I promised a long time ago to give some sort of explanation about what I mentioned at some point as understanding the mechanics of chess. It is not so easy to do, but when I mention it to fellow grandmasters, they seem to understand what I am talking about.
It has a lot to do with understanding where the pieces belong. Understanding sequences and how they relate to each other.
To give an example:
Here the main line is 6…Qe7 7.Qe2 Nd5 and so on.
I was talking to a GM yesterday that mentioned this as being an example of a pattern of moves that can be used in several situations. He immediately mentioned two:
Apparently this is popular among juniors in the US. A lot of knight moves are usually played, but it White plays 4.Qe2! then after 4…Qe7 5.Nd4, he is fighting for an advantage (with an extra pawn as a secret weapon).
Famously in Miles – Christiansen the players had agreed a draw in advance and were just making moves here. Miles polished the e2-square before eventually playing the harmless 6.Nxe4 and after a while the game was drawn. It was inserted into Chess Informant without annotations and Viswanathan Anand used it as Black against Zapata. After 6.Qe2! Black will lose a piece as 6…Qe7 is met by 7.Nd5. Anand resigned.
This is just one example. Another is this: giving up a pawn to get the d5-square for a knight. Here they are all tactically based (because it is fun), but they need not be.
21.d6 exd6 22.Nd5 Rxe1 23.Rxe1 Qxc4 24.Bxd7 Nc5 25.Bc6 Qxa2 26.Qh6 1–0 Karjakin – Karttunen, Men Kemer 2007
25.d6 exd6 26.Bxf6 Bxf6 27.Nd5 Qxb2 28.Rd2 Qe5 29.Qxe5 Bxe5 30.Rxe5 dxe5 31.Nf6+ Kg7 32.Rxd7 1–0 Volkov – Sohib, Dubai 2012
14.d6 cxd6 15.Nd5 Nxd5 16.Bxh8 Nb4 17.Qc3+ Nc6 18.Bg7 and White won.Zhao Zong Yuan – Fitzpatrick, Tweed Heads 2009
There are lots of these small tactical, thematic patterns. Thousands. And together they create a foundation for understanding the mechanics of chess. You get a sense of how the pieces relate to each other, where they want to be placed and which pieces you need to exchange.
Unfortunately I do not know of any short cuts to mastering the mechanics of chess; you pick it up as you go along. Probably the main thing I would mention is the Yusupov books and the Grandmaster Preparation series. But it is always like this.
A small additional point
Just because I am a big believer in strategic patterns (a better word than rules) I am weary of those who want to teach chess without the inclusion of a lot of practical examples. I am not sure this abstract knowledge is that beneficial without a concrete reference. On the other hand; I do not believe that concrete examples without pointing to the underlying strategic patters is ideal either. Use all knowledge and ideas; this is my method.
Another ‘pattern’ I noticed is moving a Knight to G5 only to swing it back to E4 to exchange off the black knight on F6. Normally done with a knight on C3 in-case black takes. I learned that one in the Karpov-Kamsky book. It has some very interesting observations like this where Karpov shares these typical strategies/maneuvers.
Dear Jacob,
Thanks for the post. My belief in serendipity (one of the really beautiful word in English) is confirmed yet again. I was just last week reading a book by WGM Irina Mikhaylova, “Thinking in Schemes-A Strategy of the Champions” (2012) in which she deals with something similar (??) (I think so). Here is the link to her webpage (http://www.chessy.ru/page1_2.html) and book and link to Introduction to “thinking in schemes” (http://www.chessy.ru/UserSite/www.chessy.ru/pub/Ts-122-1.pdf).
What do you think? Similar to what you mean?
In general terms I see two motivs: A) pin; B) field clearance. Both can be found in any basic book on combinations. In short: these two motivs direct the moves, creating a sequence. Now you can either remember the sequence, or you can remember the motivs and deduce the sequence from the motiv. I have been taught to remember the motivs, because these are more general and can be used in more circumstances. For instance when you mention the “mechanic” “giving up a pawn to get the d5-square for a knight” the specific details of the d5 square and the piece being a knight is in my opinion not really part of the used chess knowledge.
But maybe I am missing your point entirely :).
@weng
I think Mikhailova’s book is similar to Jacob’s book “Positional Play,” although I found the problems very difficult, so I have set it aside for future use. I didn’t know there was an updated version. Is it in English? I have the Russian version from 2008.
@Indra Polak
The more difficult the sequence, the more difficult to deduce. Let’s create a math analogy. There are only 4 basic operation types. Still it can be helpful to know how to differentiate.
There is also another point to it- if you know the sequence, you probably also know the outcome. Knowing the subsequent position can be helpful to visualize things further down the line. Moreover it safes time, because you aren’t forced to deduce.
Well I must admit that some sequences I simply “see” and the deducing part is not really there. Maybe the pattern matching takes over after having seen the sequence often enough.
I agree with Indra Polak. A common scenario is the “poisoned” d4-pawn due to a check (most commonly Bb5+, Bg6+, or in the case of a castled king, Bh7+). The most common example I can think of is in the Milner-Barry Gambit (1.e4 e6 2.d4 d5 3.e5 c5 4.c3 Nc6 5.Nf3 Qb6 6.Bd3 cxd4 7.cxd4 Nxd4? (7…Bd7 is correct, and if White castles on move 8, now you grab!) 8.Nxd4 Qxd4?? 9.Bb5+.
Like Indra Polak mentions, once you’ve see the pattern numerous times, there is no deduction to the process. It’s pretty simple, no need to calculate once you know it. If the Bishop has a check, the pawn is poisoned.
Another common theme is trapped Bishops and Knights on the 5th rank, netting a pawn, in cases where White can “sacrifice” the Bishop on h7 if the King is the only way to capture back, and then execute a fork with the Queen on h5.
I think what it amounts to is whether you are talking “standard” sequences or “one-off” scenarios. It’s like comparing the 15 “Essential Chess Sacrifices” (book from 2003 by Gambit) that are standard and common versus some one-off scenario where maybe a Knight sacrifices itself on a6 for a pawn (not common). Scenarios like the latter (the rarer, one-time scenarios) are where deduction is necessary, unlike say, the Greek Gift Sacrifice, where it’s more “general rules” that the sacrificer must ask himself, like can a Bishop get to f5 for defense, or does Black control g3 giving White a problem after …f5 once the Queen goes to g4 to threaten a discovery on the Black King on g6, or can the King run without material loss.
So if you are “simply seeing” and not “deducing”, hopefully these are your stock scenarios and not “laziness” in more complicated situations that are very specific to the position at hand.
In the scenario above where Anand lost in the opening (after 6.Qe2!), this looks like the 5.Nc3 Petroff (assuming the “last” move made was 5…Bf5??), and if that’s correct, I’m shocked that a GM wouldn’t know that 5…Nxc3 is played for a reason other than randomness!
@weng
I have to admit I have not read her book. I do not have time to read everything; sadly.
But I am not only talking about pattern recognition; it is also about a general feeling for the rhythm of chess as well as the pieces interrelations.
@Jacob, Thanks for your reply. I understand about time and not enough of it to read everything and anything. I also agree that this mechanics of chess is the heart and soul of chess. If I can use a different metaphor (and not abusing it), it is like the dao/tao or zen of chess. However, without wishing to sound too mystical, as someone who is himself trying to learn chess as well as wishing to learn how to help others learn chess, we have to “analyse” ie break it down, and like you say, the mechanics of it.
I think there are two things/ideas I wish to link/connect with here:
1) Following from the recent books on excellence and the debate of talent vs practice, there is undeniably a state of affairs where one achieves the “unconscious competence”, which is I think akin to the “gen feeling for rhythm as well as piece interrelations” or My mystical “dao/tao/zen/way”.
2) In that case, research suggest that the “way” to get to “unconscious competence” is via deliberate practice and lots of it. Mind you, it is not just “practice” but Ericksson’s deliberate practice. Here is where I think the guidance of a coach/trainer is invaluable!
3) If we wish to classify these notions (classification is a std analytical tool to achieve some semblance of control), then perhaps the following schemata may be a useful starting point?
Opening>Variation>Sub-Variation>Tabiya>pawn structures?
Opening>Variation>Sub-Variation>Tabiya>priyomes?schemes of thinking?Understanding sequences and how they relate to each other and where the pieces belong.
4) A consequence or corollary of Jacob’s point and the above is that there has been in the past few years (since I have started noticing) comments by GMs (especially those trained in the Soviet system) that such and such is an “ugly” move or when asked about a move, responds “I do not even think about such a move!” (I think it was Tiviakov who said this).
5) This can be further extended into the human vs engine style of play and especially the criticism of the younger generation which grew up with chess engines as their coach, that their play is engine-like.
6) I found WGM Irina’s thesis of some interest because I think she is one of the first to quantify chess expertise (measured in terms of Elo and titles (IM and GM)) (see her short article which is available on ChessOK (Convekta) website). Not saying that her figures are correct but it does provide a series measurable goals/targets for anyone whos wishes to embark on a training regime to achieve an IM or GM title, which is not just based on the undefined “chess knowledge” (you know it when you see it).
I do apologise in advance if the above is out of line or outside of the topic nor is the above intended to make any commercial endorsements.
@weng
I am not afraid of someone mentioning other people’s products here, when debating the issues put forward. I think it is unpleasant when someone is straight advertising a competing product, but it only happens once or twice a year.
The “ugly” move debate is quite interesting. But we should in general keep an open mind. I actually wanted to write something about “computer moves” after the debate on Attack & Defence. I might have one or two simple points to make…
I do fear that I have not fully reached a level of conscious unconscious competence of the mechanics of chess; which I feel I have of other areas of the game. Or in other words, I am not very good at explaining this subject, concerning the interrelation of the pieces yet.
One of my students said about pieces “sometimes they dance”. That is hard to fathom from a 13 year old, but there it is…
This all is pattern recognition in psychology for many students. Take the first three examples. There are common features: the half-open e-line, the uncastled king, the pawn (or knight) of the color having moved, the queen on the board of the side to move. Belonging to this features are tools or operations, like attacking the pawn (or knight) with the queen move.
Thinking a lot about this I think this explains how important it is to have fitting training material according to your knowledge or skill level. We do this work of categorizing the features and adding the operations to the features automatically in all areas of life. If a chess position was too difficult it is important to find, what the stumbling block was. Finding it means knowing better what to train (repeat) till you find the solution automatically.
If you like to read some scientific psychological literature about this take
De Groot – Thought and choice in chess
and the articles of Simon and Chase.
Hard stuff, but really great stuff by great thinkers. Easier is Chess for zebras by Rowson, his 2nd book, as he integrated the cognitive science in a great manner.
I found the De Groot book a bit tough going and I am not sure I truly agreed with all the conclusions; but it is a long time I looked at it. I definitely did not agree with a lot of Rowson’s stuff. We read the same books at university and understood them differently. In general I am worried about this thing of taking the local and making it into universal principles. For example: because we can see that pattern recognition is happening when we play chess, chess is all about pattern recognition. I know a lot of science makes such jumps at time, but often it is just wishful thinking. (This is not particularly meant towards Jonathan’s book, which I am not wanting to debate deeply here. I think Jonathan has a very unique way of seeing things and that makes Zebra’s a bit off the mark for me. On the other hand I think 7 Deadly Chess Sins is a Masterpiece.)
### ERROR FOUND ###
In the book “Build up your Chess 1” by Artur Yusupov, lesson: 17. Stalemate motifs, I found an error in exercise section.
Namely in position Ex. 17-2, on page 168, Yusupov claims that after 1.Bg4 white draws.
Really? My engine Komodo TCEC 2013 World Computer Chess Engine Champion Edition claims that Black has mate in 10 moves 🙂
Check it by yourself, here is FEN: 5k2/3B2bK/6P1/4n3/8/6P1/8/8 w – – 0 1
1. -+ (-#10): 1.Bg4 Nc6 2.Bd7 Nd4 3.Bg4 Nb5 4.Bd7 Nc3 5.Bc6 Ne2 6.Be4 Nxg3 7.Bf3 Nf5 8.Bg4 Nd4 9.Bd7 Nf3 10.Bc8 Ng5#
So Black doesn’t take the Bishop on g4, he mates instead 🙂
Editors?
I remember being concerned about moves other than 1…Nxg4 when I did that exercise too, but chose 1.Bg4 anyway because it clearly fit the theme of the chapter. Looks like RĂ©ti’s been cooked.
Maybe QC should start a special area up for all mistakes found by readers and publish them?
@LE BRUIT QUI COURT
We will check it.
@Michael Bartlett
No. But if people contact us, we keep corrections on file and put them in future editions.
@Jacob Aagaard
DeGroot IS tough stuff!
Take the thing and thinking with pattern recognition the other way round. Pattern recognition is a model about something universally happening and chess is the local. Plus: Try to fit the models to yourself. Quote: “I don’t think like a tree.” Wasn’t it Rowson writing this down?
@Jupp53
I remember “I don’t think like a tree” being Jonathan Tisdall in “Improve Your Chess Now” but maybe Jonathan Rowson said it too.
Anatoly Lein might be the source of that “I don’t think like a tree” quote.
It seems to me that this pattern recognition thing is just a reflection of what we have learned or have met before, whether we could draw a conclusion directly from it is still in doubt. Chess is in any way still a concrete game, we must always depend on ourselves, try to see every good moves on the board. But I think there’s another thing which hurts my playing strength very much. Just let me call it the rhythm of chess. Sometimes we have to play on both sides and it’s very hard for me to decide the correct decision there, to attack or to consolidate a little bit, like for example when to play Kb1 in a opposite castling position, or the right time to push White pawns to h4, g5, h5 while Black is trying to throw every piece on my face in a Najdorf line. Anyway, I’m an amateur and it’s just my own thought. Maybe it’s just the same thing. And It’s still a long way for me to really understand the mechanism of chess.