How Do You Practice Chess Visualization?

I use a 2.5" x 2.5" paper chess board that I cut out from a score sheet. I'll look at the paper and move an imaginary knight around the board, for example: find Nf5 [Ng1-Nf3-Nh4-Nf5], find Nf6 [Ng1-Nf3-Nh2-Ng4-Nf6] or practice openings, checkmates, tactics...everything and anything.

I find it much easier to calculate variations without pieces on the board. While playing internet chess, it's difficult for me to imagine my White dark square bishop capturing Black's knight on f6, look at the board, and NOT see that knight  anymore, because it is STILL there. Well, it's not completely that bad, but it gets much worse when there are multiple captures, and needing to see the resulting position after those multiple captures.

So, I entered "chess visualization skills" into google, and the search engine filled in the blanks - "how do you improve chess visualization skills". I then stumbled onto a site I toyed with a year or so ago, www.chess-strategies-tactics.com. I did the sample lesson, which was not easy for me, and it was only a level two exercise. Here it is.

You are told to visualize 1)...Qxd5,2)Rxg7 Rxg7 3)Bf6, and then told to visualize the winning move for Black.

I couldn't do it, even after staring at this  puzzle for 5 minutes. I just couldn't see the position clear enough that far ahead. Once I moved the pieces in Chess Base, I saw the winning move imeadiately, so it wasn't because I couldn't see the tactic.

How Do You Practice Chess Visualization? When did you develop decent visualization skills? After how many years? Was it something that just sort of happened as you grew up with the game over many years, starting as a child?

I started playing guitar at age 7, and I can visualize my fingers on the strings just fine. I can practice new scales, chords, anything - in my head - and then play it right the first time. I don't need to physically practice the guitar anymore to learn something new. I couldn't do this when I was 7, but at 53, and after 1000's of hours practicing the guitar physically, it's not too hard to see it in my head, and then play it. 

A very well thought out post! I hope I can provide a helpful response...

In my experience, it seems that the best way to improve visualization is to simply "do tactics". Of course in your example it is not too dificult to find ...Qd1+, however, the ability to fully see this position before it arises on the board is, of course, more difficult. I would simply recommend focusing the majority of your time, whatever percentage that may be, to doing as many tactic puzzles as you can on tactic finder. (Also, don't get caught up in the time or make the first move until you are sure you see the entire puzzle because this will defeat the purpose! :D)

Other than this, one visualization exercise I have come across is to, without any board place a knight on a random square and name another square. In your mind, try to find the shortest method to arrive at the second square. 

I'm not sure that was of any help... 

P.S. Focus on the forcing moves first!

When I first became interested in chess I was 18, so not a child.  I had trouble seeing 2 or 3 half moves ahead.  In the opening I'd often have to re-check tons of stuff... knight takes pawn, pawn takes knight and... crap where are those pieces again?

So after many years of just playing this got better slowly.

Then as zkman said I really started improving with tactics.  After about 6 years of playing I noticed a marked improvement when over the course of a month I did all the tests sections in "The Ultimate Chess Puzzle Book"  A test was 16 puzzles and you had 2 hours to complete it.  They started out easy and got harder at the end.  Some problems at the end I'd spend 30 minutes on (yeah, sometimes I'd go over the time limit ;)

As Zkman said don't worry about the time, and just work on it in your head until you think you have the solution all the way.  This book actually scored you for seeing certain variations, so after I thought I saw everything necessary to solve the puzzle and all relevant variations I'd stop, and write it all down (that way I coudln't cheat and think to myself, yeah I saw that, and give myself extra points ;)

I don't know how many hundreds of times the position would become "fuzzy" and I'd have to start over from move 1.  Eventually when I "moved" a piece to a new square, I'd pause a moment and "burn" it in place.  By the end of the month I could see very clearly many moves ahead, and very rarely had to re-check a previously calculated variation because I could see it so clearly I knew it was good...

I've since lost that efficiency due to not practicing/playing :p  I re-check and fumble around between variations again.  But the best way to get better at visualization is to practice visualization itself.  Tactics are one way I found enjoyable.  You could also try to follow analysis without a board, then set up the visualized position and check yourself.

I used to be really bad at visualization too, but it really does get a lot easier the more you practice. It used to seem crazy to me to try to find a good move in a position that isn't even in front of me (!), but now that I'm used to it it's not unnatural at all.

In fact, even now when I calculate I very often "reset" my calculation -- that is, I look 3 moves ahead, see a good possible resource for one side -- and I'll often start back from the original position and re-calculate those 3 moves again (plus the new half-move), just to make sure I'm taking into account every change in the position. In fact I do this quite frequently, but surprisingly it's not that big of a problem, as long as I can "re-calculate" very quickly -- the important thing is if I "lose my place" in the visualization I can quickly recover. Moreover in this way I am able to constantly remind myself of how the position is changing until I think I've got it.

But be very wary of that -- I lose my place a lot but because I know I lose my place a lot, I am able to make sure I check my calculations sufficiently before playing my move. Accuracy is so much more important than speed here -- one feature of the position you visualize incorrectly can totally change the evaluation of it.

As far as the method of practice -- it just happens naturally when I play OTB. You have to calculate lines, and see what the position looks like a few moves later. You can practice that in tactics puzzles or just by playing -- no special method is necessary.

I work at a job where I can't just break out a board or use the computer.. so I have no choice but to do it in my head (badly). 

Maurice Ashley has a great tool for this called "seek and destroy" in which you set up the pawns to be taken in as many moves as there are pawns on the board using a specific piece (such as a Knight for example)

Basically you start with 5 pawns and a piece. The pawns will not move in the position and you MUST find the correct sequence for one move to one capture. Once you get to 8 pawns, go back to 5 using a different piece.

Just a thought.

Doing all the tests. That's good discipline.

I think you really need to scrutinize why you didn't "see" Qd1 -- was it because you did not even consider checks...

Outstanding post! It gives me a lot to think about. Did I check for checks? Yes. But for some reason, I didn't "see" the queen as being pinned to the king, making that an absolute pin, where it would be illegal to move the queen off of the g-file.

For some reason, even though I knew that White's bishop had moved to f6, I worked so hard on being able to "see" that Black's rook was pinned to it's king, that I had it stuck in my head that the rook could not move, so, the queen was not under attack.

Maybe it WAS a tactical weakness, in that my brain could not process TWO absolute pins happening simultaneously. It's not a puzzle/pattern that I am familiar with.

Again, thanks for the help.

It's an interesting test puzzle... I could easily "see" the position after the moves... or so I thought.  The first move I considered was Qd1+ but then I calculated QxQ and threw it out.  What's the solution?  I re-calculated, yep, I see the future position correctly... stared at it for a while until it finally hit me that the queen was pinned.

So I saw where all the pieces were, but not the implications I guess.  Like musikmole said it's not a common theme to me either, so that may be the case.  It's interesting though there must be a distinction between knowing where the pieces are and understanding their interaction.

This is probably what happens to new players when they try to blunder check.  They see the bishop which is attacking their undefended knight... but somehow that line of force extending from the bishop and hitting the knight is unnoticed.  I think many fundamental chess skills are unspoken actually, and players either pick them up naturally though practice or not.  This is one of them I believe, easily seeing the interactions of the pieces in front of you (and after some practice doing it in future imagined positions as well).

I've broken out my wooden chess board on my bed and played the first two games of "My 60 Memorable Games" on the board. 

Would anyone support the belief doing this with a physical board in front of you improves vision? 

I just call that evaluating the position in your head that occurs a few moves later. I think the goal is to be able to evaluate a position in the future the same way that you would evaluate a position right in front of you. So if you normally take notice of pins and square coverage and different possible plans when the position is in front of you, try to do those same things in the "ghost" position in your head as well.

Yeah, it's the same thing, but certain unspoken skills take place.  Not just in this example but in tons of other positions (and other skills besides chess).  I think this is one distinction between a good and a bad teacher. 

"Do these exercises and gain this skill, it worked for me so it will work for you" while not understanding the fundamental skills involved and the student's strengths and weaknesses.

I think so.  On the computer it's so easy to click through a dozen moves.  Two advantages of the board in front of you are 1). it takes longer, so your mind has time to absorb the new position bit by bit and 2). You're physically interacting with the board.  Adding a sense to an exercise helps the brain recall.

For more skilled players it probably makes no difference though.

some books you might find valuable (authors omitted simply becuase I cannot remember them all

Lev Alburt "chess tactics pocket trainer"

kasparov's chess challenge vols. 1-3

Can you become a positional genius

Chess by Lazlo Polgar

perhaps these will spur you on as well.

In addition to what people above have already suggested, I think that just playing slow games can really help your visualization. If you play slow games and spend your time wisely (using all your time, as Dan Heisman is always recommending), then you will certainly get lots of practice trying to visualize all the promising lines you have available, and the consequences of all your opponent's replies; and since you really, really care about finding the best moves and winning the game, it's very easy to give this sort of "exercise" 100% of your energy and focus during a game.

great insight sapientdust.

The best way to improve in visualising future positions is to practise,practise,practice. How ? Like building a house, you begin from the foundation and work your way upwards. In order for visualisation to be effective it has to be ACCURATE, otherwise you're wasting your time. Also you must know when to calculate. There are not many books available to teach this much sought after skill. The Kotov, Aagaard and Soltis books are well known. A newly published and highly recommended book that uses step by step methodology to PRACTISE and improve visualisation accuracy is " The grand art of Chess Calculation ", by Friend. Again there is no quick fix method to attain this skill. Practising visualisation is paramount in order to become a master thereof.

I'm not looking at the board, in that example right now, but I don't think that's correct sequence. I believe it, rather (after Bf6), is Qd1+ and wins the Queen? Rxg5+ is illegal, as the R is pinned to Black K.

Trying to remember the move order without looking back: 1....Qxd5 2. Rxg7 Rxg7 3. Bf6 (pinning the Black R) and now, ...Qd1+ winning White's Q, since Blacks R pins White's Q to K.

Now I will see if I knew what I was seeing

there's are some good apps called "Chess Visualization" - for what it's worth they really improved my board vision by leaps and bounds. These apps are really good, but you have to use them at least for 15 minutes everyday. it took me two months of training and suddenly one day, it was like my IQ level went up by 50 points.