The Flashiest Side of Chess

Introduction

What does it take to play a brilliant move? No, I'm not talking about the stupid !! that chess.com game review gives when you play an obvious tactical sequence. I'm talking about moves that would win a brilliancy prize, moves that embody the beauty of chess, moves like this:

You might recognize this position from Topalov - Shirov, which is covered in many endgame books. If you read Dvoretsky's account, you'll find that Bh3 has a logical explanation as well as concrete variations to back it up. But I'm not here to explain brilliant moves; how do you find them in the first place?

Candidate Moves

If you don't know what candidate moves are, it's basically a fancy way of saying "the moves that you consider". Typically, we are told to include forcing moves, improving moves, and prophylactic moves. However, it's extremely impractical to include every single move under this category; even considering every forcing move is often not possible, because in calculation, you can't just come up with candidate moves yourself and go down a long, linear, line; you have to come up with candidate moves for yourself and your opponent on every move of the variation, which is why it's not a good idea to calculate non-forced variations too long. Old engines had similar problems, having to consider every move possible, so they used a technique involving hash tables to trim the number of moves to a manageable amount. Occasionally, hash table collisions would cause an error, and the engine would make an erroneous assessment. Us humans have to trim as well; we can't just consider every possible forcing, improving, and prophylactic move, and most of the time, we trim away the brilliant moves.

The Problem with Finding Brilliant Moves

For the most part, finding a brilliant move requires you to spend a lot of time on the position, and it requires you to not trim away stupid-looking moves. For the most part, this is how people think when they are solving tactical puzzles, since they know that there's a tactical sequence to be found. However, it's impractical to do this in a game because you won't know when the brilliancy will be, and forgoing the trimming process and spending a lot of time on every move will just lose you the game. This is also why puzzles have limited usefulness, because you don't approach them in the same way that you approach a real chess game, unless you have incredible self-discipline. However, there is one situation where it's completely possible to find brilliant moves while approaching the game in a normal way. That isn't to say that no one does that; the late Emory Tate was well-known for finding brilliancy in mundane positions by seeking complications at every corner.

Necessity is the Mother of Invention

Frank Marshall spent years refining his famous Marshall gambit, and yet Capablanca managed to defend against it and win over the board. Though there isn't really a single "brilliant" move to pinpoint, Capabalanca played very accurately in the face of extensive preparation and a dangerous attack. When you are backed into a corner, with a lost or extremely uncomfortable position, that's when you find the most "brilliant" resources, as you have no option but to look at everything. I could give a lot of examples from top player chess or even my own games. Just look at Tal's brilliancies; most of the time, he is much worse positionally when he starts his crazy complications. Isn't it ironic? Our most brilliant moves are usually found in the worst situations. You will typically not have a brilliancy unless you threw the first half of the game or you play in a crazy and impractical way.