How to see many moves in the future.

I have heard of chess players that can visualize a trap that will take place 7 moves down the line. Some people have told me that gm's can see even farther than that. I can see what will happen after about 3 moves then i begin to get confused. How can they visualize so deep into a game? What type of training does it take to be able to do this?

Anyone with any knowledge of such things will be greatly appreciated.

Three moves?  Wow!  I'd say you're doing pretty good.  I have trouble managing one.

First of all, it's a highly specific and trained skill (practice makes perfect).  But it certainly doesn't hurt that they have a veritable rolodex of position in their long term memory gained through years of study and play.  This includes both tactical patterns and strategic themes.  Because of this, when they calculate, they're exceptionally good ignoring poor moves.  How else could Kasparov win the first match vs Deep Blue when DB calculated literally one million positions per second?  It's because the computer was looking at hundreds of millions of extraneous positions that had nothing to do with what was going on...

And the same can be said for amateurs.  Amateurs spend a lot of time calculating and not a lot of time understanding.  GMs can certainly calculate deeply, but it's their accurate evaluations of moves that count.  Many times when we cannot, they can tell whether a move is good or bad just by looking at it (no calculation needed).

If you call the moves out, or it's a relatively simple tactics problem, I can see 10 moves (20 ply) ahead... but this means nothing during a game, as I'm usually calculating only 2-3 or so moves ahead and spending most of my time evaluating.

Another way of saying it, in an average position, to calculate all lines 3 moves ahead is something like a million positions... impossible for anyone (or anything) to brute force search a good game of chess (computers prune their move tree like crazy).  It's the understanding (especially for humans) that counts (and consequently makes deeper calculation possible).

The Way to calculate deeply is to understand simple positions, themes and ideas.

for example, The position below is white to move and win.

Now that you understand the first position, you should be able to calculate the next position down, which is 2 ply deeper than the first

...And in a similar fashion, if you understood the 2nd position then you should be able to understand the next position, which is a further 2 ply deep.

Blackadder, that's an excellent post.

Some very good posts here.

A couple things to add about visualization, if you are struggling to accurately "see" 3 ply  then take a step back and really work on basic stuff like what color each square is and be able to see all the possible squares that a Knight can attack etc. Being able to accurately visualize the lines of force that a piece controls can greatly increase your ability to "see" a position.

A lot of people don't really even see a position in their head but they hold the interaction of chunks of pieces.

I am stoked because when I really started getting into chess a few years back I remember chess columist commenting on one ability to gain a higher level of play. Most of the folks out there think chess genius is something you are born with.. I disagree you can go as far as you want at any level. The question really is how bad you want to inprove and is the desire a truth statement to the core of what you really want. Cause if you really do want to be the best you will find a way and then one day you too will believe me...

There are quite a few people who want to be the best, and who work their butts off in an attempt to be the best, yet only one of those will be the best.

A strong desire to improve, will to win, and willingness to put in hard work to be the best will go a long way, and without that, I don't think natural talent will get you anywhere close to the top.  But some people do have more natural ability than others, in chess just as in other pursuits.