Getting More Out of Puzzles
Suggestions for making puzzles translate to real games
As I work on wrapping up my book on evaluation (very close, I swear!!!) one question I’ve gotten is, “Why do I need to evaluate? Can’t I just compare the moves to each other and choose the best one?”
It’s a very reasonable question, but in practice I find that evaluation impacts the whole thought process. There’s a big difference in the mentality between being in a dominant position or a desperate one. It impacts what sorts of moves you look at, how ambitious you have a right to be, how long to take on the clock - really every part of the decision.
This week I saw two positions on Twitter that illustrate the impact of evaluation on decision-making.
To give you some time to think about your answer before seeing the solution, here’s a picture of Bobby Fischer.
The best move, as most people got on Twitter, is 1. f6! starting an unstoppable attack on Black’s king. But most people gave this move with no further variations to support it. Of course, you probably don’t put as much thought into a Twitter puzzle as your tournament game, but I’d argue that playing f6 in this position without calculating is a huge thought process mistake, even though it turns out to be a good move.
To understand why, let's take a step back and evaluate the starting position. This is a huge difference between puzzles and games: in a game, you’ve already been playing for many moves and potentially many hours, so you should already have a pretty good idea of what’s going on in the position. Who’s winning? What trades would be beneficial for you? What about pawn breaks? What is your opponent up to? Hopefully you already have pretty good answers to these questions!
But in a puzzle you’re dropped into an unfamiliar position with no context, so the first step is to orient yourself. So what’s going on in this position? In fact, White is crushing, and it all comes down to the knight on d5 vs. the bishop on d8. White’s knight is occupying an ideal outpost in the center of the board and the dark-squared bishop can never challenge it. The difference between these pieces is so huge that it defines the entire position. White can combine play on both sides of the board, while Black has no counterplay whatsoever. In the card game Magic there’s a strategic concept called inevitability. Having inevitability means that in the long run, if nothing weird happens, you should win. In this position White has inevitability.
Okay, let’s return to 1. f6. After 1… Bxf6 you’ll be more or less forced to trade the knight for the bishop. Remember, the knight is so much stronger than the bishop that it basically guarantees a winning position for as long as they both remain on the board, so this is a terrible trade. Does this mean you shouldn’t do it? No, it just means you need to calculate. You’re cashing in one advantage, a dominant minor piece, for another advantage, a direct attack on the king. This could be a great way to end the game on the spot, but before giving up a huge static advantage, you should be sure what you’re getting in return is actually worth it.
The key line is 1. f6 Bxf6 2. Nxf6 gxf6 3. Qh4 Re6 4. Rh3 when White crashes through on h7.
If you’ve seen this line, you can play f6 confidently. But if you didn’t see this line, and especially if you didn’t calculate anything, you were throwing away a winning position on a gamble. You might argue that White’s attack is so obviously crushing that you don’t need to calculate. Okay, but if this is a tournament game and you have some time on the clock, there’s no reason not to make sure. If you’re right, the game is basically over. If it turns out you’re wrong and you spot a defense for Black, it’s not too late to do something else and win in a different way.
This time, let’s admire a picture of another world champion, Tigran Petrosian, while you think.
Interestingly, the suggestion of several players, including at least one master, was 1... Rxf3+ 2. Kxf3 Qxd4 3. cxd4 Nxd4+ 4. Kg2 Nxc2. I have to admit I was also attracted to this line initially. But while this sequence looks like a puzzle solution, the more you think about it, the less chess sense it makes. It’s basically a very fancy way to trade a rook for a knight and two pawns. (And yes, I did try to trick you by using the picture of a world champion known for exchange sacrifices.) The final position of that line is actually quite good for Black, but something about it still strikes me as fishy. Putting your queen en prise with a not-that-forcing move should set your spidey sense tingling. And indeed White can blow up the whole operation with 3. Qxg6+ to get an extra knight out of the deal and emerge up a whole rook.
So it turns out this fancy combination would be an excellent way to lose the game on the spot. Going back to the initial position, what’s wrong with 1… Raf8? This simple move brings the last piece into the game, creates a rook battery, and threatens the knight on f3. Given that f2 is also weak, it’s hard to imagine how White will be able to deal with this. White can try a desperate counter-sacrifice with 2. Nxg5 hxg5 3. Rxg5 but this can be refuted easily, for example 3… Rf3+ 4. Kg2 Nxe5 (4… Rxf2+ is also winning). So it seems the most obvious move, the one you might play in a bullet game, is winning easily. But knowing it’s a puzzle, you might be tempted to look for something that looks more like “a tactic.”
Wrapping up, there are two main things you can do to make puzzles more applicable for real games:
Start by orienting yourself to the position. What are the major positional factors? Who’s winning? What are you trying to accomplish? A rook-up position is very different from a rook-down position. Likewise, a risky sacrifice might not make sense if you have a big positional advantage, but if your back is against the wall it could be your best option.
Seriously consider the possibility of playing a “normal” move. Depending on where you get your puzzles from, it could be the case that the answer will always be a tactical shot, but even if that’s the case it’s better for your actual chess game to consider quiet moves as possibilities. Not even Mikhail Tal played a tactical haymaker on every move!
thanks for the article...this adds something to what I have been trying with my puzzle work. Where do find the best game like puzzles are?
thanks