How to Reassess Your Chess: Mr. Silman, could you answer these questions?

I have the third and fourth edition of JS's HTRYC. Already in the third edition I noted that JS did not give very detailed guidelines how to analyse the imbalances in the practice. At all, I am missing some more information how to use the book. Let's take one example of the fourth edition:

The game is Topalov vs Ivanchuk MTell Masters 2008, diagram 31 in HTRYC, fourth edition, black to move.

Now, should the reader go first through all the imbalances? Or should he begin to calculate forced moves, like how to defend the pawn on b5? 

We are discussing the HTRYC in a club, and it is for us (or let's say, at least for me) still not clear how to approach the positions. I remember that in the third edition JS was explaining that it is important to use a fantasy position before beginning to calculate (already in the beginning of the book), but later he wrote to calculate forced moves, or just play a move which was obvious the best without even calculating anything. 

I can of course draw my own conclusions from what I know until now from his books: calculate when you have to calculate, but go through the imbalances when there is nothing forcing in the position and try to find a plan based on the imbalances. But there are still some open questions here. In the position above it is clear that black has to defend the b-pawn if he wants to keep material. Should I first deal with this, or should I first go through the imbalances? JS went through some of the imbalances:       

"This position is better for Black, but that assessment might prove somewhat surprising for those players who noticed the active Bishop on e2, the weak double-isolated pawn on b5, and white's central and kingside space advantage. However, Black's not without his own perks. After all, he does have a queenside spatial plus and white's pawns on a3 and c2 are both weak and under pressure". (Silman 2010, part two) 

One of the imbalances is weak squares (holes). First I thought the e4 is such a hole, and could be perhaps be used by the black knight (via d7-c5). This square was not relevant for the game!    

http://www.chessgames.com/perl/chessgame?gid=1492879

The game continuation was very difficult to predict, as the knight did some backward jumps, in some variations allowing a white pawn making a fork on the knight and queen (but was not a problem for black). 

Returning to my initial question: how can I know that the e4-square as a hole is not relevant for the game? Should I be able to calculate already several variations to come to this conclusion? I am not criticising JS, I just want to understand the best way to use this book and how to use the concept of imbalances in a real game.  Or to put in other words: I know in retrospect that the e4 square was not relevant. Should I have enough calculation's skills to see in the initial position that the knight should jump to d7 and then to b8? Or is d7 just the only move that doesn't drop material, and the decision how to proceed from there will be taken after white's next move? 

Another question from somebody in our club was: when should I begin to search for the imbalances? I tried to answer this question arguing that you should begin to think on imbalances as  soon as you are out of the book. This can be a little bit exagerated: let's say your opponent plays 1.h3 and you you are already out of the book - in this case you play something that you think it will fit to your repertoire, in the case that you have one :-) But the best thing would be if some strong players (or perhaps even JS) could give some advice here.

By the way, I am aware that I have to get used to the analyses of imbalances in such an extend that I should even not be aware that I am searching for imbalances. Yes, but at the moment I need something like a clear guideline how to approach the positions. 

Related to imbalances and how and when they apply, since I currently don't have a copy of Reassess, but from Amateurs Mind chapter 1:

"You have to read the board and obey its dictates."

So what's reading the board? Basically understanding the basic components (imbalances) of the position.

Silman: "Before you get carried away, let me remind you! DON'T look for individual moves! In fact, never calculate until you understand the basic components (imbalances) of the position."

Reading the imbalances, plusses and minuses of the position based on the different categories, will suggest a plan to "strive to make them better than the imbalances your opponent will be using."

Then you start calculating on how to implement that plan either offensively or defensively.

In Kotov's "How to think like a grandmaster" he describes a like process, without using the term imbalances, either in the intro or chapter one.

Okay, I know this statement, but in several other parts of his books he says to check first forced variations. For example, in the third edition of HTRYC he wrote "In general, calculation is necessary to verify the tactical worth of a move that seems to take care of the needs of a given position" (p. 39). This is obvious in the diagram 30 in this edition, where both players were trying to mate the other, so you have to calculate this, as everything is forced. Willy Hendriks, assumes that we "don't first judge the position and then look at moves. It all happens at the same time".  

I mean, clear guidelines how to approach the positions in JS's sense - I have to do this, as I am studying the book. By the way, piece placement or pawn structure is part of the imbalances.

You said piece placement is the most important, but I think that we can not assume this for every position. For example, in some positions is queen safety much more important - even with horrible piece coordination, if you have nevertheless mate in two this is the only important aspect in the position, right?  And I didn't write the book, I am only studying it. The book is trying to give a kind of guidelines how to evaluate positions.

In the 4th edition, there is a comment about "tactical issues" in the paragraphs before diagram 18.

In the position above one important factor is also development: black has most of his pieces ready to jump on white's position (only the knight needs a better square). 

Yes, I know. My problem is that Silman should not suggest in some positions that we should not calculate at all, and some pages later he says you should calculate. I think he is probably not completely aware that weak players are sensitive for this kind of contradictions, while better players simply jump over this and pick up what they need.

The basic concept Silman is trying to get across us that each player us going to have plusses and minuses according to the position. This starts straight out of the opening.

Best play comes from correctly assessing the position, coming up with a good plan based on that assessment and then calculate correctly to get moves to realize the plan.

In other words, working backward, each move has a reason, a reason based on a plan with a certain goal, a goal based on what the position dictates. Every example is very very likely to follow this process. In your page 39. example, the key phrase is "take care of the needs of a given position."

So board read first, plan next, then move. That's what GM's are referring to when they say stuff like " most logical move " or "move in keeping with the position".

Silman didn't invent the process, Kotov wrote about it and Reshevsky that I remember. But I'm sure it goes way back before them.

The passage goes like this:

"Make sure you are aware of any crude threats by the opponent, and also do a quick search for any basic tactical themes that might be present. This is usually done subconsciously by players 1800 and up. However, if you are lower rated, are prone to blunders, or feel that you have serious tactical issues, then it's a good idea to take a few moments to get on top of this stuff". (4th edition, text previous to diagram 18). 

Perhaps Silman wrote the book for people who are 1800 or up, and we patzers are losing our time with the book if we are not able to see tactical things going on on the board. But I don't think so. The book is good for everybody who wants to invest enough time to understand it. Perhaps Silman should stress more what he wrote above when he wrote about the list of imbalances: only go through the imbalances after you check all relevant threads or potential combinations for both sides (what means perhaps you have to calculate something before even thinking on the imbalances - good players probably do not separate these two steps).

Related to calculation in general, when a move is forced defensively, you don't have to calculate at all. When a move looks to gain an immediate, sustainable advantage and you ruled out any effective counterplay. You just double or triple check your line, so not much calculation. Other differing situations will necessitate different calculation needs.

I'm not sure what calculation example Silman is dealing with but it may relate to awareness of over calculation.

You know, what you said about "check the lines" is for me already "calculation". I am really bad in visualisation, and I have a huge difficulty to see the position in front of me already after two moves. This has something to do with my brain, I have a horrible orientation too. I am trying hard to improve this by doing tactics, and probably I am doing better than some years ago, but I am still weak in these things. 

Assessing threats is part of assessing imbalances. Usually under minor pieces and/or initiative. Reading the board is not necessarily static. Potential of each piece is important to consider.

For example, it's important to see the potential of a really great, maybe strategically winning, opponent knight outpost before it happens. Silman deals with this with the tricky area of getting in your plan but needing to be aware of opponent's plan too.

@torrubirubi,

Every good chess player wishes he could calculate better. One exercise I do is take a GM game, find an exchange or an exchange sequence of maybe 5 moves, play it on a board slowly a few times, then play it over in your mind with your eyes closed. Then try to do it later in the day and the next day. When you get good at it, you can try to change the moves in your mind and then see if it works on a board.

Easiest is starting with opening positions. Like Lasker's Qe2 Ruy Exchange or Petroff or Scandinavian with early exchanges. Never know, might be helpful.

Did you know he is a member here and you can send him a message directly?    https://www.chess.com/member/silman

/ He may ignore it but at least there is a chance he may respond.

As a rule, always go from the general to the specific.  Make sure you understand what you should be doing in general before trying to figure out a specific move.  The general understanding will help guide your search for the best move.  For instance, if the position is wild and you are up material, trading pieces will limit counterplay, lower the risk of a blunder and guide you towards a simplified endgame win.  Therefore, all other things being equal, you should value variations that trade pieces (especially the Queen) over ones that don’t.  This will help you discard unsuitable moves and lead you to the best one, or at least a good one.

You don’t need to do this every move, as things may not change much if at all.  This general overview, though, should always be front and centre.  Many people do this automatically for very rudimentary concepts (eg, I’m down lots of material but his King is weak, so I should avoid Queen trades).  With training and practice, this can become more refined and encompass more chess concepts automatically (eg, good and bad bishops, pawn majorities, open files, etc).  GM Simon Williams suggests taking time to consider these general factors after every pawn move and/or piece exchange, because these fundamentally change the position.

Once you understand the position, that’s when you can calculate as deep as necessary.  It also depends on time, as it is much easier to do when you are analyzing a game at home than playing a 3min blitz game.

These are for sure good exercises. I played already chess with colleagues, where I agreed that I would play blind the first 9 moves and then I would be allowed to see the board. Funny that people get very wild in an attempt to confuse me, playing horrible moves. Usually they are already losing after 6 moves! I tried already playing opening lines in my mind, going as far as I could, but usually I do much better if I can have an empty board. This helps.

"Now, should the reader go first through all the imbalances? Or should he begin to calculate forced moves, like how to defend the pawn on b5?"

Sorry I don't have time to explore the position in detail but, as long as you have sufficient time on your clock, you should look at the imbalances before doing a knee-jerk reaction and defend hanging pieces or pawns.  It might be that you have a strong move or your opponent will put himself in an awful position by grabbing material.

Of course if you can give mate in one you can ignore material. I was talking more about the balance between calculating / checking imbalances / searching for single moves / checking forced moves. It seems that good players are not that discipline in their thoughts, and they jump from calculating lines to imbalances and back to calculating and evaluating positions after a couple of moves. As they have a huge number of patterns saved in their memory, they can afford to skip whole lines and focus on more relevant things, while we patzers are wasting a lot of times with irrelevant things.  

I think you're absolutely right there: the thinking process isn't that conscious, or only that conscious on rare occasions. I tend to apply Silman's thinking technique of imbalances, or Kotov's technique of candidate moves whenever I feel I'm stuck - it helps you to look at the position with fresh eyes and unearth new ideas and concepts. This is also another reason why I'm a little bit critical towards Silman as a pedagogue (while still considering HTRYC 4th ed is a great book). I can't help feeling that he regularly loses his intended target audience (1400 to 2000 USCF rating or 1200 to 1800 Elo rating, if I remember correctly).

One advice that you always hear from strong players is to play over old, well-annotated, master games - the more the better. Even if they're not well-annotated or you don't understand everything (unlikely for anyone who isn't highly proficient), you'll quickly get a much better instinctive feel for where your pieces should go even you don't specifically know the specific opening or variation. I remember doing exactly this when I was a junior. In my first Swedish championships, I failed miserably, scoring a miserable 2/8 (+1, =2, -5) in one of the junior qualifying pools. It probably corresponded perfectly to my actual playing strength at the time, but this hadn't stopped me from expecting a much better performance, perhaps as much as 50% (I especially remembering smarting very badly at two defeats against opponents of roughly my own rating, victims ripe for the picking in my eyes). Totally unrealistic in hindsight, because I hadn't done any real, structured work on my chess up to that point. That summer I started playing over master games regularly. In fact it became my main form of chess training (another error with hindsight: I should have paid specific attention to strategy and tactics, too). It had an immediate effect. By that autumn, I probably was about 70 points stronger, and by the next summer I was 300 points higher than I had been exactly one year earlier, and scored an undefeated 5½/9 in a senior section at the Swedish Championships. From then on I basically jumped 100 to 150 points per year until I was about 19.

In general, you should calculate only forced moves/variations, then stop calculating until there are no more forced moves, then evaluate the position. In the diagram you posted, there are three candidate moves you should consider first: Qd7, Qc6, Rh5 and Qxc2+ (yes, I know this loses the knight, but this is a capture and a forcing move so it's one of the first move to evaluate).