How many moves a GM can think?

Yeah, exactly. Some positions are way harder to calculate than others. If it's just a pawn endgame or something, yeah, you're just changing a few things in your head, it's not so bad. But in a sharp enough middlegame position, you can miss something really simple even on the second or third move. That's always humbling, haha. Oh, I found this mate in 9, I go here, he goes there, then I sac, then he does this, and then I realize that back on move 2, oh, he can just fork me with this move. 20 minutes wasted lol.

It's also surprisingly common that I just kind of forget what a piece controls. Like if in a position, a queen starts on d8, but as I calculate it ends up going to e8, I'll often put pieces on h5 like it's no big deal, because the current position has the queen on d8... it's like I blend the calculated position and the current position together in some weird irrational way.

Yeah, I'm sure that's common.

Another one I'll do is after a file is opened, and it's a main point of the line too, on one of the moves you pretend it isn't because like you said it's on the board.

---

But when I'm not rusty and I've slept well and all that, I'm actually fairly confident of my ability to visualize. My problem seems to be that I calculate too much in my games. I trying to find ways to help me cut things short.

One of my new questions I'm trying to work into my routine is "what am I choosing between?" That way I don't do something like spend 10 minutes on one move and I haven't even considered 1 other candidate yet... sometimes a candidate can be clearly superior but you have to give it brief look to realize.

Or sometimes there really is no other move so it's pointless to analyze.

But I'm pretty bad at just going with the flow so to speak. If something looks interesting, I can lose track of time and I just keep looking at it. A really bad habit!

Yeah, those kinds of "back to basics" questions are actually really good to do. I've only started doing them in the last year or two. I ask myself "is this the kind of position where finding the right move is necessary?" or "do you actually have to refute your opponent's counterplay or can you just stop it?" In fact regarding the latter, I blew a totally won endgame against an expert. I had some extra pawns, and was very close to queening, but my opponent had some counter threats. I was sure that my play "deserved to be" faster, but that meant calculating and we were in mutual time pressure. My play actually was a lot faster, but it was sharp enough that one or two bad moves can turn that around, and that's basically what happened.

Then after the game I realized I could have played a slow looking move that solidifies my position, and I can push the pawns at my own pace and win. Even with little time, it would be an easy win. And I could understand that easily. The problem was, even in the time pressure, I had to calm down and ask that common sense question, and I would have found the right strategy easily. Hard to do when you think you have found a forced (but tricky) win.

I really try to drill that stuff into my head. Keep asking myself questions. Am I getting too carried away with a tactical threat and don't actually have a plan? Does it really matter that I find the best move in this line? Am I missing a basic one-move tactic? It's surprisingly hard to do that because you're so engaged in refuting your opponent's ideas you just get distracted and stop playing wisely/practically. I think masters are just a lot more practical than me. They're more aware of their mind's faults, but that's why they steer the game into situations where it's harder for them to go wrong.

"If something looks interesting, I can lose track of time and I just keep looking at it. A really bad habit!"

Same! Like if I feel like a move shows a lack of understanding on my opponent's part, I feel like it just "deserves" to be refuted... even though it would often save way more time to just be practical. I just can't stand an unsolved puzzle. I mean, maybe take a minute or two to try to refute it, but don't pin your whole game on doing so, else you better find it or you will be down 20 minutes on the clock for nothing (and even if you find it you're still down the 20 minutes!). Especially unfortunate since you would have probably outplayed him later on anyway, without taking that think.

Bring practical. That's a good way to say it... I think the next breakthrough for me will be improving the practical side of how I play.

Yeah. It's funny. When 2200 players analyze, it doesn't always seem that different from me, they'll still occasionally forget things like, oh yeah, that's hanging, never mind. Yet you never seem to catch them like that when you play them. They seem to be a lot more mindful about creating a position in which you always have a solid go-to move that won't hang anything. Whereas someone like me might not care about that as much, and go into an unharmonious position, trusting that I'm "too good" to allow a tactical opportunity. But of course I often end up hanging stuff when I do that. The thing is, a master might also err in that type of position too, but he rarely gets himself into that kind of position.

Or if he does... he knows how to exchange the right pieces to quickly get out of it :)

It's hard to explain... I mean obviously masters are much better tacticians and stuff, too. But I think part of their ability to avoid mistakes is their ability to steer the game so that it's easy to play.

it depends on the positio

it depends on the position

I think a lotta people who have asked, are asking, and will still ask, this question (including me), really wanna know about just pure visualisation and depth of the lines a GM can 'see', or hold, in his head without losing the image. In other words, how many moves deep is a strong player CAPABLE of seeing? The question is really about visualising capability, not necessarily about deductive chess logic or candidate moves or whatever. Hence words like 'long' or 'deep' in such questions.

Unfortunately, many answers are usually in the form of "depth doesn't matter, width does" or "don't worry about depth, only computers think like that" or "just stick to short 2-4 moves lines and you're fine". Answers like these can really be misleading to tell ya the truth, especially as they are proffered by the very strong players themselves!

You wanna know the real answer to the question of depth? Nearly INFINITY, for GMs. You know why? Because they can play blindfold chess! What are you doing when you calculate in your games? Playing blindfold! And if you can hold an entire game of around 40 moves on average in your head while playing a blindfold game (mind you, you're calculating side variations within the 40 moves mainline of the game you're playing on the virtual chessboard in/of your brain), you therefore have the CAPABILITY to see 40 moves ahead, without getting foggy-headed! That's 80 ply in computer-speak.

Anand said in an interview that he could see "50 or 70 moves far, it doesn't really matter". Let that sink in, guys.

GMs don't necessarily stop calculating a deep variation when they need to because of any inability to hold the position in their heads, despite the typical advice you might get.

So, OP, there's your answer.

Hmmm... not true

I can play a decent game of blindfold chess - but I can't calculate anything close to "40 moves" or 80 ply in a chess game - and neither can a GM. 

In a blindfold game, you know that each of those 40 moves was actually made on the board. 

When calculating, there are 10^120 ways a game can go if you give it 40 moves... 

Small differences, you know :-)

My personal record is 26 ply--I'm quite proud of the resulting endgame win.  But when I show it to stronger players, rather than duplicating my 26-ply calculation, they look ahead about 6 ply and say "and then it's a winning pattern."  I didn't know the pattern and had to count the whole thing out.  So in some cases a GM may calculate *fewer* moves than an A player.

(I missed a branch, too.  Luckily it was also a win for me.)

Well I got to a king and pawn ending once everything was forced so I saw about 25 ahead.

My point, Solskytz, is that IF NEEDED, a GM can, and even will, calculate as far as he wants. Actually, as far as his visualisation 'super'powers enable him to. I'm not necessarily saying he will begin to consider all 10^120 move candidates in any given game, surely not. There's more to calculation than visualisation, I know that.

Here's Anand's take on it in this interview, just watch the first 40 seconds or so.

https://www.youtube.com/watch?v=EfX8US2iOVQ

If you're now speaking about visualizing any single line - then I'm with you. 

Calculating (or "thinking"), of course, is a very different story - of which visualization is one (very very important) element. 

I think about 3 moves ahead that's 1 for each side = 6 and about 3 different options *3 = 18 moves ahead. Call me a GM.

But the other opponent is also thinking 20 moves ahead, and since no one is perfect, they do not move 20 turns fast. Even if they could think 20 moves fast, they probably would still spend some time double-triple checking.