Friday, April 24, 2020

The Kotov Syndrome

In my last post, I discussed my slow approach to chess studies which included:     

  • setting up positions on a board
  •  developing a scoring system,
  • using a clock while evaluating a position 
  • doing my own analysis methods.

I started practicing these while revisiting Alexander Kotov’s popular Think Like a Grandmaster (TLAG) book committing to setting up each diagram on a chess board and going through the whole thing. Eleven years ago, when I was blogging about Zurich 1953, I did a brief post on highlights from Alex Kotov’s games of that match and provided a brief biography. 

In this post, I will get into the first section of this book as this really highlights Kotov’s approach with his “tree of analysis” calculations.  I’ll touch on what I thought were strengths in this section as well as big gaping holes leaving me scratching my head.

On the “Soviet School of Chess”

I’m a dinosaur. I learned chess back in the days when what was considered normal was playing  with analog clocks and playing chess in clubs where my older opponent would smoke a cigar and exhale in my face while I was deciding what my next move would be. We didn’t have databases, we had Modern Chess Openings (MCO) and chess informants.  The “soviet school of chess” was often referred to when looking up lines in the informants and getting books translated from Russia.  When Kotov and Keres published The Art of the Middle Game, it was first translated and published in 1964 by Dover publishing. This was a big deal.  Kotov saw a market in the US with the rise of interest from Bobby Fischer and continued to write books that would be translated in the 70’s. He has a whole series of his Grandmaster series due to the success of TLAG. Thus the “soviet school of chess” had a cadence and lots of fodder from this prolific Russian writer. Interest in gaining insight into how a Russian Grandmaster might think was fueled by a cold war and a world chess scene dominated by the triple K’s ( Korchnoi, Karpov and Kasperov).  Yes, I am old. Now, get off my lawn!

My 2020 motivation was to revisit this book and hopefully develop my ability to calculate positions deeper. I wanted to do more than a visualization exercise. I used to practice visualization training by playing blindfold against an old chess computer on the lowest level. I would play until my mind got jumbled and would have to “peak” at the position. I’d eventually get good enough to play that level until I could beat it quite regularly and move to the next level up. BUT THAT WAS 30 YEARS AGO.  Since then, once in a blue moon, I’d see how well I could play a game blindfold. For instance, back in the early 00’s on a family road-trip, I challenged my kids to a game in the car with my travel set. They had the board I had the road. (Yes, distracted driving… I know… but hey, play a gambit!) I managed to get pretty far along until I had to get a positional clarification.  Most recently, I attempted to play someone at work whom I sparred with quite regularly but seemed to be beating as well. To make it more interesting for both of us, I suggested I play blindfold in our timed match. I played the whole game ending with a winning position only to lose on time. Bottom line, you should try it, it’s not that hard and good visualization practice.

On the Tree of analysis
All that sounds like fun parlor tricks to play with your friends and all, BUT what I was noticing was difficulty in positional evaluations when I was branching and attempting to come back to the position at hand. Referenced in the book, Kotov parrots advice from Blumenfeld on visualizing the board using one’s imagination:

 “No matter how strong is your imaginative faculty, it is clear that the picture in your mind must be feebler than the one you get by looking at the board.  So, when your opponent moves, even when he has made the move you expected, you should never make the move you intended without further thought.”  --Blumenfeld

Basically, our mind’s imagination has a tendency to lie to us. Beware!

Enter Kotov. In his discussion of Tree of Analysis he covers various tree types:


  • Bare trunk
  •  Coppice
  • Thickets  

When do you invoke the powers of calculation? Am I to do this for every fricking move? No.  Consider the course of the game. You have your “book line” you are playing in the opening. When the first player takes a turn out of the book, that’s a good time to calculate. But further on into the middle game, coming up with a plan is one thing. Calculating the path to get there is another. A sharp position with complications on both sides is the key time to look deep into a position for the right move. Other positions require merely what he calls “positional Judgement” and Kotov loosely qualifies closed opening positions under this heading where its more a matter of positional posturing of the pieces rather than open combat in tactical play. So um, Good luck! This is an example of the Kotov-Ambiguity Factor I ran into.

He makes a bold statement “When analyzing complicated variations, examine each branch of the tree only once.” Here is where the Kotov-syndrome comes in. I can only speak for myself but I know others may have suffered the same fate.  The clock is ticking, I am evaluating a line of a move I think looks good (more on that later), I get to a position in my poor memory and replay the steps over and over. I finally realize that move won’t do so I look for another… rinse-lather-repeat. I finally realize 30 minutes ran on my clock and I panic, I look for ANOTHER move and make THAT MOVE  without any thought and hit the clock. You can deduce the result from there.   

The assumption Mr. Kotov seems to make in this first chapter is that the reader can already evaluate positions really well and are able to select proper candidate moves. Thanks for having that much confidence in my abilities Mr. Kotov.  You do realize I am blunder-prone and am trying to change that? 

Sure, there are trees with bare trunks where one move really sets things in motion. A slight variation is where forced moves are played. Combinations are another factor where a singular trunk would come in. These are easier trees to evaluate but knowing when a combination is in a position is another skill to be developed.

A coppice (shrub) is a position where you may have several options but they can be a bit easier to evaluate as each branch is more or less one or few move variations that presents an easier evaluation. Think endgames and positions that are fairly uncomplicated.  

It’s when you come into the thickets (each branch may be 5 or 6 moves deep), a complicated position that could go either way, is where I struggle the most. These are positions that offer many variations which diverge into quite different lines of play. These are the positions I most likely fall victim to the analysis paralysis “syndrome”.  I can’t for the life of me find good candidate moves, let alone ANY candidate moves worth evaluating.

On Candidate moves:

YAY! There’s a section on this in section one. I was delighted to see this and couldn’t wait to get to this section. (Is the expectation set right?) How, then, does a chess player choose which move to play in a given position?  To which, Mr. Kotov’s reply

“There is no easy answer; each player goes about choosing in his own way.”  

Thanks a lot pal.  This falls about two thirds the way through the first section on analysis and all he does is provide anecdotal stories of Lasker, Petrosian and other soviet masters’ unique approaches. For instance, he quips, Lasker considered developing the ability to analyze accurately by seeking out moves that are either most necessary and/or  most interesting. If I wanted platitudes, I would have hung out in a corporate conference room and read the placards.  

 I almost threw the book away at this point. This is the critical juncture of the whole book and his trees of analysis. Being able to come up with candidate moves to evaluate in the main position AS WELL AS each variation you need to examine is a skill that we now find varies from maser to master.

Parsing out some of the techniques in his homage to other player’s anecdotal ways on how to develop an eye for picking candidates requires:
  • Better understanding of strategies 
  • Knowing how to recognize tactics ( some of us have been down this path)
  • Need to develop  a Bird’s eye view of what’s happening in the position

I would have preferred the book to start with chapters on how to develop the skills around picking candidate moves BEFORE getting into a lengthy discussion about the tree of analysis.  I think this was the biggest shortfall of the book.  For the next sections, he does a cursory discussion on Positional Judgement, Planning and Ending which are necessary elements in getting to candidate selection.  He then devotes a whole chapter on A Player’s Knowledge. I skimmed that section. I had reached my end point and felt that this was rather trite and full of platitudes.

Don’t get me wrong, I trudged through the next three sections in the book to see what I could learn from Kotov’s Soviet school of chess. Next post I’ll wrap up the TLAG discussion and get into the following sections on Positional analysis and Strategy, which lead me to a couple side diversion and my eventual decision to transition to Silman’s How to Reassess Your Chess (future posts to come).
Thanks for listening.. now GET OFF MY LAWN!!!


LinuxGuy said...

The more I think about it, I realize what a distractive effect that Kotov's book has had on those who read it. While there are great examples in the book, and it's a tremendous book to read for improvement, people get thrown off by his ideas and prose. It's more of a context issue than an issue over anything that he is saying in particular.

There is so much more to say that I am trying to write a book about chess improvement at the moment. There are a lot of great books out there, like this one, on chess improvement, and they try to put it together systematically, but there are these unintended cracks in the edifice, or they lead one to the next question, such as "Where to candidate moves come from?" It wouldn't be such a thing, if so many didn't have the exact same complaints about this book, imho.

ChessAdmin said...

At least Kotov didn't lie: he wrote a book about how a Grandmaster thinks, not about how someone below that level thinks.