Stoyko and Phaedrus combined

This afternoon I struggled with the position below. If you want you can use it yourself as a Stoyko exercise. The position is taken from Khmelnitsky's Chess Exam.

After you have analyzed the position and written down every line, you can compare it with Khmelnitsky's analysis. If you highlight the space between the brackets you can see the line he recommends for white in this position.



(This is Khmelnitsky's analysis: 1 Rxd6!? was played in the game Baburin - Basas, Andora 1998. White got two pawns for the piece, neutralized blacks initiative, and gainde psychological momentum. Black never recovered from the initial shock and quickly fell apart. After 1 ... fxe3 2 Nxe3 Bf8? [already a losing move. Better was 2 ... Nf6, but after 3 Rc6 and 4 Rxc5 white has excellent chances.] 3 Rxg6! hxg6 4 Qxg6+ Ng7 5 Ng4 Be7 6 Nh6+ Black resigned)

When I had checked my analysis with Khmelnitsky's solution, I tried to win this position with the move recommended by the author and using my own Phaedrus Exercise. It was really challenging. Amazingly enough I already had real trouble winning this against Fritz 10 with a fixed depth of 3 ply.

Phaedrus Exercises

Somehow during the Christmas holidays I lose focus and become a bit more contemplative. This week I am at home, taking care of my children. While you may think that this gives me a lot of time to work at my chess, there actually is surprising little time to do so. There is (the) continuous (threat of) disturbance. Young children in one way or another constantly seek for attention, confirmation, comfort or just turmoil.

Still, all in all it is a great distraction from work and the daily routine at the office. Children are very direct, and it is wonderful to take care of them and give them the opportunity to play and have fun. On the side there is also the chance to do a bit of chess practice, but not the exercises that require focus and concentration (like step 6 exercises or Stoyko exercises) . Instead I have used my time mainly to contemplate a bit about the problems with positional chess training.

Unlike tactical training, it is hard to make positional exercises. The unforced nature of the positions makes the moves more debatable. And sometimes it is more a matter of how to approach a position, than a matter of specific moves. So in most books about positional chess one will find no, or only few, exercises. Most of them give model games that are well analyzed and explained. While I acknowledge that these are very suitable to improve your knowledge, I do quesion if they improve your skill.

Lets take the following position:

This is a position from the game Spassky - Fischer, Santa Monica 1966. It is given in Gelfers Positional Chess Handbook in a chapter about good bishop against bad knight. I will give game moves with Gelfers annotations.

35 h4 Nc4 36 Ke2 Ne5 37 Ke3 Kf6 38 Kf4 Nf7 39 Ke3 39 Bd5 is better (Gelfer). 39 ... g5 40 h5 Black has rid himself of the weakness at g6 but his knight is restricted to watching the passed h-pawn. 40 ... Nh6 41 Kd3 Ke5 42 Ba8 Kd6 43 Kc4 g4 44 a4 Kg8 45 a5 Kh6 46 Be4 g3 47 Kb5 Ng8 48 Bb1 Nh6 49 Ka6 Kc6 50 Ba2 1-0

If you have played through this carefully, you might have the impression that you have learned something that you can use in your games. But how can you be sure that you have? To test this I suggest an exercise that can be done by players of all strengths.

Setup the diagram position in Fritz or any other chess engine you have. Fritz has the option to chose a level. For starters chose a level with a fixed depth. Start with 1 ply, or if this is not challenging enough for you, with 3 ply, and try to win the position against Fritz. If you do, move up a level and play out the position with the engine set to 3 ply. Than 5 ply, and so on until you can't beat the machine anymore. As soon as you stop winning, analyze the games. What went wrong, and where. As soon as you think you know how to do it better, try again at the same level. At one time or another, no matter what you do, you will not win anymore. This is the time to stop and move on to another position.

I do not make claims about the rating points will gain with this exercise, but I do believe it is the best possible way to study most books about positional chess.

To make a name for myself, I do however claim the name "Phaedrus Exercises" for this method of training. Maybe I am mistaken, but I do not believe that (however obvious it is) until today anyone has ever suggested this exercise.

Happy holidays to all A.C.I.S. members and other readers of this blog!

No sneaky preview

In last weeks post "searching for Steve Stoyko" I explained the dilemma I encounter when I select positions for my Stoyko Exercises. A Stoyko Exercise is the most challenging when one starts it without having a clue about the game continuation or the given analyses. When the position meets these requirements, the exercise comes really close to a real otb chess game. But the exercise also demands that the position is complicated and/or dynamic. And last, but not least, it is best if I can compare my exhaustive analysis with those of a much stronger annotator.

But finding positions which meet those criteria is impossible without a glance at the position and the continuation and analysis, which at least gives away candidate moves. I absolutely don't want this to happen, so I have found ways to tackle this problem, but not to my complete satisfaction.

At the moment I am using positions from two sources. I get my positions from Agaards "Excelling at chess calculation" and from Pata Gaprindashvilli's "imagination in chess". These positions are in general sufficiently complicated to offer a real test to my calculation powers. Because the answers and analysis are given at the end of the book, I do not have to fear getting involuntarily preview at the solution.

There is however a small negative side to this method. The positions I get are not only highly tactical, but they also have a very clear "solution". Contrary to real games, every position can be solved to a clear advantage (win) or equal position (draw). So there still is an certain urge to obtain a database with positions that do not lead to a clear cut conclusion. Alastairs suggestion to let another player do the selecting might be the best way to get exactly what I want. Teaming up with a player of more or less equal strength and exchange positions is probably the way to go.

Thanks to everyone who gave me suggestions to find Steve Stoyko. Your help is greatly appreciated.

Searching for Steve Stoyko

As described in my last post, I am currently doing two Stoyko Exercises every week. The aim of these exercises is to improve my calculation. I love doing the exercises, but there is a practical problem that I encounter in the search for suitable positions. How to find good positions without spoiling them?

I do not have a coach, so I have to rely on books as a reference for checking my calculations. Of course there is also the possibility of using a chess engine. But I very much prefer a human assessment of the lines I found as a reference. The main problem with engines is their evaluation of non tactical positions.

Because of this, I want to work with positions that are exhaustively analyzed in a book. But to find them, you have to look at them. And even a glance involuntary gives away moves and lines before you start exercising. I think I have found a few ways to get around this obstacle, but I would love to hear some suggestions from other players to tackle this problem.

Step 6 and Stoyko exercises

At the moment I use two training methods. The first is step 6 of the steps method. I already did this book once (over a period of more than a year) and had an overall succes rate of about 85%.

At first I thought it would be best to concentrate on the exercises I failed to solve. But when I started to pick up the book I noticed that I didn't remember most of the (1300) positions. For this reason I decided to go through the book again. Not surprisingly my succes rate is now about 95%, and I also solve them a lot faster than I did the first time. I consider this work to be largely a kind of maintenance of acquired patterns.

The other method I recently started is doing 2 Stoyko exercises a week. I have chosen this method to improve my calculation abbilities. Last season I noticed that most of my opponents calculated deeper than I did. So this is probably my weak spot.

Below a description of the Stoyko exercise that I copied from Dan Heismans website.

FM Steve Stoyko suggested this very helpful exercise. First the reader should find a rich middlegame position. You can find them in many Kasparov, Shirov, or Speelman games, or in the books The Magic of Tactics, Genius inChess, or How to Think in Chess. Take out a couple sheets of paper and a pen or pencil.The idea is to write everything you can possibly visualize from the position, like you were playing the game without a clock and you had to see and record everything before you move.

Write down every line that you look at (no matter how bad!), along with that line's evaluation. This should fill up several sheets of paper and take 45 minutes up to 2+ hours! If you chose a sufficiently complex positions dozens of variations should be considered. Consider lines to as much depth as you think is significant.You can show your judgment of the evaluation (who stands better and by how much – you don’t always have to say why) with any number of methods:

1. Traditional: =, ±, ∞, …
2. Computer - In pawns; negative means Black is better: +0.3, -1.2, …
   
3. English: White is a little better, Black has compensation for his lost pawn, etc.

When you are done, take your analysis to a good instructor, player, or software program. Look at each line to see how well you visualized the position (any retained images, illegal moves, etc.?), and also compare your logic (was that move really forced?) and your evaluation.In general the Stoyko exercise, if done properly, should help you practice and evaluate the following skills:

1. Analysis
   
2. Visualization
   
3. Evaluation

Steve claimed that each time he did this exercise he gained about 100 rating points!

Why are we seeking?

Blunderprone has started a new circle of chess players on the road to improvement. I think this is a wonderful initiative and it certainly seems to have rejuvenated the chess blogging scene. When I started reading all the reactions on different blogs, one stood out as the most fundamental. It was Robert Pearsons post in which he questioned the point to seeking improvement.

It struck me and reminded me of a question a that co-worker of mine once received when he said he was going to the tennis club. "Are you going to play or work on your backhand" he was asked ironically when he parted.

I understood why this question came up. This co-worker is a very competitive guy who doesn't seem to enjoy playing when he loses. And when he talks about his games he is always pretty self critical.

But isn't this something that everyone who is part of, or feels related to the A.C.I.S., recognizes? We want to improve, so we can be more successful. And most of us already have made all the improvement that can be gained within our comfort zone .

What we recognize and appreciate in each other is the ambition to improve, win more games, gain rating points and (above all) make sacrifices to achieve this.

So the answer to the question "why seek improvement" for me is: "to beat all of those lazy (and maybe more talented) bums who don't". Maybe it is some old fashioned Calvinistic work ethic that makes me think that justice has been done when the player who has worked the hardest, wins the game.

65 points in 400 days

When I started blogging last year my rating had reached an all time high. I hade made a jump from 2040 in november 2007 to 2068 in february 2008. This succes inspired me to do two things:
1. Start blogging.
2. Working through step 5 extra, 5+ and step 6 of the stepsmethod.

I worked through the steps while commuting in the train. Not an ideal enviroment, but it sure beats doing nothing or slumbering as far as chess improvement is concerned.

Today the provisional version of new Dutch ratinglist was published. It shows that I have made it over 2100. My new rating is 2105. Since I started to get serious in november 2007, I have gained 65 points. Not even close to the 400 that MdlM made in his stint, but believe me, I am really pleased this accomplishment.

Answering Caquetio, Tanc and Polly

Before I return to the core of my subject of my last post, I will first take the time to answer to the analysis that I received in the comment section of this blog. I asked my readers to give their suggestions for white in this position.

The position is taken from the game Andersson - Vaganian, Skelleftea 1989.

The Caquetio Knight said:
1.Qd2! Putting pressure on the d6 pawn.
Plan white: Rfd1 increasing the pressure on d6, b3 protecting c4, f3 protecting e4 if necessary. Look out for the appropriate moment to play Nd5 or exchange the dark bishop on h6.
Plan black: black’s counter play is on the Q-side with a6, b5 and the bishop looking down the long diagonal and on the long run the breaking move f7-f5. But the rook belongs than behind the lever pawn. For now if he plays 1…Rd8? 2.Rfd1 Ne5 3. b3 and black is in trouble cause f4! is in the air.
1…Nd4! 2.b3 (2.Bxd4? cxd4 3.Nd5 d3!) f5!

Tanc (happyhippo) came up with the following analysis:
a. The key to the position is to note that Black's 2 most active pieces are his g7 bishop and e6 Knight. The Queen is strangely offside at b8.
b. Black has a Knight that is likely to come to e5.
c. Black's other Knight is also coming to d4 soon so I need to find a way to counteract it. If I allow these 2 Knights a chance to come into the center, White's position would be difficult.
d. On the White side, the b2 pawn is weak.
e. Black is very unlikely to trade the strong Bishop on g7 for the Knight on c3 else it just opens up the dark squares around his King and that is suicidal.

I first thought about the move f4 then f5. to pry the position on Black's kingside open. But then I run into the problem of Black responding with Nd4. This is a monstrous Knight and needs to be removed. How do I do it? I cannot shuffle my Rook nor move my Queen to attack the d4 square once Black put his Knight there.

With this in mind, it seems tempi is critical here. How about Rc2? I now protect the b-pawn.
The move looks good. I'm also threatening to play Nb5 next attacking the d pawn so it looks like Black is forced to play a6 on the next move to stop to defend this crucial pawn. Afterwhich Black will surely play Nd4 on the next move and Rd2 Black's Knight is now threatening to overwhelm the position with Ne5 and those 2 Knights will be a handful to deal with.

So after 1 Rc2 Nd4 2 f4 (to stop Ne5). This is probably what I would play.

Polly agrees more or less with Cauqetio
Hippo: I'm not overly afraid of the knight coming into d4 though it does stop the idea of piling on the d6 pawn. Perhaps Bg4 pinning the knight on e6. It might provoke f5, though I think that's a lousy move for black.

It's funny looking position with the knight on e6. It looks like a Maroczy bind type position, though as Black I would not have been so eager to get rid of my e pawn. I can't even imagine the move sequence that had black capturing on e6. Having played the Black side of the Maroczy I know that one needs a lot of patience to get a break for Black. I've also reached the White side by transposing against players of play c5 against my English.

I like Qd2 because it not only gives White a chance to put a rook on d1, but also gives the possibility of Bh6 trying to trade Black's dark squared bishop.

Nd5 also looks appealing just out of general principle

This is paraphrased (to avoid copyright issues) what Aagaard has to say about this position:
He points out whites completed development and think it is hard to say what the best position is for the queen, so it would be best to leave her where she is. He feels that the rook on c1 needs to be improved, and that b2 needs protection. Blacks position is in his eyes in a bit of trouble, though the knight on e6 and the bishop are well placed. He might consider a6 followed by b5, but he also has to take care of his weaknes on d6.

The only way for black to defend this weakness is by utilizing the control over d4. So white should attack down the d-file with Rc1-c2-d2.

He gives the game continuation with the following analysis.

15 Rc2! threatening 16 Nb5! 15 ... a6 16 Rd2 Nd4 17 Bxd4 cxd4 18 Rxd4! Bxd4 19 Qxd4 leading to the following position.



Here, according to Aagaard, white is slightly better, and brings under the readrs attention the white threats Nd5 and Bg4.

When I tried to solve the position, my thoughts were about the same as those of Caquetio and Polly. I saw the apparently strong outpost on d5 as a significant feature, and realized that the pawn on b2 needed to be defended before the knight could be played. So in my opinion playing Qd2 (which also connected the rooks) seemed to be more than obvious.

After reading Aagaards analysis I do however agree with his judgment, and must confess that Rc2 is way stronger, so cheers for Tanc (happyhippo) who also thought this was the move to be played). The key to this position is the exchange sac. This is what I had missed, and what really gives a bite to Rc2.

Though Andersson gets the credits for finding this exchange sac, I think Aagaard also deserves some credit for his explanation. However I believe that this position is a very bad choice for an exercise. Even Temposchlucker, who seems to like this book, agrees with me. In my next post I will try to make clear why this position is not suitable as an exercise for self improvement, which at the same time is my answer to Temposchluckers comment.

Looking closer at "Excelling at positional chess" 1

Please look at this position.

It is white to move, and this is the first exercise in Aagaards book "Excelling at positional chess". What move would you play?

Space and capacity

BDK's planning exercise and the follow up made me grab Michael Stean's classic "Simple chess" again. I especially felt the urge to read what Stean had to say about space and the advantage it is to suppose to offer. And again I was impressed by the clarity of his explanation. This a small part of his treatise of this subject:

Space is not an easily definable or recognizable concept. The visual impression you obtain by glancing at a position and estimating who seems to have the lion's share can be misleading. The following is nearer to the truth. Any given pawnstructure has a certain capacity for accomodating pieces efficiently. Exceed this capacity and the pieces get in each other's way, and so reduce their mutual activity. This problem of overpopulation is easy to sense when playing a position, it "feels" cramped. To take an example, compare the next two positions.





They do, of course represent the same position, but with two pairs of minor pieces less in the second case. In the first diagram black is terribly congested. There is no way he is ever going to be allowed to play b7-b5, while alternative methods of seeking some breathingspace by (after due preparation) ... e7-e6, or ... f7-f5 would compromise his pawnstructure considerably. White on the other hand can build up at leisure for an eventual e4-e5, safe in the knowledge that as long as he avoids any piece exchange, his adversary will never be able to free his game.

The second diagram is quite a contrast. The size of blacks forces is here well within the positions "capacity". As a result there are no spatial problems at all and black can very quickly seize the initiative by ... a7-a6 and ... b7-b5 or even by ... b7-b5 as a pawn sacrifice, e.g. 1 ... b7-b5 2 cxb5 a7-a6 3 bxa6 Rxa6 with tremendous pressure. We see from this pair of positions that blacks structure is very good, but his capacity is small. Visually white has a spatial advantage in both cases, but in the second the eye flatters to deceive. In fact he is grossly overextended. A vast empire requires an army of equal proportions to defend it.

I have nothing to add to this, besides a glance at the position BDK gave us in his planning exercise.


Indeed, white has space? But why should this be an advantage, considering the explanation given above? It seems to me that blacks forces are will within the capacity of his pawnstructure

Blindfold?

Temposchlucker asked me: I assume you have tried to play blindfold chess. What's your experience?

I have to confess that I did try a lot of methods to improve my visualization and calculation. And amongst those was also playing blindfold chess. I was not very good at it. I managed to play blindfold without playing illegal moves, but it was not something that came naturally.

The way I played was that I would reconstruct the complete move sequence from the first move onwards, to rebuild the chunk I wanted to analyze. When I had done this (I did this fairly quickly) I was able to analyze the position a bit.

As far as I can remember I never felt that I improved. Not in playing blindfold, nor in regular OTB games. The only thing that made me better in visualization is the platform technique that Tisdall described in "improve your chess". Not that it made my visualization and calculation deeper or faster, but mainly because it made it more effective and efficient.

What doesn't kill you makes you stronger!

This is the material I have worked with the last year. The workbook is almost completely worn out. I accompanied me during a year on every train ride to and from the office. There never was a book that forced me to work as hard and continuously as this one!

The workbook without its cover.

The torn and worn out cover.

My notebook.

My notes.

My notes in closeup.

Patterns, patterns and patterns!

In my last post I made it clear that I believe that I am relatively good for a player of my strength in pattern recognition, and that I am relatively weak in visualization and calculation. This is something I see confirmed in the post mortems with players with ratings comparable to mine, or sometimes even significantly lower rated players. They have seen more variations, calculated deeper and I more often than not think that I can see that they calculate a lot faster.

I once was fortunate enough to witness a post mortem between Timman (at that moment ranked third in the world) and Kasparov. Kasparov won that game in which the following unusual ending ocurred:


In this position it was White (Timman) to move. Kasparov (with the black pieces) won this game even though almost everbody thought it was a draw. During the post mortem it was stunning to see the difference between kasparov and Timman. Kasparov gave one variation after another in an amazing speed. Immediately rebuffing almost every suggestion Timman made to improve upon the game.

Witnessing this I strongly felt that Kasparov brains just worked much faster than Timmans, and that the only thing that could compensate Timman for this disadvantage was maybe experience and knowledge (read patterns). Unfortunately for his Dutch fans however, Timman never could prove that his experience and knowledge were superior to Kasparovs, and after this match he never again was able to seriously challenge Kasparov again.

Nevertheless, Timman of course was a world class player. So it is possible to be good at chess, even if your main strength is not calculation and/or visualization. And I think that this explains my progress from a steady 1900 player to a steady 2000+ player. The period in which I accomplished this, is the period in which I mainly concentrated on improving my pattern recognition. I started to calculate less, but the variations I calculated were much more relevant than before.

And up to this day I still think that the road for improvement for me is to work on my pattern recognition. I do step 6 of the stepsmethod now for almost a year. I have nearly completed the workbook. My succes rate is about 75%. There is hardly a solution in the book that is deeper than 11 ply, and neither will you find solutions with more than 3 to 4 variations. As long as I keep missing 25% of the patterns within this horizon, there seems to be little reason for me to try to improve my abbility to look deeper.

And besides, before I started to concentrate on pattern recognition, I have tried a lot to improve my visualization and calculation skills. It did not improve my rating, and neither did I have any other indication that it brought me significant gains. But maybe this is caused by the fact that I did not use an effective method. So for now I focus on patterns. The question remains however what to do when I have finished step 6.

More on calculation and patterns

Temposchluckers attempts to shed some light on calculation and patterns (and the following posts) has inspired me to share some thoughts of my own on this subject with the readers of my blog. This is not primaraly meant as a comment or criticism on Tempo's observations and conclusions, but more a reflection on my own dicoveries and during the past year.

I have already stated several times that I think that for a player of my strength (2050+), I pretty much suck at calculation and visualization. So if my rating has some validity at all (and I like to believe is does) I have to do other things relatively well for a player of my strength. It seems to me that I am pretty good at pattern recognition.

Of course pattern recognition, calculation and visualization heavily interact. And is is very difficult to divide between these three. I have asked myslef if it is at all possible to find the best move without any knowledge of the appropriate pattern? Imho this is hardly possible. One could make a case that in a position where it is mate in 1 move, the best move can be found simply by exhausting every possible move that gives a check. But one could also say this checkmate is a pattern that is eventually recognized.

Because of the number of possibilities, which rises exponentially with every move, sequences of more than 3 to 5 ply cannot be found by calculation and visualization alone. At least not by humans. Computers however can.

But for us mortals, calculation and visualization have to be guided. This is largely done by the search for patterns. A player who finds the relevant patterns does not have to calculate and visualize very as much as a player who doesn't. The latter is clueless. He will have to calculate a lot of move sequences to find as much as a decent (not losing) move, and have a small chance of finding the best.

Computers however can use brute force. And they are so powerful now, that their ability to prevent weak moves within their horizon, outweighs the importance of even the strongest human players to (occasionally) find the best move beyond that horizon.

So in order to improve there are two leads. The first is to improve my calculation and visualization, and the other is improving pattern recognition. In my next post I will tell you the choice I have made the past year (in which I gained over rating points, reaching an all time high).