The Calabrian Sacrifice

Oct 10, 2015, 4:03 PM |
I should start by saying that I am not a good chess player. I am enthusiastic, but certainly won't be threatening Carlsen any time soon! However, this means I get to play some rather satisfying combinations (for me anyway) that would certainly never been seen at a higher level. Perhaps one explains the other. One of my favourites is the Calabrian Sacrifice.
I first stumbled accross it in an online game a few months ago.

I'm certainly not laying claim to Bxh7+! but after succeeding with this move in subsequent games, I view it as a significant combination that is overlooked by some players at a similar standard to myself. Georges Renaud and Victor Kahn (2015, p63)  believe the Calabrian to be "brought about by a series of moves which seems so abnormal to the modern player that we suspect [they] had been composed". Despite this, I want to explore why it has worked for me and what this may reveal about the psychology of players at my level. But first, a little background to the Calabrian. In 1619 Greco published the following game:
This position represents the ideal outcome of the Calabrian Sacrifice, rather than a watertight method of achieving it. Black has three nontrivial responses to Ng5+ as follows:

    1. Kg8 (hiding the King).

        In all three of these variations, white forces mate.

    2. Bxg5 (taking the knight)

        Once more black cannot escape checkmate.
    3. Kg6 (The king goes for a wander)
Although black avoids immediate mate, white has traded a knight and a bishop for a queen and a pawn.
For the Calabrian to have chance of success white needs to be in the following position:
  • the light squared bishop on the b1-h7 diagonal to sacrifice itself on h7 with check;
  • the dark squared bishop on the c1-h6 diagonal (but not on h5 or h6) to defend the knight on h5, and allow the possibility of discovered check if the king is on h6;
  • the queen on the d1-h5 diagonal, prepared to leap into the attack on h5;
  • the knight able to reach g5 with check after Bxh7+;
  • a pawn on e5, controlling f6;
  • additionally, if black still has the dark squared bishop on e7 or f6 there need to be a pawn on h4 defended by a rook.
The last requirement is not strictly necessary as I found out in this game:
At this point I will reiterate that I am not a good chess player. I happily acknowledge that in both of the games I have presented to far I benefitted from my opponents' blunders. Yet I still believe in the effectiveness of Bh7+ for the following reason. In what appears to be a relatively familiar, usually docile opening the bishop sacrifice comes as a complete shock when black does not anticipate it. Suddenly black realises that all of white's pieces are pointing directly at the now vulnerable king. Moreover, after the pugilistic Ng5+ and Qh5 black is reeling. The mind homes in on the squares immediately surrounding the king, and black tries anything that will block or chase away the queen. The abrupt shift from balanced game to onslaught induces mistakes. The next game perfectly demonstrates this. Having manouevred into a position where I could unleash my favourite sacrifice I couldn't wait to play the (awful!) Bxh7+:
Why have I waffled on about a few games I have lucked into winning? The Calabrian sacrifice is a useful weapon to have as white, and important to be aware of as black. I think there are three main reasons why. Firstly the reason I have been able to set up my pieces in preparation for Bxh7+ is that I assume black has not seen the sacrifice in a while, and at first glance dismisses the trade of white's bishop for black's h-pawn as unsound. In the spirit that an ounce of prevention is worth a pound of cure being aware of the possibility of the Calabrian allows black to defend against it. For example, the simple h6 stops Nh5+ as well as giving black some luft. I love opening with the Queen's Gambit, and the postitions I have analysed have all arisen when black as played e7-e6 during development. e7-e5 makes white work incredibly hard place a white pawn on e5 and kick the knight on f6 away. Fianchettoing the bishop on the kingside is also effective. Secondly, the psychological pressure of the attack lures black into making errors. I have read numerous times that one should evaluate the position objectively, and take time and care to calculate possible moves accurately. But as a budding chess player the memories of fatal back rank attacks and unforseen checkmates are as fresh as they are painful. It is overwhelmingly tempting to react instinctively. I'm not saying one should rely on an opponent's mistakes, but the shock and awe aspect of the Calabrian should not be discounted. This is particularly true in speed chess. Finally, even the threat of Bxh7+ can be useful in forcing black to commit too many pieces to the defence of the kingside.

By way of illustrating the second point, I hope you'll indulge me in sharing one more game with you that I am quite proud of. The time control was 1 minute each and I achieved checkmate with about 20 seconds still on the clock.

In this 1 minute game it was very efficient to have a setup (the Calabrian) to aim for. I wasted little time in the opening moves. The attack and the time pressure forced black into numerous mistakes and although I made some myself, I at worst would have forgone mate to capture the Queen.

I must end by reminding you that I am not a good chess player. Nevertheless, I hope you enjoyed my take on the psychology of Calabrian and are able to use it in some of your games! If nothing else it is a very pretty and immensely satisfying combination to study. Please post any interesting Calabrian games (or other cool sacrifices) in the comments. I look forward to learning from them!

Renaud, G., and Kahn, V., 2015. The Art of Checkmate. London: Batsford (a lovely book bursting with interesting mates)
Silman, J., 1998. The Complete Book of Chess Strategy. Los Angeles: Siles Press (this book is fantastic! Thank you Mr Silman)
The Computer Analysis engine (it amuses me greatly that the moves with which I am happiest are invariably the ones that disgust the computer the most)