I think it would be a good idea to improve your visualisation skills, as this will reduce the amount of calculation errors you make. In order to find the right move, one first has to 'see' it. Trw's advice of always looking at all checks and captures will definitely increase your chances of seeing the right move in the position; at your level, you will need to see not just the very first move (this is most important, but you seem to be adept in this area), but the following moves in your calculations. 

And if you miss a strong move many moves into your calculations, don't beat up on yourself - even the world's top players make mistakes in their calculations!

The first thing you should do is improve your tactical vision further - there are plenty of books and CDs on the market that provide a bounty of tactical puzzles. 

Secondly, solve more complicated puzzles which require you to not just see the first move, but to evaluate it correctly and see the key winning or drawing move many moves into your calculations. 

If you train yourself to do this for puzzles, it will become much easier to calculate effectively in your own games.

Good luck!

I thought 1.Rf7 was winning too, so I ran the problem through Fritz which quickly found not only that 1.Rf7 was best, 2.Qh7 probably wins too!

Here's what I have:  




1.Rxf7 Kxf7 (1...Ne3+ 2.Rxe3 Qd5+ 3.Ref3) 2.Ne5+ (2.Qh7+ Ke8 3.Ne5 Nf4+ 4.gxf4 Qd5+ 5.Kh3 Kd8 6.Rg1 Rc8 7.Rg8+ Kc7 8.Qxe7+ Kb8 9.Rxc8+ Kxc8 10.Qe8+ Qd8 11.Qxe6+ Kc7 12.f5 Rxa4 13.Qf7+ Kc8 14.Nc4 Kb8 15.Nd6 Ra3+ 16.f3! Ra7 17.f6+-) 2...Kf8 3.Qh7 Ne3+ 4.Kg1 Qe8 5.Qh8#

Your solution works, just not nearly as well.


So I would say that what's wrong with you is not your tactical skill, but your confidence in yourself!  (btw, I have a similar rating and similar problems! )

Sounds like you are on the right track since you have recognized the problem, a good first step! I am only around 2150, but have a few comments nevertheless:

Gerbarts wrote on 06/03/08 at 18:22:24:

Ask yourself if you would have played 1.Rxf7 (obviously the winning move if it works) in just one minute in a real tournament game. I would have taken more care to check everything! Obviously you did not really use this puzzle to simulate a game situation. Since you say you are good on tactical motifs, maybe you should shift your training to using more time in complex, unclear positions, as you would in a real game. For example Dvoretsky's books (f.ex. Positional Play, Tactical Play, Attack and Defence) have a lot of game-like situations marked as questions and exercises, they could be particularly helpful becuase he always stresses prophylactic thinking (see below).

As an aside I once solved this position myself, and I was immediately drawn to the question "Is there any difference between 2.Qh7+ and 2.Ne5+?" Any difference would have to be different defensive tries available to black. This simple list of candidate moves was obviously a great aid to finding the very best move (Thank you Kotov!).

Gerbarts wrote on 06/03/08 at 18:22:24:

I disagree strongly with this statement, and even think this attitude may be a part of your problem. A big part of a GM's thinking process is prophylactic thinking; insterting self-instructions like "What is he planning", "What is his best answer to my intended move", "What could I have missed" and so on. Avni's book "The Grandmaster's Mind", where he interviewed the strongest players in Israel, really opened my eyes to this and is highly recommended. Also his earlier "Danger in Chess" is a good survey of the reasons we overlook resources for the opponent.

In fact, sometimes you see otherwise good players forget to search for all their opponents' ideas, and then all their experience and reputation can't prevent disaster. I'm thinking of games like the recent Ivanchuk-Bu (1-0) from Sofia and Carlsen-Topalov (1-0) from Linares. Even 2700 players have to find that "110%" motivation and put in hard work at the board to maintain their level, and so we should also strive for that lower down the food chain.