Pawn-up endgame, won or not?

I own a few endgame books and while they're full of manoeuvres, none of them state whether an endgame without pieces where one side is a pawn up is normally a win or a draw. It would be useful to know this to decide whether to trade down to such an endgame...

I think that's probably a draw, you could exchange off so you have one pawn and he has none, but I don't think you could get your king in front of the pawn, which is necessary for the win.

I kinda think it should be a win...the pawns can make a passer on their own without the king help, so perhaps white can use his king on the kingside (forcing black to do the same) and let the c-pawn march on its own. If black tries to stop it, then white chomps the remaining black pawns and does have his king in good position to force promotion. 

In my games, when I'm a pawn up, I assume I have a won game. If my opponent can hold, then either I was wrong, or played it wrong. At the least we both get more practice, and even if I was wrong, my opponent can still mess up as well. 

As I understand it, king+pawn endgames are the most decisive type of endgame, so I would think usually a pawn advantage should be good enough.

Yeah, I was hesitant to call a draw because I knew a pawn up should be winning, I just couldn't see a way to penetrate through.

Principle of two weaknesses, my friend. Gotta let the c-pawn be on his own as a threat and use your king elsewhere instead of trading down to one pawn when, you're right, it's hard to get your king in front.

Yeah, there are a few obvious methods of trying to win, but there should just be a theoretical verdict on this and I can't find it :P

I think this is difficult for White even with best play on both sides.  I read in an endgame book somewhere that 4 vs 3 pawns is tougher than 3 vs 2 which is again tougher than 2 vs 1 to evaluate/analyze clearly for a win/loss.

Exactly what Shivsky just said except the opposite :). It is very easy to win up one pawn with many pawns on the board because it gives you more opportunities to make passed pawns. This endgame is very easy to win, create a passed pawn and keep an active king. If someone would like to play through an example I could show you just PM me.

It all depends on how you play.

I must have heard it wrong. Would be glad to see how easy it can be from this position.  Though I still believe it is impossible to calculate anything with precision with these many pawns?

White should win fairly easily. Here is a sample sequence:

The reson why you will never find this type of position in an endgame manual is that is terribly impractical. I mean all the pieces and the wing pawns are exchanged but the central pawns still in their starting squares?! You are much more likely to find something similar with all the pawns on the kingside (and not on their starting post).

P.S. I would be absolutely stunned if this particular ending isn't easily winning for white.

Well in an OTB game of mine a while ago this position arose after move 27:

This is good general advice that can be used as a starting point for more concrete analysis. It also helps you to make a guess if you just cannot say whether a particular exchange is good or bad. However, each exchange must be evaluated independently.

As an example take an ending where both sides have a rook+bishop with an extra pawn for the attacker. If the bishops are of opposite colour then the attacker usually shouldn't exchange the rooks as the drawing tendencies of pure opposite coloured bishop endings are very high. If the same ending is with the bishops of same colour it's often a good idea for the attacker to exchange the rooks but exchanging the bishops has to be evaluated very carefully as pure rook endings also have very high drawing tendencies.

I'm 99.9% sure it's won (and quite easily as well). In general, a healthy extra pawn is sufficient in pure pawn endings most of the time. Problems mainly arise with very reduced material or if opponent has some compensation (in an ending the compensation obviously takes quite different form than earlier in the game).

The pure rook ending with 4 vs.3 on a wing is a theoretical draw in most cases altough defending it is far from simple and in practice the attacker has quite good chances.

I would imagine that the pure knight ending falls somewher between these two extremes. I imagine that the ending with rooks + knights is probably better for the attacker than the pure rook ending but perhaps worse than the pure knight ending (not sure though).

Post #14 should be a win for white. After the forced exchanges white plays e4 to restrict black's pawns, puts another pawn on f4. After that it's a matter of using the extra pawn as the key tempo that pushes the black king off the key squares. White will either (a) generate a passed pawn or (b) get behind the black pawns to pick them off. If black can get white to waste his tempo it's possible to draw, but I would not want to defend that position by any means.

A very interesting question. I'm not exactly sure myself, but logically thinking, White should be able to win any pawn endgame when he is a pawn ahead. An exchange of pawns until Black has no pawn left would always make sure a win for White.

However, this depends on which ranks the pawns are and how far they are already advanced.