this problem in under three minutes - which would be a long time in a blitz game, and it's not saying much since in a real game one could be much more mentally tired OTB at this stage of the game, and there is no green light to terminate your opponent's position, only the instinct for survival at times.
Hi Tempo,
ReplyDeleteBack from my chess.com game bingeing and learning the Benko gambit. In looking at the problem and it solution ; it occurred to me the answer could also be arrived at by asking the principled question. How can I bring one more piece into the attack ? Cheers, Jim Takchess.
I would NEVER consider Nc5 as a candidate move!
ReplyDelete1. My king is not safe - and moving N opens it much more (Bb5+ comes to mind).
2. After Nd7 moves Bg5xf6 ruins my pawn structure (around the King).
3. Nc5 covers the line of attack by my Queen (very familiar like playing Qa5).
4. It does not capture anything (nor gives check).
5. It "does not" threaten anything (really?!)
The only move I would consider:
a) b3 - directly attacks the pawn around the King (b3-c2 is a one move capture)
b) Rc8 - to bring another piece into horizontal attack (Q+R)
To sum up: this kind of task is impossible for me to solve. I would rather play Bc5! to develop another piece and have an additional free square to my King! Is there any hope for me?
Tomasz:
DeleteI think your No. 5 is incorrect. Nc5 creates TWO threats: checkmate AND fork of the King and Queen. White can only prevent one of these threats from being realized by playing Qe3, losing the Queen for the Knight after ...Nb3+. In essence, it is a "first order" THREAT preceding a direct attack. No. 3 also overlooks that the Knight is only temporarily blocking the action of the Black Queen. On the next move, that line of attack against the White King will be cleared.
I guess it boils down to which mindset you play with: first, defensively, and only when everything is "safe" do you look at offensive - or vice versa. One of the things I have been trying to do for a long time is to NOT consider how "unsafe" my side of the board is until after I "see" what I can attack.
The answer to your last question: YES, there is hope for you - and for all the rest of us!
I added REDUX II to the post.
ReplyDeleteI have been trying to post at your blog for 4-5 months already, but NONE of my messages are accepted (published) in your blog.
ReplyDeleteDid you add me to the black (ignore) list my friend? :(
No, my friend. Blogger just stopped notifying me of new comments! I will dive in it tomorrow to find out the why.
DeleteYour proposed approach (looking at ONE Point of Pressure at a time, depending on what captures your attention first) has two advocates that I know about: GM Jonathon Tisdall [Improve Your Chess NOW] and GM Valeri Beim [How to Calculate Chess Tactics – A revealing look at the nuts and bolts of chess thought]. Both GMs develop an approach to analyzing positions which is in contra-distinction to GM Alexander Kotov’s approach to calculating variations [Think Like A Grandmaster].
ReplyDeleteGM Beim summarizes GM Kotov’s proposed method (pg 64) as follows (emphasis added):
1) In beginning our calculations, we must first of all LIST ALL OF THE POSSIBLE MOVES IN THE POSITION – the ‘candidate moves’ – so as to ensure that we do not look over some important possibility.
2) Having done this, we then CALCULATE EACH VARIATION IN TURN. The ORDER in which we do this depends on the character of the player and the characteristics of the position. Every player has his own way of doing this. One prefers to start with the most difficult lines, and only then turn to the easier ones, while another player prefers the opposite.
3) ALL OF THE POSSIBLE LINES CAN BE PICTURED AS A ‘TREE OF VARIATIONS’.
4) The main rule in calculating is that the player must train himself during a game to GO OVER EACH BRANCH OF THE TREE ONLY ONCE and must not be tempted to return to lines he has already looked at.
GM Beim uses most of his book thoroughly (and respectfully) refuting GM Kotov’s method altogether.
GM Tisdall relates a relevant question asked by GM Antony Lein:
“Do you think like a tree?”
He then proposes to calculate using ‘variation processing’ which is exactly like the process you have described. Pick what seems to be the most important variation and then calculate along it as far as needed to see if it works or not. If it does not, then pick the next most important variation and repeat the process. (Notice that JUDGEMENT must be developed in order to decide which variations are ‘most important’.)
system II has as specialty that it can walk though long tunnels in a sequential way without being distracted. If you lay all the branches of the tree of analysis behind each other you get in fact one long track that is ideally suited for system II to follow.
DeleteSystem II goes haywire when it must keep track too of the administration of the branches of the tree of analysis. System II cannot multitask. If it tries to multitask, it fills the free slots of the Short Term Memory and stalls.
To lay all the branches after each other and make system II think that it is one continuous track, we must develop a method. That is what I'm after now. The hope is that when system II is happily single tasking, system I gets the free hand to work its miracles.
“Do you think like a tree?” It often feels that way.
DeletePerhaps that's why we are called "wood pushers"? LOL!
DeleteShould read "wooden"
ReplyDelete