Stepping Stones


On holiday, so only had the chance to put up a post today.
One of the best ideas I ever came across in chess literature was Jonathan Tisdall’s idea of Stepping Stones presented in his only chess book, Improve Your Chess Now! Like most other great ideas in chess literature, it is a verbalisation of something strong players were already doing. But rather than being something that you can just come across while training/playing and start using some of the time randomly, putting it on the page makes it something simple and concrete that you can train.
The basic idea is that you find out which positions you need to pay extra attention to in your calculation, and fix the position in your head so that you can approach it in the same way you approach the initial position.
Last night I gave a pro bono lecture in Copenhagen with all income going to a local chess club. We discussed candidate moves a lot and the way that compounding of candidate moves is a great part of calculation.
Take for example the decisive moment in Kramnik – Aronian in the Candidates tournament in London. Kramnik thought for a long time before taking on d5 with the queen, trying to squeeze an advantage out of the endgame (unsuccessfully).


It was quickly established with computers that Kramnik could have won with a strong punch on the light squares: 24.Rxc8 Rxc8 25.Qxf5 Rc7 26.e6! Black would have no way to deal with this other than 26…fxe6, when White plays 27.Qh5! with ideas such as 27…g6 28.Qe5! hitting the rook and preparing Bh6, winning. For example 28…Rd7 29.Bh6 Bd6 30.Qxd6! and Black is mated.
I am sure that Kramnik saw all of this rather quickly. But why then did he not play this and win the game? Was he afraid of risks? We know from looking at his other games, that this strange attempt of pocket psychology is nonsense. There might be some preference for technical advantages, but not so much that he does not calculate attacking lines.
It took me some time to work out what it was Kramnik had missed, and I had the chance to test it with a 2600 student, who knew the position, knew that White was winning, but could not refute one move, the move I suspected was difficult to refute, 27…Qb5!.
This is where the technique of stepping stones can be useful. If Kramnik had put this position in his mind and taken the time to do candidate moves in it, he would have won the game:

When I put this position on the board at the lecture, it only took seconds before a few people found the winning move. Still, very strong players have failed to find this win – one of them even knowing that there was a win!

I will put the solution in the comments later. The main point has been made. Train stepping stones and increase your level of calculation greatly.

23 thoughts on “Stepping Stones”

  1. @Jacob
    Thanks for taking time to post such an interesting thing during your holidays!
    I think I found it so I won’t write the solution here and will check my variations against yours later :p

  2. 28.Be4! is the important winning move. Why this is so hard to tell is a debate and probably different from person to person, but it is definitely because of …dxe4 – which does not work after all…
    The main line is 28…h6 29.Bxh6! Bc5!? 30.Bh7+! Kh8 31.Bxg7+! winning.

  3. There is one problem with chess problems compared to actual games. I have noticed, in “Grandmaster Preparation: Positional Play” that sometimes, the obvious move that I find in under 10 seconds is the right move. But in most other books and Websites, the “obvious” move is never the solution, and I mean “NEVER!”.

    That’s the one downfall I see to chess problems. I have occasionally carried that habit to my over the board games, and when the move is right there in front of your eyes, obvious as can be, while I might ultimately play it, I go in the tank for like, 15 minutes, trying to figure out what’s wrong with it.

    Like the last poster, I won’t disclose my answer so that others can figure it out, but if I’m right, it literally took me 5 seconds to find the move, and about another 30 to 60 seconds to confirm my answer! This probably means my answer is wrong! (See what I mean? LOL!)

  4. I agree with Phille… The most “annoying” thing when I solve problems at Quite often, it’s just “simply take the queen and you’re up material”, but I never even consider those moves as I’m looking for “fancy” solutions and neglect the obvious ones.

    I also found Be4 very quickly, though I stopped after Bxh6 (didn’t consider Bc5). I would have never calculated the variation from the original position that precisely. I might have played up to 26. e6! out of pure luck, though 🙂

  5. When I do puzzles, I try to make them useful for the practical player, more than necessarily surprising or fancy. With Positional Play, one of the points of the book is that the exercises are meant to be 1-move solutions with no complications (in general, not all of course), exposing weaknesses in your intuition if you get them wrong…

  6. Concerning the theme of stepping stones, I wonder if a good start is
    to focus on a position where the tree of variations is not so
    wide, then try to calculate as far as possible and set up the
    resulting position.

  7. I think (at least this happens to me) the difficulty in this particular example is the pin on the Qb5, which is a new motive compared to the previous positions. That makes it difficult to see imo.

  8. I knew this method (read Tisdall’s book years ago, which is very good by the way), but somehow fail to use it consciously in my games… Thanks for the reminder!

  9. @kieran
    Or you can take positions where the same theme is the same to start with. For example the Greek Gift sacrifices in Attacking Manual 2, though training material for this does not really exist.

    Another idea is to look at the puzzle position and then turn around and keep it in your head. Analyse it, write down your lines and turn around and go through them looking at the board.

  10. @Jacob Aagaard
    I often solve puzzles like that anyway. I try for a few minutes without success, but then the position is in my head and I solve it later on, in a quite moment. Or the solution suddenly pops up, while I’m doing something else.

  11. My biggest problems are connected with such a “block” when looking at some variations. I am able to see 3-4 moves ahead and evaluate the variation quite correctly, but when I see the other variation and some unexpected move (like shown at the diagram 27…Qb5!) appears… I cannot find the refutation. What are the best ways to overcome this kind of “blindness” Jacob? Could you provide any hints? Thanks in advance – your examples and methods of training are really helpful to me (even if I disagree with part of them).

  12. Very nice ! It is also very difficult : he who can figure clearly the second position just looking from the first diagram certainly has GM level. In the first position, the idea of Be4 linked with a checkmate on h7 looks surrealist. Vlad, you are not walking alone…

  13. @Tomasz Chessthinker
    I think practicing stepping stones will be good training for this. And then to remember that this is the big problem most chess professionals struggle with – because at times the good defensive move is really just a party stopper.

  14. A way I could find them is as follows: first identify the weak spots, the objects of attack. Clearly white’s main hitting potiential is on the white squares since only the Kn and the Q are defenders and there is a lot to attack on them: d5, f5 f7 and h7, Ra8, Rc8. Blacks K-side looks weak since its white squares are only defended by the Knight. Therefore the plan should be: attack blacks kingside on the white squares. Ok. Which moves come in handy? Qf5 ofcourse, excehange onRc8 first to gain a tempo on the unprotected rook on c8. Rc7 is natural to guard he 7-th. Then what. We are looking for candidate moves to attack the white squares. Naturally e6 is such a move. So we start to analyze. fe6 Qh5 eyes e8 en threatens to win a piece there. g6 fails to Qe5 and Bh6. Nice! What else. Wait! Qb5 covers e8. But the Q is hanging there, which I can use to get my last piece that is not contributing into the attack for free with Be4. This must be the move, now we have all pieces in the attack and black must weaken its king protection even further. I have only invested two pawns (one doubled and one backward) and have a dangerous attack. I think I would stop analysing after Be4! and evaluate that position as clearly better for white.

  15. @Indra Polak
    This sounds very much like the story you would tell after, when in reality this is not what I think is going on when you are looking at the position. Sorry to be a spoil sport. There are also other narratives, such as include all the pieces in the attack.

    I think I will write an article about this for Monday to give my view. Obviously I am not sure you would not find the position like this, but I think there are questions there you would not ask yourself during a game, unless you knew what the solution was :-).


    🙁 MY REVIEW: Pump Up Your Rating by Axel Smith 🙁

    According to ChessCafe rating system: *** stars or „Uneven“ book

    Part I Positional Chess

    Chapter: 1.No Pawn Lever – No Plan
    – pawn levers are only one of positional factors, so I don’t see big emphasis on this theme
    – he omits to solve perennial question: how to improve in positional chess, which books or positions to use (those who don’t know how read Hellsten’s Mastering Chess Strategy)
    – moreover he doesn’t mention any method

    Chapter: 2 Fair Exchange is No Robbery
    – as said before, this is only one of positional factors, but good that he makes accent on this cause many go astray when exchanging pieces

    Chapter: 3 Auxiliary Questions
    – somewhat odd questions to ask ourselves
    – I use Dvoretsky’s algorithm and Aagaard’s from GM Preparation Positional Chess

    Chapter: 4 Calculation
    – general guidance is not merged into effective formula like in Jussupow’s „Tigersprung“ series
    – too many phases, too much theory with less effort at the end

    Part II A Training Program

    5 The List of Mistakes
    – good chapter

    6 The Woodpecker Method
    – excellent treatment of Tikannen method, but nothing new (compare De la Maza and others)

    7 Openings
    – specialized overview of Nf3 system
    – he doesn’t cover why and which opening to choose as amateur, semi-professional and professional
    – where is detailed explanation of major openings regarding structures, dynamics, theoretical burden etc
    – how to build fluid and coherent repertoire? nothing about that!

    8 Theoretical Endgames
    – good chapter

    9 Attitude
    – be warned: don’t be unfair player who uses tricks when he wants to win
    – don’t quit university for chess, it’s biggest mistake in your life!!!

    Publisher, where are endgames is PGN for download?

    SUMMARY: for novices who don’t know how to train this is excellent book, for others nothing new under the sun. Read Irina Mikhailova’s “Modern methods in training chess players”.

    I’ve also just finished the book but really enjoyed it and would give it 4-5/6. In general, the book is very entertaining and Axel is extremely enthusiastic. Especially the chapters 1,5,6 and 8 were excellent. I definitely see your points but there are also several of the books to which you refer which I don’t know!

    In my opinion, the biggest weakness is that we aren’t really told anything about the target group (strength of the readers). Being a relatively weak player with 1946, I feel the chapter on openings is at least for extremely ambitious players, and in chapter 6 he keeps on using 2300 as example when time limits are set for exercises; if you look at the cover of the book, you wouldn’t think that this is the typical reader I assume! One of my friends who is 1500, won’t learn very much from it I suppose and he bought it immediately when it was published.

    I hope that the book will be a success because the author has really worked hard on this book and it is very personal.

  18. I disagree with le bruit qui court
    The book is clearly a work of love and very inspiring.
    Of course nobody can write a chapter of complete positional chess
    But several ideas are quite interesting. There is no magic formula
    For improving but this book show a way to do so.
    I think that also lower rated players can learn a lot, even the chapter
    On openings can be usefull, they need not the amount of detail but a little bit of
    Own theory hepls against higher rated players

    I like this book

  19. @John Hartmann
    Positional Play is easiest, then probably Calculation, Attack & Defence and finally Strategy. Edngame Play will be on par with Calcultion and Thinking Inside the Box will not be an exercise book, so probably easiest of all :-).

  20. An Ordinary Chessplayer

    neiman :
    … he who can figure clearly the second position just looking from the first diagram certainly has GM level. …

    False. My highest rating ever was around 2300 FIDE and in around two minutes I was able to visualise the position and settle on Be4. Many candidate-master strength players like me can do the same, Soltis has made this point in his writings. There are many skills required for GM chess, I prefer not to reveal precisely which are my weakness(es). But if you see me thinking a long time in a chess position, it might be because I am trying to count the pieces for each side!
    Back when I was a club player working through Informator’s Encyclopedia of Chess Middlegames Combinations, I called this calculation technique “staging”. Recently looking through Alburt’s Chess Training Pocket Guide, I saw that he called this a “support position” (something like that) without explaining the term. The Massachusetts (USA) legend John Curdo called it “look-ahead”, which is a little imprecise but he probably liked that. If you can stage like this, there is no theoretical maximum to the longest single variation you can calculate, but the clock sets a practical limit.
    As to candidates, I looked at them in roughly this order: Bf1, (illegal so…) Rf2, Bxd5, Bf4, Bh3, Be4 and just stopped generating candidates there, this move is too obviously strong to bother with any others (1… g6 2. Bxg6! or 1… h6 2 Bxh6!, although in a tournament game I would hesitate a bit). I don’t use the proper Kotov method of generating all the candidates first before calculating, instead I look at the black countermove as well before moving to the next candidate. No long variations though.
    I think one real GM skill is knowing when to continue calculating a forcing sequence, versus when to stop, stage, and generate candidates. That is, identify the critical positions.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top