The Truth About Doubled Pawns, Part 3

Last week we analyzed games where doubled pawns were a liability in the endgame and unless they were compensated by something else (like a pair of Bishops for example), the game was doomed.  The next basic endgame is another example that in many cases a pair of doubled pawns is almost the same as having just one pawn.

It is easy to demonstrate many examples where doubled pawns are just bad in the endgames, but is it possible to find a position where such pawns would be  (quoting Martha Stewart) "a good thing" ? Why, of course!

The endgame Rook + 4 pawns vs. Rook + 3 pawns is very common in chess.  Even though it is a theoretical draw, you have to be very accurate since sometimes it is not that easy to defend such an endgame.  The next game is clear proof:

Enter double pawns and you don't need to be Kasparov to make a draw since the strong side simply cannot create a passed pawn without trading practically all the pawns:

The finish of this game is both funny and instructive, therefore I'd like to offer you to solve this little puzzle:

Finally I want to show a very famous position where all of Black's pawns are isolated, and also he has two sets of doubled pawns!  But thanks to his isolated doubled pawns he won the game!