When to push pawns queenside?

So I'm often faced with this situation. I'm white, I've castled kingside. So has my opponent. All our pieces are out. Center is occupied.

At this point I'm always uncertain about pushing my queenside pawns. I don't know when I should, when I shouldn't.

Same type of thing with castling queenside and pushing the kingside pawns. Is it just generally a good thing to push these pawns as long as they aren't left hanging? I don't know how to decide.

I don't think there's any rule of thumb related to queenside pawn pushing in positions where both players have castled kingside.

The general flow of a game goes something like this (sometimes you can skip numbers, this is just a rough sketch):

1) Increase piece mobility (move off the back rank and centralize / on open lines)
2) Play a pawn break / exchange pawns to open lines (lines are rank, file, diagonals)
3) Infiltrate onto the opponent's side of the board
4) Come into contact with weak pawns / squares near the opponent's king

---

Ok, with that in mind, we can see a game of chess is really all about piece activity (pieces are non-pawns). You take your pieces through increasing levels of activity until you win material or checkmate your opponent.

First let's look at when to choose the queenside.

-

-

In the above diagram white's main pawn chain "points" to the queenside. This gives white more space (and black less space) on that side, and so it's natural to seek play on the side of the board where your central pawn chain "points"

This is also the so called "orthodox" or "carlsbad" structure. White typically pushes the queenside pawns forwards to exchange them off, which will leave black with an isolated or backward pawn that white's pieces can gang up on.

Notice how this fits in with the rough outline I gave at the beginning. White is exchanging pawns to open lines and eventually hopes to bring his pieces into contact with weak enemy pawns.

Here's a different kind of example.

-

-

This is a bit of an anti-example to reinforce the idea that pushing pawns should play the role of increasing piece activity. Black's move b5 wasn't bad, but black doesn't have many pieces on the queenside, so b5 was not played with the intention of the pawn break b4 soon after. With few pieces to make use of the open lines, a queenside pawn break doesn't serve much purpose at the moment. The black queen came to b6 to touch the weak e3 pawn, so that's nice. Meanwhile the b5 pawn keeps a white piece off of c4, so that's nice...  but mostly what black is dreaming of is the d5 pawn break. A lot of white pieces are pointed at the black kingside giving white pressure there, and the classic response to a flank attack is to open the center.

Thanks for the excellent reply llama47. Very helpful. That 2nd position is basically when I would be unsure as white. First idea for white that comes to mind to me is c4 with the hope of exchanging pawns and moving Nc4 or Rb2.

I have to post slowly, in pieces like that, otherwise chess.com eats diagrams. I think I'm close to done now, lol.

No problem. I actually meant c4, then Nc4 or Rb2. Accidentally wrote a4.

Yeah, after white plays c4 it would be nice if black captured and brought your knight to c4 with a gain of tempo right?

But why would black help white like that?

IMO one of the most useful test for any candidate move is asking "if my opponent ignores my move do I still like it?"

If black ignores c4, do I like my pawn on c4? What's it doing? Well it's blocking my bishop, that sucks. I guess I'll eventually have to play cxb to open my bishop back up, but opening a file on the queenside automatically lessens my pressure on the kingside (you can think of it like pressure in a balloon, if there's only one big hole, then all the air rushes through there. You want that "hole" to be your f file on the kingside against white's king. A hole on the other side will steal some air).

So yeah, generally you choose a side of the board to play on (kingside, center, or queenside) and then focus on that. Your opponent will usually choose a different area and that will be their counterplay.

You choose an area where you have more space (due to your advanced pawns) where you have more pieces, or ideally both space and piece superiority.

In the first diagram white typically chooses the queenside, while black tries to counter in the center or kingside.

In the 2nd diagram white will choose the kingside, and black will try to counter in the center (a queenside counter is not impossible, but with so few pieces there it doesn't make as much sense).

Now... sometimes counter intuitive things happen... sometimes both players build up on the same side, and both open liens, claiming they have an advantage, and that can be a big messy fight... but no matter what happens, the bedrock this all falls back on... the underlying logic... is that active pieces win chess games. If your pieces are touching weak pawns and/or squares around my king, and my pieces aren't doing the same, then you're winning. You can play the craziest moves you want as long as in the end you can satisfy that condition

Thanks for the tips llama47. Very helpful!

It's a really hard question, cause it almost can't be answered in a vacuum. Llama makes a few very good points, but I feel that it's not complete. For example, I really like ...b5 in the 2nd example.

I think there are mainly 3 different reasons to play a pawn move. 
1. To make your pieces better
2. To grab space
3. To cause weaknesses, remove your own weaknesses or to defend against weaknesses, including making the opponent's pieces worse. 

Examples of 1: you want to open up a file for your rook to improve them. Or you want to open up the position to get your bishops into active positions. These are your typical pawn breaks. Example of 2: the 2nd example of Llama's first post: black just grabs space with ...b5. Space can be a nice little advantage to have, as it's easier to play with your pieces on that side of the board. An example of the 3rd reason is for example the Carlsbad structure. White wants to smash his 2 pawns against the black pawn chain as trades will create weaknesses that he can later attack. Pawn breaks can also have this sort of effect.

So when you play a pawn move, it should have one of these reasons behind them. But the question of when to play them is still more difficult and depends a ton on the position. If you focus on improving your pieces (and making their pieces look bad) and on creating and attacking weaknesses (and removing your own weaknesses) you'll do pretty well in playing natural (pawn)moves. If you do find yourself in a position where it's not so clear what to do, usually the answer would be some sort of pawn move. Either just grabbing more space (you see this a lot in positions where one person has complete control over the centre and can't make their pieces even better) or advancing pawns in a search of creating weaknesses (rush of the a- or h-pawn is not uncommon). 

Push on the queenside if you exhausted all developmental moves.

Thanks. It's the grabbing space part I need to work on. 

     The first diagrammed position is ripe for the classic "minority attack". White pushes the two pawns--exact moves determined by what black does--in order to create a weakness in black's queenside pawn structure. Perhaps an isolated pawn, an undefendable hole that a white piece can occupy, or better yet a backward pawn on an open file. The idea is to tie black's pieces down to the defence of the weakness, limit black's space to create congestion in his ranks, and eventually win a pawn to set up a favorable ending. Or perhaps black will be tempted to open up his king's position in an effort to create counterplay on that side of the board.