Why do high level chess players consistently make really strange looking moves?

Sort:
Deranged

One thing I've noticed that both novices and titled players have in common: they all make strange looking moves that break all principles!

Intermediate level players (rated above 1000 but below 2200) tend to play more normal, solid-looking moves, but people above 2200 rating are always putting their pieces on awkward squares. For example, I see high rated players do the following things a lot:

1) Putting their knights on the rim of the board

2) "Undeveloping" their pieces (eg. moving a piece back to its original position, instead of completing development).

3) Refusing to double up their rooks, instead keeping them separated across the board

4) Making strange pawn moves on the flank instead of developing their pieces

5) Playing weird openings like 1. b3, even against other grandmasters

Can someone please explain this to me? Why is it that when I watch a game between 2 titled players, their moves have more in common with 600 rated players than 1600 rated players?

wannabe2700

I doubt that very much. Look at this game https://www.youtube.com/watch?v=T1l-_dE_65U 

Do you think the lower rater player played more normal moves than Carlsen?

Rocky64

Both novices and high-level players break principles but for completely different reasons. Beginners do it out of ignorance. Strong players are aware of the principles and follow them most of the time, but also know when it's appropriate to break them. This is true in any field, not just chess. Intermediates follow the rules, but experts are the people who can tell you when a situation is an "exception to the rule".

Deranged
Kallatroh wrote:

I doubt that very much. Look at this game https://www.youtube.com/watch?v=T1l-_dE_65U 

Do you think the lower rater player played more normal moves than Carlsen?

That's one of the more normal games. But there are many other games where pieces just get shuffled around, even when development isn't yet finished. For example, one of the "lessons" on chess.com is trying to tell me that I should play like this:

 

Shadow_Dragon_86

A lot of the moves you have described as dubious are aiming to create weaknesses by sacrificing time.

Shadow_Dragon_86

Also Nh5 is a classic King's Indian idea. Correct me if I'm wrong (I play the King's Indian) but, 2... Nf4 can be played because it can create light squared weaknesses on the kingside which black can easily exploit if white chooses to kick the knight. It is primarily about the long term plan, not about the short-term positional advantages. And you say that Black's light-squared bishop is inactive, but look at the monster of a diagonal it controls that can be utilized later for an attack. Finally, black's play seems to be directed towards the King's Indian pawn structure, which is perfectly in line with principles.

wannabe2700

Closed positions are of course full of maneuvering. Less space, more weird moves. In the King's Indian defence white usually tries to break though in the queenside and black through the kingside. The pawns in the center point that way. 

1.b4 is just one possible line. It's preparing for c5, as it would put pressure on black's position and open lines for white.  White only has one piece left to be developed and it's not clear where the bishop should go. Be3 would be hit by a future f5, f4. 

Nh5 the rim isn't some illegal area. Usually it's just a pit stop. You can see that the knight is getting closer to the white king. The main idea though, is to play f5. Playing f5 is just a must. How else would black get any space whatsoever?

2.Re1 white just doesn't want to give up his bishop pair yet or make the life of the h5 knight easy.

Nf4 not the main move. f5 3.Ng5 Nf6 is the normal continuation. There's no easy way to push the knight away as g3 would create a weakness on white's camp. A problem of Nf4 might be that it's hard for black to continue his attack. Often black plays f5 and then f4.  Also notice how 2...a5 3.bxa5 Rxa5 4.Nd2 Nf4 5.Bf1 would lead to the same variation anyway. 

3.Bf1 saving the bishop pair

a5 What would you like to develop and where? There just aren't many squares available at all. Bd7 would be a very slow move. The idea of a5 is to hinder the plan of c5 breakthrough. 4.a3 would just lose a pawn.

4.bxa5 Rxa5

5.Nd2 again our idea is to be able to play c5. The knight is heading to b3.

Ra6 the rook would have to move anyway

6.a4 More space is better. It's logical for white to play in the queenside. Many players seem to put their rooks on a3. You can see the rook would be a lot more powerful on that square. 

 

None of these moves were forced. They were just one possible but logical continuation.

Shadow_Dragon_86
Kallatroh wrote:

Nf4 not the main move. f5 3.Ng5 Nf6 is the normal continuation. There's no easy way to push the knight away as g3 would create a weakness on white's camp. A problem of Nf4 might be that it's hard for black to continue his attack. Often black plays f5 and then f4.  Also notice how 2...a5 3.bxa5 Rxa5 4.Nd2 Nf4 5.Bf1 would lead to the same variation anyway. 

Precisely, and a confirmation from a better player like you is better, so thanks.

Squishey

It's because the really strong players know how to think concretely and calculate alot deeper and know where the pieces should go depending on the plan. This is often a result of prior preparation or deep position understanding. For example, they might undevelop a piece because they see the piece have no prospect in that particular position being on that "normal looking square", so they look for another plan with the piece and found a subtle one.

To put it simply, they just understand chess deeper, calculate deeper and see further and they have reasons justifying the principle breaking moves which is unique to each specific position.

Deranged

So 2 things I still don't get:

1) Why is it that they arrive at so many positions such that awkward moves are considered superior to principled moves? I thought that the reason a move was considered a principle move was because it was the correct move in >95% of cases, and that awkward moves were only correct in <5% of cases, hence why we're taught to follow principles. But it seems that they're making awkward moves more often than not. If awkward moves are correct so often, then why are we even taught principles in the first place?

2) The whole "move x accomplishes y" doesn't make sense to me. Literally every move in chess accomplishes something. Even if I play an awkward move like 1. f3, I could list a bunch of things that that move accomplishes. So when a titled player tries to annotate a game by saying "this move accomplishes these things...", it's like yeah, I get it that those things are accomplished, but what makes those things more important than other things? For example, why is it more important to take a tiny bit more space on the queenside than it is to get piece activity? Why is it more important to perform a 3 move knight maneuvre than it is to develop 3 pieces? What makes an awkward move superior to a non-awkward move, when both moves accomplish different things? No one ever tells me this. They just say "this move accomplishes x, y and z", but they never bother to explain why x, y and z is more important than a, b and c.

wannabe2700
Deranged wrote:

What makes an awkward move superior to a non-awkward move, when both moves accomplish different things?

Those awkward moves just aren't awkward to me in any way. I can't relate at all. Post some possible alternative moves for the game you posted. Words don't really make it clear.

blueemu
Deranged wrote:

What makes an awkward move superior to a non-awkward move, when both moves accomplish different things? No one ever tells me this. They just say "this move accomplishes x, y and z", but they never bother to explain why x, y and z is more important than a, b and c.

Usually the Pawn structure dictates which factors are important and which are peripheral.

Take the King's Indian example from your own post #4 above. The interlocked Pawns give the game a closed character, so the development of new pieces is not nearly as vital or as consequential as it would be in an open or semi-open position. Instead, Pawn breaks are dominant themes... and this leads directly to White's idea of b2-b4 (supporting c4-c5) and to Black's idea of moving the f6-Knight to allow f7-f5.

Deranged

Well here's how I'd have played it out:

 

blueemu

Why are you so anxious to trade your GOOD Bishop (on c1) for Black's BAD Bishop (on g7)? Are you expecting to launch a mating attack on the K-side... where Black has more space and more minor pieces already positioned?

I get the impression (perhaps mistakenly) that you just disregard the Pawn formation and try to impose the same general plan on every position...

wannabe2700

That was a completely weird variation. Where would you put your bishop after h6? Why would you play nh5 when you found it unnatural before? gxf4 wouldn't weaken black's king at all, only Nh5 would feel weak.  Why would you play Bh6 as it's not even developing another piece? You are contradicting your own logic.

drmrboss

1. b4!! ( there are many choice in KID Classical) but it is Leela's choice, so I like it! (why Leela chose b4! she knows in this particular position, b4 gave her the highest chance of winning, according to her 60 millions games in the training)

2. According to Leela's strategy, you dont need to exchange that bishop on g7 with your active DSB, so you wasted Bg5 and Qd2, two moves ! grin.png And if you count Bh6, then you waste 3 moves to exchange his bad bishop.

3. exf5 is horrible, you exchanged your valuable "e" pawn vs his "g" pawn. You might need to read basic theory in KID, white cant stop black's king side push. White have to win in queen side faster than black can attack your king.

After move 4, your position is horrible enough to lose or get draw if you are extremely lucky.

 

There are other posts of Leela choices if you are interested in checking.

https://www.chess.com/forum/view/chess-openings/my-idea-against-the-kings-indian-defense?page=2

 

 

drmrboss
 

 

drmrboss

I dont know how you plan strategy, this is my strategy. I consider majority of strategy based on pawn pattern.

In this diagram, direction of white pawns goes into queen side, direction of black pawns goes into king side. The main strategy would be going into "c5" as fast as possible for white and "f5" as fast as possible for black.

 

As a white player, if you are really scared about getting attack on your king side, dont play "d5". Instead, play dxe5 or keep the center tension.

Squishey
Deranged wrote:

So 2 things I still don't get:

1) Why is it that they arrive at so many positions such that awkward moves are considered superior to principled moves? I thought that the reason a move was considered a principle move was because it was the correct move in >95% of cases, and that awkward moves were only correct in <5% of cases, hence why we're taught to follow principles. But it seems that they're making awkward moves more often than not. If awkward moves are correct so often, then why are we even taught principles in the first place?

2) The whole "move x accomplishes y" doesn't make sense to me. Literally every move in chess accomplishes something. Even if I play an awkward move like 1. f3, I could list a bunch of things that that move accomplishes. So when a titled player tries to annotate a game by saying "this move accomplishes these things...", it's like yeah, I get it that those things are accomplished, but what makes those things more important than other things? For example, why is it more important to take a tiny bit more space on the queenside than it is to get piece activity? Why is it more important to perform a 3 move knight maneuvre than it is to develop 3 pieces? What makes an awkward move superior to a non-awkward move, when both moves accomplish different things? No one ever tells me this. They just say "this move accomplishes x, y and z", but they never bother to explain why x, y and z is more important than a, b and c.

 

1) It's a matter of style. Alot of strong players choose to play complicated and sharp positions which requires deep calculation and concrete variations rather than relying on general intuition. Alot of times these strong players have to be quite creative or do something counterintuitive in order to solve a problem. Weaker players lack that creativity or experience, so they need decent generalised principles to guide them for a while until they get strong enough to arrive at their own solutions. Think of principles as training wheels to rely on for a while.

2) It all comes down to concreteness again. Because mainly the strong players see further ahead, calculated more variation or studied the position in depth so he might know that some goals are more important than others. For example, there might be a position where having a strong central knight outpost might not lead to anything, so the strong players find another more awkward move which offers more chances, and he knows this based on experience or prior study of the line/position.

 

Once again, chess is very concrete, each situations should be treated uniquely. If you want to play more "classically principled" positions, try and study some of Capablanca's or Carlsen's games. They prefer more simple positons, and thus their chess is flowing and harmonious.

Deranged

I don't understand why we want to expand on the queenside. Why does it matter which way our pawns are "facing"? Why can't white attempt a pawn break with f4 and why can't black attempt a pawn break with c6?