The Butterfly Effect

  • GM Gserper
  • | Mar 24, 2013

"Knowledge is power" famously pronounced Francis Bacon in 16th century. I doubt many people would argue with this statement since the more knowledge you have, the better for you, right? Well, almost. Today we'll discuss a very unusual situation where extra knowledge can actually hurt you in chess. I am going to talk about something I would call the 'butterfly effect' in chess.

The well known saying goes like this : A butterfly flaps its wings in Brazil we get a hurricane of the coast of Florida. I am not going to touch such a complicated subject as the chaos theory, but the point is that a tiny, almost irrelevant change in one place can lead to a humongous difference in some other distant place with no visible connection between two of them. How it works in chess? Let's consider the famous Legals' mate position:

If you've just solved this little puzzle it is great, if you knew it, then it is even better. Now suppose you got an almost identical position in your game with only difference is that Black pawn moved from the a7 to a6, what would you do?

Since the whole White's combination takes place in the center and the King's Side, who cares about that little change on the Black's Queen's Side? After all the little Black pawnie is as far from the action as it is just possible, so we use our knowledge and just execute the combination,right?  Wrong! Try to figure out why the famous combo does NOT work now.

Here is another trap which became famous after the next game:

A position from the next game looks pretty similar to the Capablanca's trap, so should we just follow the footsteps of the great Cuban and trap White's Bb3?  Black decided to take the bait and immediately regretted his decision.  Can you figure out what happened there?

So, did the knowledge of the classical trap ruin Black's game?  Yes and no.  Yes, because if Black didn't know the Capablanca's trap, he would probably never go for it.  But from the other side, let me quote the famous phrase (and by the way, even though it is technically correct, I still hate it!): "Guns don't kill people, people do".  So, we cannot blame chess knowledge for such defeats, we can only blame people who use their knowledge blindly. Yes, chess knowledge, just like any knowledge is obviously a wonderful thing, just use your own head before you apply your knowledge!


  • 3 years ago



  • 3 years ago


    Butter fly effect! This in fact is the Style of The  great Cuban Capablanka.

  • 3 years ago


    excellent article!

  • 3 years ago


    liked it. thanks! :)

  • 3 years ago


    Good article. Explains why most amateurs don't really improve at chess. If you don't have the talent, the intuition, you can kiss chess goodbye.

  • 3 years ago


    Can't you do the same thing in the first puzzle by eating the knight and bishop? You don't need the other pawn in the end do you? Or is it because of tempo? Because the both lead to the Bh5 after white does pawn h3

  • 3 years ago


    At ultras: a6 prevents Qb5+ to get back the knight.

  • 3 years ago


    Thanks, i-c-k-e, but I was talking about the black knight taking the white knight in game 1 (instead of black bishop taking the white queen). Still no checkmate... Thanks anyway...

  • 3 years ago


    all right!

  • 3 years ago


    Nice article and thanx for the info.

  • 4 years ago


    ...or not thank you for the knowledge?

  • 4 years ago


    Thanks for the knowledge!

  • 4 years ago


    @jcm1978: Thanks! That's what I get for trying to study diligently on my phone...Embarassed. I forgot the game was posted by a GM and not a novice.

  • 4 years ago


    I have disagree with part of the premise here. Knowledge did not lead to anyone's demise. Rather, poor calculation and misreading the position did that.

    I, as a low rated player know that knights move in an "L" shape and can fork pieces. So forking king and queen is awesome knowledge to have! Forking king, queen and another knight :'( doesn't work to well. That didn't happen because I had too much knowlege lol.

  • 4 years ago


    For Mark vs. Halasz, wouldn't Queen to d5 be a better move than bishop to f7 as the first move in the puzzle.

  • 4 years ago


    @ananga1 the queen check can be blocked by Be7 which protects the rook in Steiner vs Capablanca, but going back to the 2nd game this doesn't work as a defense because the follow up move to Be7 is Ng5+ forking king and bishop 

  • 4 years ago


    Thank you for the arctical

  • 4 years ago


    Loved it!, thank you grand master

  • 4 years ago



    If there is no pawn on a6 :

    6. Nxe5! Nxe5  7. Qxh5 Nxc4  8. Qb5+! to take the knight and win a pawn in the exchange.

    if there is a pawn on a6 :

    6. Nxe5! Nxe5  7. Qxh5 Nxc4 and you just give one piece to the opponent and probably the game : the b5 square is protected so you cannot take back the knight with the fork.

  • 4 years ago


    In Diesen-Halasz 1973 Black can try 10...Be6!? 11.Ng5+ Qxg5 12.Qxg5 Be7.  He is heavily down in material but the unbalanced material gives him faint survival chances and he can complete his development.  In the game line, Black is the exchange down with King stuck in the centre. 

Back to Top

Post your reply: