LiChess puzzle 94536 (from this game) begins with this position with black to play:

[fen "r1b2rk1/ppq1n1pp/3p4/3Ppp2/1bP5/3B1N2/1B1P1PPP/R2QR1K1 b - - 1 13"]

The fork 1...e4 is obvious, then white responds 2. Qb3 hitting black's b4 bishop. We need to save the bishop, and my candidate moves were as follows:

  1. 2...Ba5 which I eliminated after some analysis, since it allows a removal of defender tactic with Rxa5 at some point.
  2. 2...Qb6 which I quickly eliminated since it pins the e pawn.
  3. 2...a5
  4. 2...Bc5

I couldn't distinguish between 2...a5 and 2...Bc5, and I fortunately played the correct move 2...a5 (which I thought was slightly better since it guards the c3 square, preventing white's battery). The move 2...Bc5 is faulty because of the following:

black's dark-squared bishop becomes trapped via the moves 3. Bf1 exf3 4. d4! Bb6 5. c5! and black cannot capture on c5 since 5..dxc5 e6+ drops black's queen.

After seeing the solution, I feel there is no chance of me recognizing this in an actual game---I'd probably declare there to be virtually no distinction, and play a move to avoid wasting time. Maybe someone at a higher level than me might find it in a tournament game.

Question: At what level would a player be able to correctly analyze how ...a5 is superior to ...Bc5 if this puzzle arose in a tournament?

I'm trying to improve, and get an idea for how my improvement is going.

  • How much time are you giving yourself for this type of a puzzle "in a tournament game"? This is quite tricky to see, but I would expect an 1800 player to figure out the differences given 10+ minutes. – NoseKnowsAll Mar 27 '19 at 8:24
  • In a tournament game, you choose how much time to spend on each move: it’s not fixed, but the clock is ticking. As I say, when playing a tournament game, I wouldn’t recognize there is a distinction and just make a move to save time. (Note that I’m around 1800 FIDE currently and didn’t find it, even when presented as a puzzle where I know in advance there is a distinction.) – Rebecca J. Stones Mar 27 '19 at 12:46
  • @RebeccaJ.Stones: that's a good point, during a game you wouldn't know that there must be some big difference. It's an interesting question but I wonder how it can be answered, maybe by a strong coach who had a lot of different strength pupils. The crux is seeing that 4.d4 is an idea that means the calculation isn't finished yet. – RemcoGerlich Mar 27 '19 at 13:42
  • Also, if the fork cleanly wins a piece, black wins the game. There's no reason not to use a lot of time to double-check. – RemcoGerlich Mar 27 '19 at 13:44
  • In a real position, pretty much almost everybody will not figure out why Bc5 is bad because they will not bother to analyze it. a5 is clearly better without doing any calculation whatsoever. Therefore all their time will be spent on seeing if there is any refutation to a5 and considering you are winning a piece, you will not have to look for a better move. – Matthew Liu Mar 28 '19 at 19:04

Some things make it difficult....

  1. several motifs are involved: two clearances, trapped piece, discovered attack (and you have to see all them in order to understand the problem)
  2. the particular form of these motifs is non-standard (meaning there are more usual patterns for any of these motifs):

    • moving the bishop to f1 to clear the d pawn, is rare;
    • trapping the bishop with just one pawn (more common is a theme with a white pawn on b4 and a black pawn on c7);
    • clearing the diagonal by playing c5 (this is essential for the discovered check);
    • discovered check with a pawn is not very common either
  3. Moving one of the double-attacked pieces is not what one might come up first, because it often loses a tempo. One might be tempted to have an attitude like: anyway I'll lose one of the two pieces, perhaps I could make a useful move with one of the other pieces.

  4. One would have to calculate this defensive tactics from the black, i.e. from the attacker side which is for many people more tricky as (at least lower rated) players tend to do a lot more tactics that ask for winning/mating, etc than doing defensive themes.

  5. In a tournament game (unlike in a composed study) you never know whether there is one good move or several equal (or almost equal) moves.

Some things in favour of finding the correct move:

  1. Since e4 has the potential to win a piece and therefore the game, the black player in a tournament game would want to make extra sure and invest plenty of time to check any defensive options.

  2. The move a5 once and for all makes sure that the bishop on b4 is save, because it cannot be attacked by white pawns on that square and at most one would lose a pawn on any trade on b4 (after having checked that the pin along the a-file is not a problem).

Overall I think that because of the multiple non-standard motifs involved it is a very difficult problem in a tournament game. If you want a number, my estimate would be around 2200 ELO, but this might be more about how careful the player is than about what his rating is.

Given enough time (10+ min or so), I believe that a lower rated player of around 1800 should be able to find the solution if told that there is a unique solution.

Not the answer you're looking for? Browse other questions tagged or ask your own question.