I mentioned in another thread that I had been looking at an Andrew Soltis book, "Studying Chess Made Easy", but in a superficial way. I found it interesting enough to go a bit deeper, and came up with a very interesting find.

First, though, to tie up a loose end or two, let me emphasise again that he really does plug the intuition-over-logic aspect, and in one instance my deeper dig hit paydirt when he mentioned research that apparently claims atypical grandmaster knows about 100,000 patterns. Clearly such a large number can't be learnt by sitting down and trying either to understand them or to memorise them - unless you want to spend 30 years doing ten every day, he says. The way the patterns (which are rather fuzzy, by the way) are instead absorbed subconsciously by looking at games and doing problems, and the trick there is simply to be it "provoking", which may be nothing more than setting a challenge by saying e.g. "White to play and win". You can also do things like trying to complete a set of problems in an ever shorter time. The important thing is to leave the subconscious to do all the really hard work.

He does, however, put a good word in for memorisation, and the book is packed with good advice that would benefit any amateur go player. Apart from the topic of this thread, one special thing caught my eye was his introduction to the long and very important chapter "Living with TMI" (i.e. Too Much Information. He states: "As you improve as a player, you may be surprised to find that choosing a move becomes harder, not easier." Right on! Brimming with good sense, he discusses things like conflicting proverbs.

I'd say that chapter alone is worth the price of the book, but the same claim was made by a chess reviewer I very much respect (Peter Doggers) about the concept of priyom.

It's a Russian concept and it seems that it was Soltis who introduced it to the English-speaking world. He rendered it as "priyome", presumably to stop Americans mis-pronouncing it as priyam, but I prefer to stick with the standard transliteration.

What fascinated me about this is that the term seems to be more or less an exact equivalent of tesuji. Priyom is a common word in Russian and perhaps "take" is the best choice of word for the underlying sense in each case. I more or less gave up chess before I got my Russian degree, so I never came across its chess usage, and I get a feeling that its chess usage may even be relatively recent anyway. Various writers on the internet (who include several who copy Soltis and try to pass off the ideas as their own - but we are familiar with that in the go world, too) go into contortions trying to define the term (e.g. a combination of function and structure) but we go players recognise it instantly as a tesuji - with a proviso.

But a standard meaning in ordinary Russian is just "way, trick, technique, manoeuvre". Examples in chess include a specific shape that calls for a specific move. But they also include looser things such as "if the opponent's rooks can't on the back bank, push the edge pawn". Too many go players think of tesuji as a brilliant move, and so the proviso I just mentioned is that you may have to remember that tesuji can also mean "way of playing", so the match is perfect.

In fact, being a little fanciful, you could even say that the Japanese term combines function (te) and structure (suji) in itself, so long as you remember to think of structures as being fluid as well as static, as is the case with suji (or haengma).

Assuming my impression that it's fairly new in Russian chess is correct, I suppose we can even speculate whether chess borrowed it from Japanese go.

It seems to have been a tool of Russian chess trainers, and they apparently instructed their pupils to record each priyom in a notebook. Understandably, there has been great curiosity outside Russia as to what these notebooks might look like. I'm curious too. I gather it's not the whole position that matters - it might be simply the position of four or five pieces. Soltis's own discussion of the topic doesn't shed light directly on the notebooks but is a well written essay nonetheless.

So, does anyone know more about priyomy - history, usage, notebooks? Do Russian go books use the term for tesuji?

This would be a good time to say hello to the amateur world again, Sasha

I don't have this Soltis book. However, I do have GM-Ram, Essential Grandmaster Knowledge, by Russian GM Rashid Ziatdinov. I highly recommend it. IM Silman's review: http://rashidchess.com/gmram.html and GM Kavalek's review: http://www.huffingtonpost.com/lubomir-k ... 73180.html

At a simpler level, but with the same concept of knowing a few fundamental positions, is American GM Lev Alburt, Chess Training Pocket Book 1 & 2.

Okay John, you got me to purchase a digital copy. Let's see if I get my chess above club-casual now that I'm at it

_________________
Geek of all trades, master of none: the motto for my blog mostlymaths.net

I'm off to the chess shop too tomorrow!

The GM-RAM looks interesting. The review says:



1400 is a typical amateur tournament player (6-2 kyu in go??). Providing it is done in the sense of following the flow of the game, I could buy the proposition that memorising one go game will make you able to be that level (I assume a little bit of practice is required to cement things in place, but that's reasonable). And memorising a go game is surely harder than likewise for a chess game?

A little more on the 1400 mark in chess (is this some sort of special benchmark?) from Soltis: "Until you are at least 1400-strength, most of your games will be decided by blunders.. most of your games will be decided by blunders." He then gives a section of examples where he tells you to ask first: "Was the opponents last move a threat?" Also, help yourself by turning the board round. If there is no threat, your next move is unlikely to be an outright blunder. Eliminating one blunder from each game will, he says, improve your rating 100 points (which is massive).

That sounds so simple and obvious it's almost daft, but how many of us neglect it? But I tried on the Soltis examples and was astounded not that it made a difference (I expected that) but at how much difference it made. Apart from looking at your own games, he also recommends using the "Was the last move a threat?" technique in looking at master games. I'm not likely to try that, but I have hunch it would work big time in go as well.

On the topic of lists he is very scornful. Apparently Tartakower made a list of more than a dozen types of sacrifice (irruptive, rolling-up, etc). Too Much Information (and probably meaningless), says Soltis. The only list you need to know is that there are four types of move:

1. Tactical moves
2. Repositioning moves
3. Exchanging moves (= trades)
4. Moves that change pawn structure.

I postulate that this tiny list can be replicated in go by counting honte and thick moves as "repositioning moves", and for "pawn structure" substitute moyo, and in all cases changes "moves" to "moves and sequences".

Here's the rub: "Statistical surveys of [middle games in] master tournaments show that repositioning moves are the most common. In second place are tactical moves." Further, pawn (and moyo?) structure moves have the special feature that they so often change the evaluation of the position, they require special study. (It might be useful to mention here that what most westerners think of as a moyo is usually an o-moyo, or big moyo, in Japanese; the true meaning is simply a territorial framework, of any size.)

So the idea - mainly for stronger players, I think, as the purpose is to deal with TMI - is to get into the habit, when studying master games, of categorising moves on the basis of this very short and so manageable list, and then asking questions such as: was this trade forced? Who does it favour?.

Seems potentially useful in go, no?

Seems so. I was going to categorise mistakes (following Axel Smith's Pump Up Your Rating with his 'list of mistakes' and D.E. Knuth with his bug categorisation.) But I expected it to be more like 15-17 items long (Knuth's is 15, not too long, not too short.) Its reverse (which could be condensed, since I think moves can be categorised more broadly than errors) could easily be move categories.

_________________
Geek of all trades, master of none: the motto for my blog mostlymaths.net

What is he, the Charles Atlas of chess teachers? Isn't this obvious poppycock? Look, I don't play chess beyond knowing the basic rules. Give me a link to one of those classic games. I will memorize it and proceed to play much weaker than 1400. So what do I have to do to actually get the 1400 that's owed to me? Study with a good teacher for years? Play 100 tournaments to "consolidate" the information that I learned in that one game? If so, how can it possibly be claimed that the game mattered?



Good lord, no! I did that when I was 30k, and it didn't make me instantly an SDK.

And drinking an extra glass of water a day isn't going to make me or anyone else look like Arnold Schwarzenegger, either.

This is the essential problem we amateurs face. We want to look like this:



...but our workouts are more like this:

Speak for yourself. I think most of us have no desire to look like that.

_________________
Still officially AGA 5d but I play so irregularly these days that I am probably only 3d or 4d over the board (but hopefully still 5d in terms of knowledge, theory and the ability to contribute).

Not wanting to pick on your post, but I have been a chess master for 30 years, and I think GM RAM is nonsense.

Sure, there is a finite number of positions you must know to be strong at chess, but I guarantee it is hugely in excess of 300. You need to know more than 300 specific endgame positions alone: this does not include openings (you need to know literally thousands of positions these days), middlegame (you need to know many typical positions, positional ideas, strategic themes, well over 300) or tactics (you need to be able to recognize many hundreds if not thousands of combinations).

Just by way of example, a Russian GM from years past, Paul Keres (who at one point was probably the strongest player in the world) wrote an excellent endgame book, Practical Chess Endings, intended to teach essential endgame knowledge to club players (i.e., below 2200 ELO). The book contains 333 "essential" endgame positions, and contains no fluff whatsoever.

Another example: when I was progressing from expert level to master, one of my main training methods was do to tactics problems from the first edition of the Encyclopedia of Middlegame Combinations. I went through this book many, many times, and it had over 1,800 tactics puzzles (like tsumego in Go) in it. At one point I could solve them all at sight; not just knowing the first move, but immediately seeing all of the mainlines and tries for both sides.

The opening knowledge needed to play chess successfully even at my level these days is simply immense. A barebones repertoire for both the White and Black pieces would fill a number of full-length books, and you would still need to do ongoing maintenance work.

The middlegame is harder to quantify because there is so much going on and so much you need to know. But one excellent book on just one aspect of the middlegame, The Positional Chess Handbook by Israel Gelfer, analyzes 495 positions. And this is only a portion of essential middlegame knowledge.

GM Ram is just another gimmicky book that understates the difficulty of getting strong in order to market itself. It's pure snake-oil. If you only needed to know 300 positions in order to be a GM there would be a lot of 10-year-old GMs out there. There are none.

I don't believe that there are any shortcuts to getting strong at Chess or Go. There are no magic 300 positions or even 1,000 or 2,000 positions. Play games against strong players, analyze them, work hard on tactics/tsumego, learn strategic ideas and practice applying them, learn some openings/joseki and how to play the typical positions that arise from them, work hard on removing the biggest errors from your endgame play, etc., etc. The most important part is probably playing serious games against strong players and analyzing them.