Thursday, July 21, 2016

Rules

This is the second or the third time I encountered this position. I devoted quite some time to it in the past. I might even have posted about it, I'm not quite sure. If so, I apologize that I post about it again. My excuse is, that I have apparently learned nothing from it, since I got the solution wrong. Again.
I think it is an important position, and I think it should be possible to see the solution at once. Maybe I can devise a rule or two during the process of learning how to see the solution.

 White to move
r1bq1rk1/pp2ppbp/2np2p1/8/2PNP1n1/2N1B3/PP2BPPP/R2Q1RK1 w - - 1 1
solution

The last move of black was 1. .. Ng4. The black knight is hanging. The problem is, that if white takes it, the white queen becomes overworked. If white wants to stay ahead, he needs an extra tempo. Where does this extra tempo come from? After 1. ... Ng4 2.Bxg4 Bxg4 the white queen is hanging. If the queen takes on g4, black will get his piece back by capturing the knight on d4. White can gain a tempo by 3.Nxc6 in stead of 3.Qxg4. This power move accomplishes the following:
• White gets rid of his problem piece
• It captures a piece
• The black queen is under attack
Rule: use your problem piece to capture
Rule: choose the capture that poses a new threat

This might give you the impression that you can change the move order. Why doesn't that work?
1. ... Ng4 2. Nxc6 is answered by blacks power move 2. ... Nxe3
Now it is black who abides the two rules above:
• He gets rid of his problem piece
• He captures a piece
• He threatens the white queen
Rule: capture the piece of your opponent that can capture a piece of you with tempo first

If black takes the knight first 1. ... Ng4 2.Bxg4 Nxd4 then the rule "use your problem piece to capture" applies. 3.Bxc8 gets rid of your problem piece.

1. I did a search for this position using Fritz 11's game database and found 21 games(!) with this position. The highest rated game was 470701: Ralf Akesson (2515) vs Mark Heidenfeld (2335). So if you found this difficult to "see," consider how it must feel to be rated 2335 and fall into this position!

470701: Akesson,R - Heidenfeld,M, EU-chT (Men) 1997
rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1

1. Nf3 Nf6 2. c4 c5 3. Nc3 Nc6 4. d4 cxd4 5. Nxd4 g6 6. e4 d6 7. Be2 Bg7 8. Be3 0-0 9. 0-0 Ng4 10. Bxg4 Bxg4 11. Nxc6 Bxd1 12. Nxd8 Bxc3 13. Rfxd1 Resigns

In my initial appraisal of the position, I thought that it had arisen from a Maroczy Bind against the Sicilian Dragon. I hadn't considered that it was reached via transposition from an English opening.

This type of position is discussed by Yuri Averbakh, Chess Tactics for Advanced Players under the section The mutual two-fold attack, positions 83-92. Perhaps I am merely confused, but I consider Averbakh's "two-fold" attack to be essentially the same as your "duplo" attacks.

I excerpt the following observations by GM Averbakh:

This extremely tense situation in which two pieces on each side attack two hostile pieces deserves special consideration.

. . .

Mutual two-fold attacks require precise calculation and circumspection, for they can easily boomerang if not handled with due care.

I just got home (late) from work and am too tired to discuss the given position in depth (at 1:30 AM). I'll try to give my personal thoughts (perhaps illusions?!?) later, when I'm not tired and I have the time to think it through. My granddaughter is visiting until Sunday afternoon, so it might take a few days; Papa and Rizzy have lots of "play time" scheduled!

Suffice it to say that I used a kind of counting, although counting different things than we have been discussing so far. Obviously, counting does not address everything in this type of position but it does give some rapid visual clues of where and what to look at. Sometimes that's all the hints that are available - or needed - to spark the correct train of thought.

I like the summation of your thoughts in your proposed "rules." My only caveat is simple: if we try to capture the essence in rules, will we eventually create a large set of rules that will overwhelm us? If the "rules" approach works, then (obviously) we need to grind them into LTM for them to become subconscious/intuitive. To me, it is the same problem that occurs with comprehensive and detailed thinking processes, like Chuzakin's System. (No offense to anyone, especially Chuzakin.) I'm still trying to just "see" what should be "obvious" without having to load up STM and perform lots of logical thinking/calculation.

1. The two-fold attack of GM Averbakh means that there are two fronts where pieces are in contact with each other. A duplo attack is a double attack, a discovered attack, a skewer, a pin or a simultaneous attack (attacking two targets with one move). These play no role in this position.

The main tactical theme of this position is very simple: black did an incorrect sacrifice with Ng4. You have to find a way to scoop up the hanging knight while finding the right defense for your own pieces. Seemingly you are a tempo short. So you must find an extra tempo in all lines.

I tag the posts where I write about rules, so I can find them back easy. In the end, I expect to have 10-20 rules or so. Then it is necessary to find the common factor in these rules, so I can bring them back to 2-3 rules. Something like "look for captures with gain of tempo" or something like that.

Automating the rules is the ultimate goal. It helps to unload the STM. You have to calculate every line in any case, in order to see if you have an exception of the rules. But calculation with a STM that is not already on the brink of being overloaded is usually no problem.

All the rules I find are related to maintaining the initiative. About the moves with an extra tempo. The main goal is to develop a sense for such moves. Consider the rules to be a pair of training wheels. In an ideal case, you only need those temporarily.

2. (Rizzy is down for a nap, so Papa may get to make a short reply. . .)

In the section The mutual two-fold attack, the first example (diagram 83) is taken from the game Simagin - Zagorianski, Ivanovo 1944 (which I have been unable to find in any of MY databases).

[FEN: 2r1kr2/p3n1R1/1p1p4/2p1p1p1/2P1B3/2PP1P2/P5K1/7R b - - 0 1]

The preliminary moves are 1. ... Rf7 2. Rh8+ Kd7!. Averbakh's comment: "This has brought about a mutual two-fold attack (diagram 84)."

[FEN: 2r4R/p2knrR1/1p1p4/2p1p1p1/2P1B3/2PP1P2/P5K1/8 w - - 0 1]

Is this position representative of the "tit for tat" idea, as well as sufficiently similar to the position given in this post in terms of applying these "rules" to it? An inquisitive mind would like to know. . .

[FWIW (not much), I am somewhat confused regarding the differentiation between two-fold attacks and duplo attacks, in light of Averbakh's discussion of forks, pins, discovered check and discovered attacks with check in the section on two-fold attacks, prior to getting into the discussion of the generalized double attack (which includes direct attacks, first order threats and second-order threats in all possible combinations). Therefore it is not surprising to me that I have failed to progress to an advanced level, since I obviously have not grasped the elementary concepts; I'm working on that!]

I still would like to discuss what I "saw" in the original position, but that will have to wait: Rizzy is now up and about!

3. [FEN: 2r4R/p2knrR1/1p1p4/2p1p1p1/2P1B3/2PP1P2/P5K1/8 w - - 0 1] is a Tit for Tat
If white does take the hanging Rf7 then black will take the hangin Rh8

THE standard method would be to move the Rh8 and gain a tempo,.. which dont work.. at this moment;)
But with the sac Bc6+, Kxc6 its possible to solve the "remaining" Tit for Tat this way
After Bc6+ and Nxc6 there is a double tit for tat with both rookpairs in contact and this can be solved with the standard method removing of the rook at g8 with a gain of tempo: Rxf7+
After Bc6+ : Rxc6 is a tit for tat RxBc6 vs RxRf7
.... and so on ...

In my eyes there are Situation given by the configuration of the weaknesses
Here they are : all rooks, the black knight and king and the pawns a2,a7,c3 and g5
But to have such a situation is not enough, we need to have access to a method to make use of this situtation

We need both.. the weakness and the method to make use of the weakness

For example a "weak backrank" is no weakness if there are no rooks or queens ;)

But as more weaknesses as bigger the chances ....

4. @Robert

[FWIW (not much), I am somewhat confused regarding the differentiation between two-fold attacks and duplo attacks, in light of Averbakh's discussion of forks, pins, discovered check and discovered attacks with check in the section on two-fold attacks, prior to getting into the discussion of the generalized double attack (which includes direct attacks, first order threats and second-order threats in all possible combinations). Therefore it is not surprising to me that I have failed to progress to an advanced level, since I obviously have not grasped the elementary concepts; I'm working on that!]

Two-fold: It means when you reached some position and after that you increase the complexity of attack. It is used to stress out (underline) the necessity to find another resources to make the attack yet more powerful (and efficient as well).

Duplo Attacks ("Double attack" in an original sense): it means the attack on two (or more) pieces. The simplest type is of course fork by Knight or pawn.

There is a term "double threat" as well. It describes the attack on two targets. The simplest example is the attack on the (undefended) piece and a threat to checkmate the King.

I am not surpised you are confused by Averkabh terminology. My feel is he probably tried to hard to make it scientific and forgot people want to have it (and use it) in a practical way! I am not saying Averbakh is wrong, but I am sure his theory may be much better explained - using simplest and less confusing meaning (and terminology).

BTW. It is just my explanation. It does not mean I am correct nor I understood his theory!

5. @ Tomasz:

Thank you very much for that explanation! I have had Averbakh's book for some time. Unfortunately, every time I have tried to grasp and apply his concepts, I get "wrapped around the axle" (so to speak), trying to figure out his terminology and how to map it to what I already know. It has only been through this series of discussions that I felt that I had progressed sufficiently to try again to understand his concepts. I still feel like most of his discussion is over my head.

6. About the 21 master games where black made this blunder Ng4, was it punished 21 times?

7. Yes, every time.

2. I did analyse this position several times before too. If i remember corect then i found more than 100 games in my gamedatabase and even CT does have this position at least 3x in its problemdatabase

http://chesstempo.com/chess-problems/79291
http://chesstempo.com/chess-problems/78038
http://chesstempo.com/chess-problems/77922

In my eyes its a "tit for tat" position: if i win ( straight foreward ) your knight then you win ( straight foreward ) my knight. This type of position is very common and are usually the result of a "defence by counterattack".

The ( a ) Standard - Method in this type of situation is to remove our weak piece with gaining of a tempo, so Nxc6 is the first move to think about. But in this case Nxc6 dont gain a tempo because Nxe3 is creating a counterattack to the white Queen and we reach a new Tit for Tat.

I think this example is realy a position where you have to calculate!

Nevertheless.. "Rules" help us to look for the important things (first). I would call "rules" in this case "Standard Methods" for a "Situation" given by a constalation of "tactical weaknesses".

Tempos rules :
Rule: use your problem piece to capture
Rule: choose the capture that poses a new threat
Rule: capture the piece of your opponent that can capture a piece of you with tempo first

are good rules for any situation they solve 2 problems with one move
and by the way: Tempos 3 rules are Tempo rules ;))

3. In looking at the given position (for this particular blog post), I started with the original position (before 1. … Ng4. In that position, I see that White has a lead in development and more space in the center. Additionally, Black’s usual counter-strike on the queenside is stymied for the foreseeable future. In general, White has the advantage and there appears to be nothing “hanging” tactically. Ergo, Black has no business trying to sharply ratchet up the tactical complexities. So, the given first move for Black must be a mistake.

My first “counting” involved the square d4 (+2/-2). The second “counting” involved g4 (+2/-1). The third “counting” involved the square c6 (+1/-1). So, the “solution” must be related to those squares, with special interest on the g4 square because of the imbalance in White's favor. I realized immediately that if White captures twice on g4 (2. Bxg4 Bxg4 3. Qxg4) that Black could then capture Nd4, regaining his piece (with even material) and considerably more “breathing room” than before. I guess this is what attracted the attention of several master players to this mistake.

A quick check about capturing on g4: after 2. Bxg4 Bxg4 White can simply play 3. f3, consolidating the white-squared Pawn chain and forcing the Black Bishop to retreat. This might be satisfactory but “when you see a good move, look for a better move.” I wasn't thrilled to give the two Bishops to Black with a static Pawn chain in the center of the board, so I quickly accepted that the first White move must be 2. Bxg4.

This is where my “thinking” got "intuitive" (and murky). I considered (briefly) 2. Nxc6 and rejected it on this basis: the potential “chain” of tit for tat moves ends for White after 3 captures (2. Nxc6; 3. Nxd8; 4. Nxb7). Unfortunately, the potential “chain” for Black goes (2. … Nxe3; 3. … Nxd1; 4. … Nxc3; 5. … Nxe2+). This proves NOTHING in terms of what calculations must be made, but it gave me an uneasy “feeling” that 2. Nxc6 would not end well. So, I continued with 2. Bxg4 Bxg4 3. Nxc6, apparently leaving the White Queen in the lurch but counter-attacking the Black Queen. Since White is a piece ahead at this point, this "feels" right.

Now Black can capture the Nc6, but then lose the Bg4, ending up a piece behind. So, “in for a penny, in for a pound”: 3. Bxd1 4. Nxd8. White is a piece ahead and Black has no more “tit for tat” targets. Capturing the Nd8 allows capture of Bd1. Escaping with the Bd1 to c2 allows the escape of the Nd8 via b7.

Black can try for the e4 and b2 Pawns as compensation (Bg7xc3, followed by Bxe4 and eventually Rb8) but will remain a piece down. That should be sufficient for a White win.

That process took very little time to work through. It was aided by "counting" various things. "Counting" isn't everything in this position, but it sure does help in arriving quickly at an intuition of the proper direction to "look."

4. Robert said:
"In looking at the given position (for this particular blog post), I started with the original position (before 1. … Ng4. In that position, I see that White has a lead in development and more space in the center. Additionally, Black’s usual counter-strike on the queenside is stymied for the foreseeable future. In general, White has the advantage and there appears to be nothing “hanging” tactically. Ergo, Black has no business trying to sharply ratchet up the tactical complexities. So, the given first move for Black must be a mistake."

At the position before ...Ng4 Black is to move, so if you say "Black would be behind in development" then you have to say "Black is up a tempo" too.

Theory's main line ( played by 1222 Masters with av Elo=2688 ) is here Bd7 after which Black is not behind in development.

Ng4 is a standard move in this type of positions and white often playes f3 against it. White has more space so black should exchange some minor pieces ( 2 pairs is usually enough ) so Ng4 seems to solve all problems of black. The evaluation of Komodo 10.1 at depth 22 is +0.22 which is common in the opening,so black is not worse.

Robert said:
My first “counting” involved the square d4 (+2/-2). The second “counting” involved g4 (+2/-1). The third “counting” involved the square c6 (+1/-1). So, the “solution” must be related to those squares, with special interest on the g4 square because of the imbalance in White's favor. I realized immediately that if White captures twice on g4 (2. Bxg4 Bxg4 3. Qxg4) that Black could then capture Nd4, regaining his piece (with even material) and considerably more “breathing room” than before. I guess this is what attracted the attention of several master players to this mistake.

That is to short in my eyes, there is more to see:
Blacks Queen is week because she can be attacked by a minor piece in one! move ( Nxc6,Ne6 )
Whites Queen is week because she can be attacked by a minor piece in one! move ( Ncx3,Nxf2)
...
( for more details see f.e. chuzakins system HE #5 )

I im shure that the first move to calculate is Nxc6 because it is ( on "first sight" ) the most forcing move
It is necessary to "prove" that it is not beneficial first

1. "I im shure that the first move to calculate is Nxc6 because it is ( on "first sight" ) the most forcing move"
I'm not so s(h)ure about that. A capture is gaining wood. It is a permanent advantage. If you postpone a capture by a threat, which is only a promise of a later capture, you must have a darn good reason for it. A threat is temporary by its very nature. If it doesn't culminate into a capture, the threat is useless. For me it is more logical to look at the captures first, and if that doesn't work, to look at threats in an attempt to gain a tempo and make it work.

2. Is this reasoning why CCT (Check, Capture, Threat) are given in this "forcing" order? Checks MUST be answered before any other consideration. Captures might be ignored (for some time) with a Zwischenzug. Threats may (or may not) have to be answered if there is a possibility of an EST (Equal or Stronger Threat).

In the given position, there are no checks, so the question resolves to "Analyze captures or threats first?" As I noted previously, my preference within a particular category revolves around the last piece moved. If that piece is making a strong threat, then it seems most natural (to ME) to determine if it can (or needs to) be captured. Only after considering THAT do I continue looking at other things which MAY have changed in the position (perhaps a piece somewhere else is now insufficiently protected, etc.).

5. This article has finally awaken my inner chess giant Tempo! :).

I am going to share my findings here and if time permits - on my blog (with pictures). This way you can be sure HOW MUCH influence your fantastic work gives me! I would be very glad if you confirm (or refute) which conclusions and descriptions (explanations) are the same as yours. I simply tried to stand on your shoulders and see it (explain it) with a new (higher) perspective. Hope you will like it! :)

BTW. When I finish I will share it here. And when I write an article at my blog I will let you know. To make a long story short I refreshed the specific idea ("mirror captures") I was working on about one year ago. Without any doubts your works gave me the necessary boost and inspiration (not to mention great insights!). Thanks again!

1. Do you have a blog?!

2. I remember now. It was in Polish.

3. I am not sure if we can call it a chess blog, but maybe a "amateur diary" would be much more appropiate. It is in Polish, but if you use Google translate (or similiar tools) you can probably understand everything. However take notice that my blog is recommended rather to beginner players because I do not publish anything that requires deeper thinking. Just my chess remarks based on some experience and limited knowledge. Link below:

http://beginnerchessimprovement.blogspot.com/

6. @ Aox:

From long (and bitter) experience, I have learned to always focus on the last move of the opponent FIRST to see ALL the things that move threatens (H.T. to Coach Dan Heisman through his books), before I consider anything else. That is the only reason I did not consider 2. Nxc6 first, because that move IS very attractive because of its forcing nature.

I'm sure that Chuzakin's System (and perhaps others as well) can be very beneficial if thoroughly understood and assimilated, and then rigorously applied. I wasn't trying to be very rigorous in my description; I was giving the general gist of what I "saw" as I looked at the position without actually trying to calculate move sequences. It's very hard being limited to a linear description (in words) of an essentially non-linear process which is visual and (to some degree) subconscious; hence, my reference to intuition and a "murky" process.

I'm certainly happy with the results of the processes that I have learned about on this blog. I can "see" things that before were totally obscure. Often times, all that is required to trigger the appropriate investigation is a "hint" of the right direction to "look." I also can "see" the value of the recent "rules" that have been derived and generalized from the examples given. I have learned that the insights of others can point me in a fruitful direction for study and work, but as Temposchlucker once said, regarding his coach, (paraphrasing) "I found that I have to do the work myself." Unfortunately, someone else's "system" doesn't work well for me unless I take the time to thoroughly investigate it and translate it into my own internal "language" through hard work.

1. We dont have to do the work by our self we may read books and follow advice.
Your comments are full of insights taken from books and you seem to have no problems with the system of Lasker, Averbakh and others . ( For example : “when you see a good move, look for a better move.” )
Seemingly the term "system" is a problem for you but Chuzakins system is like Nimzowitschs system a collection of insights . Lasker could have named his books "Laskers System" too.

If you realy want to do the work all by yourself: Throw away your chessbooks and stop reading this blog ;)

2. @ Aox:

Thanks for the encouragement! (Yes, I "got it" that you are joking!)

I certainly do NOT want to abandon any source of insight into improvement, especially not this blog!

I guess I was unclear (murky). What I meant to say (hopefully, better this time!) is that UNCRITICAL adoption of someone else's CONCLUSIONS (i.e., their "system" or "rules" or whatever) means that I haven't internalized that "knowledge" as my own; it is at best superficial knowledge and at worst, downright harmful. The consequence is little different from memorizing specific grandmaster variations in my favorite opening(s), only to find myself totally adrift, lost at sea, while blindly and furiously paddling just to keep the boat afloat.

Perhaps a martial arts analogy will be clearer. I have a relatively large collection of books on various martial arts. I have read and studied them all for many years. They provide interesting ideas of how to respond in various situations with various techniques. But I didn't start with those books and ideas from other people. I started on the dojo mat, and worked on my skills for over 17 years. It was only after grinding the fundamentals into my body (often paying for those lessons with considerable pain) that I began exploring other peoples ideas and suggestions. In doing so, I began to generalize my own "rules" from my experience, combined with other peoples ideas. Without that personal "sweat equity," I would not have the defensive skills that I possess.

As an example, in judo (and jujitsu), a lot of the techniques are concerned with throwing the opponent. After many years of studying, I came to a general "rule" about throwing: "Create a hole, fill the hole." Without personal experience of throwing and being thrown (over and over again), that "rule" will make no sense to anyone! In every throw, tori must create a "hole" (empty space) into which uke is thrown, thereby filling the "hole." Almost every throwing technique can be subsumed in that "rule." The application of the "rule" depends on the specific situation, but most importantly, the balance/imbalance between the two players. It is virtually impossible to throw someone without utilizing kuzushi (the imbalance of uke). Once you have the experience of kuzushi, it becomes simply a matter of positioning so as to allow "Mr. Gravity" to do most of the work. The "rule" can be extended beyond just throwing to include virtually all striking and kicking techniques as well. If you cannot create an opening in the opponent's defense (or trick him into creating such an opening), then you cannot successfully expect to attack.

All of those ideas can be applied to chess, as well as to a wide range of situations. In every case, the one who would gain "mastery" must do the hard work required to make those "rules" personally applicable.

7. Now my promised article (take notice it is a draft, but I hope some of you may find it beneficial).

There are a few parts and the first thing to see it the analysis. This way you can see what I analysed and the written mistakes and errors may be corrected (while using this file).

----------- ATTACHEMENT [pgn file for easier analysis] -------------

[FEN "r1bq1rk1/pp2ppbp/2np2p1/8/2PNP1n1/2N1B3/PP2BPPP/R2Q1RK1 w - - 0 1"]

1. Bxg4 (1. Nxc6 Nxe3 (1... Bxc3 2. Nxd8 Nxe3 3. fxe3 Bg7 (3... Be5 4. Nxf7 Rxf7
5. Rxf7 Kxf7) 4. Nxf7 Rxf7 5. Qd5 e6 6. Qxd6 Bf8 7. Qd8 Rxf1+) (1... bxc6 2.
Bxg4 Bxg4 3. Qxg4 Bxc3 4. bxc3) 2. Nxd8 (2. Nxe7+ Qxe7 3. fxe3) (2. fxe3 bxc6)
2... Nxd1 3. Rfxd1 Rxd8) 1... Bxg4 (1... Nxd4 2. Bxc8 Rxc8 3. Bxd4 Rxc4 4. Bxg7
Kxg7) (1... Bxd4 2. Bxc8 Bxc3 (2... Bxe3 3. Bxb7 Bxf2+ (3... Qb6 4. Bxa8 Rxa8 5.
Nd5 Qc5 6. Nxe3) (3... Rb8 4. Bxc6 Bh6) 4. Rxf2 Rc8 (4... Rb8 5. Bxc6) 5. Bxc8
Qxc8) 3. Bxb7 Bxb2 4. Bxc6 Bxa1 5. Bxa8 Qxa8 6. Qxa1) 2. Nxc6 (2. Qxg4 Bxd4 3.
Bxd4 Nxd4) 2... Bxd1 3. Nxd8 Bxc3 4. Raxd1 Rfxd8 5. bxc3

----------- ATTACHEMENT [pgn file for easier analysis] -------------

8. This article and postition presented - I can call these: the most demanding ones as we saw recently. I love the way we can exchange the ideas and discuss the problems.

I think this specific problem (and position that represents it) is clearly shown as even some very strong players missed it! If players rated up to 2500 may fall for that - it means something important is behind this problem, isn't it?

Let's see why 1.Bxg4 works (PART 1) and 1.Nxc6 does not (PART 2). What is the difference between these two options?

The problem piece and the especially the rule invented by Tempo is simply great. I cannot refute any of these due to the lack of weaknesses.

I try to push our great discussion and discovering process... a bit further.

What did I do? I simply created the diagrams as a set of pictures - one next to the other (3 columns and rows). This way I have the ideal condition to describe any of these without the problem of missing elements (like when you are working with a program and clicking back and forth).

(PART 1) - 1.Bxg4 and why it works.

PICTURE 1: White is a piece ahead and it is Black's move. He can take the Nd4 or Bg4. If Black takes the Nd4 with the Bg7 Black does the same - he takes Bc8 with his Bg4. And you can check out by yourself - if Bg7xd4xc3xb2 or Bg7xd4xe3xf2+ is a better choice. Take notice White is a piece ahead and after Bxd4 Bxc8 the essence of the position remains the same: White is still a piece ahead! And the first conlcusiion? Unless one side is able to create stronger threat (capture the piece of higher value or checkmate the King) the mutual exchanges (with the same pieces value) DOES NOT change the essence of the position. It can only create so called "positional weaknesses". That's why you just have to check out if after one capture of your own piece you can answer with the same: if you start with equal (or better) position - the essence remains... as long as you answer back with the same degree of damage (here: if you can recapture {at least} the same value piece).

9. What about interruption? You have to take these into account as they change the whole picture (sometimes quite much).

1.BxN BxN 2.BxB and Black has three choices:
a) 2... QxB 3.BxB NxB 4.QxN and white is ahead

b) 2... BxN and now look at the difference - Black took the piece back and the position is equal! If White plays pxBc3 Black answers with Rxc8 and there is no advantage! Therefore White has no choice - he breaks the threats equality - he attacks both pieces (Ra8 and Nc6) and wins. Why? Simply because Nc6 is not defended! Now after:
3.Bxc6 there are two pieces attacked and one of these is handing (Nc6 has no protection!).

Now let's look at more stuff! 3.Bxb7! and 3...Bg7 4.Nxc6 wins a piece, but what about more complicated scenario? 3...Bxb2 - now both Rooks are hanging and they are protected. To see it better, just take off (remove) Nc6 and Be3 and you will see the position is still equal! (4.BxR BxR QxB QxB). However now White is winning due to the undefended Nc6. That's why White plays: 4.BxN! This way he wins a piece for free and the essence of the position remains the same! (Both Rooks are hanging, but White is a piece ahead). For example: 4...BxR 5.BxR QxB 6.QxB and White's got a piece more and Black hasn't.

Now ask yourself: are you able to look 12-plies ahead? If not, just try it and you may be surprised. Set up the original position and go with the variations in your mind:

1... Bxd4 2. Bxc8 Bxc3 3. Bxb7 Bxb2 4. Bxc6 Bxa1 5. Bxa8 Qxa8 6. Qxa1 - the first time it may be difficult, but try to visualise the images without the pieces after they are being captured. It should ease the process.

c) 2... Bxe3 and now both Bishops are hanging. If White simply takes Bishop (pxB), Black does the same (RxB) and the position is equal. It means White has to find something better. Do you remember the solution at previous variation (BxNc3). Now the same applies with small (difference) details. Let's see.
3. Bxb7! and now if Black escape with Be3, White simply takes Nc6 for free. Black has some interesting alternatives, but White is ahead in every line. For example: Qb6, Bxf2+ and Bd4.

Just have a look at Bxf2+ as when Black plays any other line White simply takes the Nc6 for nothing. It taking with check any better? Unfortunately, not. Here is why: 3... Bxp+ 4.RxB and now White is a piece ahead and Nc6 is hanging as well. Black has two options: 4... Rc8 5.BxR QxB and White is a whole Rook ahead or - 4... Rb8 5.BxN and White is two pieces down. Why was that happened? Because Black did not solve a problem with hanging Nc6, but gave up the Be3 for a mere pawn.

What about Qb6? Let's check it out: 3...Qb6. There is a nice small trap and White must be careful! For example: 4.Nd5? Bxf2+ 5.Rxf2 Qxb7 and Black exchanged the hanging Bishops and captured a pawn for free. What White should play? If 4.Bxc6?! then Black plays the same line as before (or even 4...QxB 5.pxp Qxp). Better is: 4.BxR! as Nc6 is protected. Now Black has to recapture lost Rook as Bishop threatens to escape with another capture (BxN!). For example: 4...RxB 5.Nd5! and a double attack at Q+B - it means White is a Rook ahead.

The conclusion: after 2.Bxe3 3.Bxb7 the best option is of course to play 3... Rb8 4.Bxc6 Bh6 and White is a piece ahead.

FINAL conclusion: (PART 1) - after 1.Bxg4 Bxd4 2.Bxc8 wins (due to the breaking threat's point [Bxb7 with double attack and Nc6 not defended].

1. Excellent analysis! 3.Bxb7 inspires me to formulate another "rule", btw: it is better to take an unprotected knight than a protected rook.

I find it easier to look at the lines for black and white separately. As if the other party makes null moves.

1.Bxg4 2.Bxc8 3.Bxb7 4.Bxc6 5.Bxa8 (N+B+p+N+R) for white and
1... Bxd4 2. ...Bxc3 3. ... Bxb2 4. ... Bxa1 (N+N+p+R) for black shows that blacks counter attack peters out before whites line of attack does. The extra tempo white needs, is provided by the fact that his line of attack is longer.

Black can opt out early by 4. ... Rc8, replacing a capture by a threat, while at the same time he saves his rook from the threat 5.Bxa8. A multi purpose move that generates an extra tempo! Luckily white can do the same with 5.Rb2, which is a duple tempo move too.

At first, I thought that the double attack 3.Bxb7 was crucial. Now I doubt that. The fact that the white bishop attacks Ra8 from c6 seems to be of more importance. It makes white's "line of attack" longer.

If I try to imagine both lines of attack at the same time, I get easily a STM overload error. Especially if I try to maintain the administration of the captured pieces at the same time.

10. (PART 2) - 1.Nxc6 and why it does not work.

PICTURE 2: White is a piece ahead and it is Black's move. He can take the Nc6, Nc3 or Ne3. At first let's refute the worst moves:
a) 1...BxN? 2.Nxd8 Nxe3 3.fxe3 Be5 4. Nxf7 Rxf7 5. Rxf7 Kxf7 and White has Q for B - a big advantage.

b) 1... bxc6 2. Bxg4 Bxg4 3. Qxg4 Bxc3 4. bxc3 and White is a piece ahead.

c) the last try is to reply with the same threat 1...Nxe3! Now both sides has the same pieces and the position is equal. Both sides can repeat the captures and this way the position will remain the same (in a terms of material wining/losing: we do not take gaining a pawn as a point of our discussion). Therefore: 2.NxQ NxQ 3.RxN RxN and neither side is better. We can make some other variations: 2.Nxp+ QxN 3.pxp (White gained just a pawn) or 2.pxN pxN - both sides are equal.

FINAL conclusion: (PART 2) - after 1. Nxc6 Nxe3 2. Nxd8 Nxd1 3. Rfxd1 Rxd8 White does not win a piece (due to unability to break threat's point). None of sides did not create a bigger threats to each other.

1. I admire your ability to make simple things look complex. I will have a closer look at it to see if I can simplify it again. ;)

2. I always admire to make simple things complex because it would not have been funny to have everything served on a golden plate ;) :). If you would know how difficult chess is for me you would probably forbid me to post any comments on your blog. But please do not tell it to Tempo! ;)

3. To be honest - I do not have the necessary stamina, perseverance and energy (not to mention the willpower!) to create a really good article like you do my friend. That's why I just share what's inside my mind. And take notice I have written this draft analysis within 2-3 hours. In addition I have a great ability (a skill?) to expand the things to the maximum. This way I can create some "hooks" that can refuted or look as an inspiration to others. Even if it is harmful, I am not going to change it. I hope none will be hurt badly ;) :).

11. @ Tomasz:

An impressive amount of calculation! I am rarely able to calculate 12 ply into the future with any accuracy, except in positions that have a very "skinny" tree of variations, i.e. where there are not a lot of alternatives (candidate moves) at each ply.

I am curious regarding your approach to a new position in which the solution is not recognized immediately.

Do you jump right into calculation as soon as you see the position, or is there some preliminary cogitation on the "vulture's eye view" which directs your subsequent calculations?

Are there any generalized "rules" that you apply while calculating that enable you to abandon a specific line without calculating all the way to quiescence?

I ask because I usually don't jump into calculation immediately, even when playing OTB. While analyzing, I also prune lines that do not accord with my (limited) knowledge, using intuition ("gut feeling") to guide my choice of variations to explore. (This should be obvious from my description of my thought processes.)

Perhaps this is ONE (of probably many) source(s) of my lack of and apparent inability to gain much higher playing skill(?).

12. @Robert

Thanks for your compliment! Of cource 12-ply is not that bad, but I want to stress out one "small" detail. I was talking about the SPECIFIC position with one single variation in a relatively very easy set up. That's why it is possible to see even up to 15-20 plies. My best achievement was to solve the task that required to see 28-ply! I think you can call it really impressive, but in reality it was a forced single line (an endgame puzzle). If there would be a lot of alternatives (candidate moves) at each ply - I could be able to see no deeper than 4-5 ply.

Now I want to answer your question: [I am curious regarding your approach to a new position in which the solution is not recognized immediately.

Do you jump right into calculation as soon as you see the position, or is there some preliminary cogitation on the "vulture's eye view" which directs your subsequent calculations?

Are there any generalized "rules" that you apply while calculating that enable you to abandon a specific line without calculating all the way to quiescence?]

1) I can recognize the position immediately or I have to solve it. If I can recall the position from my internal mind database - it requires only a few seconds to check the details. That's the reason I can solve some puzzles (2-3 movers) in about 3-5 seconds.

2) I do not know if I calculate first or look at the position closer. Probably I try to see what is the difference between my database and the position. If there is a small difference - I calculate the variation (most often it is a mating final or winning a piece).

3) Hard to say if there are any generalized "rules" that I can apply while calculating. Most often - a forced variation enables me to abandon a specific line without calculating all the way to quiescence. For example if I see the mating line - I just check if there are no blunders in my calculation. However when I calculate "a line that leads to quiescence" I try to use the rules related to the "when you capture must take the recapture first". The other one is to improve my pieces when attacking. If I have a few alternatives how to increase my attack - I try to develop more pieces to attack or squeeze the King with the pieces very close to the target. What is this unwritten rule? It is based on the concept of check. One of the ways of getting out of check is to COVER the line of attack. But there are two cases you cannot use this method of defence. The first is when there is no space between the pieces (direct contact) and the second - when you attack the piece (most often the King) with a Knight or a pawn. You simply cannot cover such an attack, can you?

As far as I can feel - I think I prune (calculate) the short variations (1-2 movers) very quickly. And If I cannot do this I try to evaluate the position and bring more pieces to the area of attack. In other words - I learn to develop as many pieces as possible and set these on the optimal squares - most often Rooks and Bishop on the open diagonals, Knights and pawns to be placed to the important squares (in case of pawns - promotion squares).

Anyway I discovered a simple rule: you cannot attack successfully unless (all of) your pieces are standing on active places! That's why I often try to make them active even at the cost of material deficit.

Did I answer your questions and concerns? Let me know because I am not sure if I caught your intention correctly.

13. @ Tomaasz:

Yes, you answered my questions quite nicely - thank you!