Edit: question has been resolved.  Mm40 proved its a draw with an endgame tablebase on top of my analysis, the book is wrong.  Camembert gave a bit of history on the gaffe, showing that its not capablancas error, but rather the error of typesetters/editors.  Along with Camembert, various others have backed up my claim that I didn't simply misread the book with their own copies of the book.

Heres the position.  Am I missing something here?  Capa says black should be able to draw if white makes the mistake of f5 .  In capas book it suggests working this out, but I can't figure it out.

Edit: Since many of you don't seem to know this, a king in front of a pawn on the sixth rank like this is always a win.  Just move the king to h7 or f7 and push the pawn.

maybe with  Ke7 you can play out a draw

According to the very reliable http://www.k4it.de/index.php?topic=egtb&lang=en, after 1...g6, white can force a checkmate in 21. You can check it out, but I think Capablaca may be wrong (gasp).

*mind explodes*  Anyways that answers my question, thanks.  Also thanks for the post Musiq, but I wasn't looking for alternatives to g6 for black because capablanca said to study AFTER g6.

Maybe there was an error in the book, which is pretty common in chess books I believe, since that same endgame tablebase at K4IT lists g5 as a draw (as opposed to f5).

I simply cannot believe Capa blew that pawn ending ... because pushing the other pawn is the mistake that lets Black draw -- answered by the g6 push you gave.  Do you have an algebraic "Fundamentals"? I'm guessing it's not algebraic and it says -- don't push pN5 and you goofed. Please double check.... please tell me you've made an error, or my mind explodes, too.

If you are working from an algebraic edition... maybe that's where the error crept in -- the lowly chess editor responsible for doing the conversion from descriptive to algebraic screwed up. Anyone have a descriptive notation Chess Fundamentals we can check in?

This is a gaffe with a complicated history. See number 4786 of Edward Winter's Chess Notes here (towards the bottom of the page) for the gory details.

I think capa talks about a unit that holds two, or something like that right?  How you should generally advance the unopposed pawn first.  If this is what he is talking about then it makes sense that the mistake would be advancing the g pawn first. 

Very interesting camembert, that should also clear up JG27s question =P.  Look at the link, it shows that capa did not err, but the editors/typesetters etc...

Indeed g5?? would be an incredibly bad move that loses, but let me clear up any questions with an excerpt from the book.

1. Ke4

[In this position White can't win by 1. f5 Black's best answer would be g6 draws.  The student should work this out ]

[He cannot win by g5 because g6 draws.  This, because of the principle of the "opposition" which governs this ending as well as all the pawn endings alreadys [sic] given, and which will be more fully explained later on.  White can win however, by playing 1. Ke4.]

I believe the final position in the diagram at post 1 IS a draw:

lol no 2. Kh2 and then push the pawn... hehe not a draw.

Thanks Camembert for finding that very informative chesshistory piece!

So, I wasn't too far off at all... as I suspected Capa got it right but the error crept in to a later edition ... wasn't even the change from Descriptive to Algebraic, just English to American editions.  Not surprising that Reinfeld et. al. published corrections of Capa's "boner" -- don't we all get a stupid little thrill out of seeing a GM blunder -- but for Capa to make an error in what for him would have been an elementary pawn position was too much to believe... Capa's accuracy in endings is the stuff of legend.

Yeah, thanks camembert. That was very interesting. Also a good catch by TrainingForNM. I have a copy that clearly says "White can't win...". Mine is US but I don't know the date. Possibly 12th printing, whenever that was.

where theres a capa theres a way!

Just brilliant!!!

You defeated Capablanca but I think Capa made a mistake as he didn't stress much on this move particularly but instead said that the student should work out this by himself.

Its amazing that even the masters of endgame make such mistakes.

2. Kh2 and then push the pawn the move you played is almost the worst move even losing the opossition.

It looks messy from historical article whether Capablanca made a mistake in this particular endgame. The endgame isn't so easy though, Capablanca takes quite a few moves showing Ke4 a win. He certainly made mistakes in other endgames in his books, for example a study in his "Last Lectures". Think it's very useful to try his exercises for the student though. Somebody has tried to post their solutions, will try to find the links.

http://www.k4it.de/index.php?topic=egtb&lang=en = copy of Shredder?

I think the position W--Kd4, p.f4,g4 vs.B--Kd6 ,p.g7 is a draw .

José Capablanca, Chess Fundamentals. (Everyman Chess). Example 8.

I quote Capa:

"In the above position white can win by 1.f5. Blacks best answer would be 1...g6. He can not win by 1.g5 because 1...g6 draws."

@ TrainingforNM: You seem to have read the example wrongly. Capa says its a WIN by 1.f5.