Hi, thx MagicTrick 
For pointing out that the position is a theoretical draw.

The following quote from the first post in the thread, makes it clear that I was not the only one who failed to check the starting position:
It refers to analysis in the subscriber section:
This is the final position of an analysis line. IM Wallace comments:
"and White wil l emerge with an extra h-pawn in a queen endgame. although I am not enough of an expert to say with any certainty whether it is a win or a draw!"


On the reason to present a clear method for black to hold the draw,
instead of referring to the tablebases: consider the following quote
from earlier posts in this thread:

P.M. I've avoided to resort to checking the more difficult positions where black gives up the pawn,
to reach a theoretical drawn but practical extremely hard to defend position.
First of all, they are extremely hard to analyze, given all the possible winning attempts.
Secondly I think it's important to considers these endgames without resorting to databases or
tablebases if possible, as the results are often hard to apply to practical play.
Still if my memory is any good the h-pawn has a reasonable large drawing zone in the top left corner,
should disaster befall black (my source for this is John Nunn, Secrets of practical chess).
By the way, against a far advanced f-pawn Nunn states the drawing zone is indeed very small,
so there was every reason to look for a continuation in plan C that didnt end up in this kind of endgame.

On your comment:
"White can allow himself a move like Kf4"
The problem was not to show white could draw, but that black could draw.

Nice analysis indeed. Now it is more clear to me. Thank you both..

Thx,

I appreciate your point!
To summarize our last discussion from diagram IIa 

9.Qg7, Qf5

Reaching the following position:

diagram IIb
http://www.france-echecs.com/diagramme/imgboard.phpfen=7K/6Qp/2k4P/5q2/8/8/8/8w-...
7K/6Qp/2k4P/5q2/8/8/8/8 w - -


10.Kg8!              Might be considered a high class waiting move
                        10.Qxh7,Qf8+ 11.Qg8,Qxh6 is an immediate draw.

10........, Kd6      Black continues with his plan to bring his King near the white h-pawn
                          His ultimate goal is to get his king on e7 (if the white queen leaves the 7th rank)
                        Nunn calls this the proximity zone. One of the two zones in this type of
                        endgame where the defending king should go (the other is in the top left corner).
                        However black should be aware of the fact that this gives white more opportunities
                        for cross-checks

11.Kh8!?             11.Qxh7,Qe6+ 12.Kg7,Qd7+ 13.Kg6,Qg4+ and white's king cant escape the checks
                           But this is not the final word as under the circumstances 12.Kh8! is better. 
                                 (See the previous post by Dink Heckler)
                                 So black should probably play the more accurate
                                 11.Qxh7,Qc8+ !(iso 1...Qe6) 12.Kg7,Qc3+= or 12.Kf7,Qc4+=
                                 Keeping to the principle of checking allong diagonals and            
                                denying white any chance for a position where he can interpose his queen.
11......., Ke6      Or even 11.....Kc6 returning to the original position IIb.
                         But not mindless but plausible checking:
                         (remember that an ill considered check led to Aronians downfall in his Q+p endgame against Carlssen)
                        11....Qc8+? 12.Kxh7! gives white fair winning chances.


[edit: move numbers corrected (when posting 'from the top of my head, I shifted the movenumbers by 1]

From diagram IIa

9.Qg7,Qf5
Okay, but doenst 10.Kg8,Kd6 just postpone the inevitable:
11.Qxh7, Qe6+ 12.Kg7, Qd7+ 13.Kg6,Qg4+ keeps the king locked up.


[edit: move numbers corrected]

diagram II
http://www.france-echecs.com/diagramme/imgboard.phpfen=5Q2/7p/7P/1k2K3/8/8/8/2q5...

@Dink Heckler,

Thx for your contribution!
Yes, your suggestion 6.Kf6! is a much better move than the 
continuations I gave to illuminate the draw. It sets black a more difficult task because
(a) Now black cant draw by exchanging queens as he's one tempo short;
(b) Black cant bring his King nearer as he cant cross the c-file as yet (7....Kc6 drops the queen after 8.Qc8+).
However I still consider the position as drawn, as I dont think it's so simple for white to pick up the black h-pawn. 

White's best plan to win the h-pawn is to approach the pawn with his King. This assumes he will put his
king at either g7 or h8. So:

6.Kf6,Qc6+!
7.Kg7,           I dont think the alternatives very meaningfull
                    7.Kg5,Qg6+ or 7.Kf7,Qg6+ and white doenst make any progress.
7......, Qg6+
8.Kh8            Now white is just two steps away from winning the h-pawn
                    But I am afraid this is not enough
8......, Kc6!     Black has bound white's queen to the defence of h6
                    and plans to play Kd7-e7. Once blacks king reaches e7 the position is a dead
                    draw as the defence .....,Qf7 becomes feasible and winning the h-pawn isnt enough.

diagram IIa
http://www.france-echecs.com/diagramme/imgboard.phpfen=5Q1K/7p/2k3qP/8/8/8/8/8w-...
5Q1K/7p/2k3qP/8/8/8/8/8 w - -

One tiny detail:
From diagram IIa the attempt to drive black's queen off with 9.Qg7 fails to 9....Qf5:
9.Qg7,Qf5 10.Qxh7,Qf8+ 11.Qg8,Qxh6 picks up white's h-pawn.

P.M. I've avoided to resort to checking the more difficult positions where black gives up the pawn,
to reach a theoretical drawn but practical extremely hard to defend position.
First of all, they are extremely hard to analyze, given all the possible winning attempts.
Secondly I think it's important to considers these endgames without resorting to databases or
tablebases if possible, as the results are often hard to apply to practical play.
Still if my memory is any good the h-pawn has a reasonable large drawing zone in the top left corner,
should disaster befall black (my source for this is John Nunn, Secrets of practical chess).
By the way, against a far advanced f-pawn Nunn states the drawing zone is indeed very small,
so there was every reason to look for a continuation in plan C that didnt end up in this kind of endgame.

[edit: move numbers corrected]

@micawber: thanks for the analysis! I also got the impression that this ending was more in the sphere of pawn endings, than Queen endings.

@Dink Heckler: Before checking out micawber's subsequent analysis, I first thought that IM Wallace was right, but of course I have missed 6.Qb8+,Ka6! (under diagram II). After that, I cannot see a sensible way for White to take the h-pawn (see also SF's analysis above). In case this is all wrong, can you please post the continuation you were looking at?

Excellent analysis, micawber!

My first instinct was to agree with IM Wallace, but as soon as you showed that Black can trade queens I knew you were right.  The most obvious line, 


1.Qg7 [1.Qb8+ Ka6 2.Qd6+ Kb7 3.Qd5+ (3.Kf6) 3...Kb6 4.Qe6+] 1...Qe3+ 2.Kf6 Qf3+ 3.Ke7 Qe4+ 4.Kf8 Qa8+ 5.Kf7 Qd5+
only draws, and your other lines confirm that.   

Although you were right, I still think you were a bit harsh on IM Wallace.  Instead, give yourself credit for finding the resources to draw.  Playing good defense is harder than finding ways to win pawns.

Great job in pointing out how hard it is to win these things!