Normal Topic Rechel-Hebden endgame (Read 5826 times)
micawber
God Member
*****
Offline


like many sneaks and skunks
in history he's a poet

Posts: 852
Location: Netherlands
Joined: 09/07/05
Gender: Male
Re: Rechel-Hebden endgame
Reply #8 - 08/13/08 at 17:21:30
Post Tools
Thx guys for the positive reactions.
I have yet to post the sequel on the knight endings. There has been a
delay because I had to put in rather a lot of overtime at work.

At MUbas:
Great that you recognized the quote! Indeed I love Chesterton
(favourite book: The man who was Thursday).
Here is another quote for you, that would raise some eyebrows 
if connected to chess:
"I am going to base the defence on the fact that the judge has a bald head"
  
Back to top
 
IP Logged
 
MUBAs Opponent
Junior Member
**
Offline


I Love ChessPublishing!

Posts: 70
Joined: 02/06/07
Re: Rechel-Hebden endgame
Reply #7 - 08/13/08 at 00:17:35
Post Tools
micawber wrote on 07/04/08 at 20:09:36:
like many sneaks and
skunks in history
he's a poet

Hey, great thread, great analysis -- and also a great quotation  Smiley  I love Chesterton, and I didn't realise there were that many people who still read him. Thank you for bringing a smile to my face when I saw this!
  
Back to top
 
IP Logged
 
Smyslov_Fan
God Member
Correspondence fan
*****
Offline


Progress depends on the
unreasonable man. ~GBS

Posts: 6902
Joined: 06/15/05
Re: Rechel-Hebden endgame
Reply #6 - 08/01/08 at 07:16:03
Post Tools
Hear, Hear, to all those who thought this was a fantastic revision!  I had to double-check a couple of lines, but everything is laid out clearly and I didn't see any mistakes at all!  

WOW!

Thanks!
  
Back to top
 
IP Logged
 
blueguitar322
Full Member
***
Offline


I Love ChessPublishing!

Posts: 138
Joined: 07/27/06
Re: Rechel-Hebden endgame
Reply #5 - 07/29/08 at 17:11:35
Post Tools
Outstanding analysis. Thumbs up.
  
Back to top
 
IP Logged
 
Zarvox
Junior Member
**
Offline


I love ChessPublishing.com!

Posts: 72
Location: California
Joined: 05/15/05
Re: Rechel-Hebden endgame
Reply #4 - 07/24/08 at 00:47:22
Post Tools
Thanks for posting this, with enough diagrams so it was easy to follow in my head! I look forward to the Knight ending.
  
Back to top
 
IP Logged
 
chk
God Member
*****
Offline


a pawn is a pawn

Posts: 1063
Location: Athens
Joined: 10/26/06
Gender: Male
Re: Rechel-Hebden endgame
Reply #3 - 07/17/08 at 09:25:29
Post Tools
I needed a good revision of this stuff, thanks!  Cool
  

"I play honestly and I play to win. If I lose, I take my medicine." - Bobby
Back to top
 
IP Logged
 
micawber
God Member
*****
Offline


like many sneaks and skunks
in history he's a poet

Posts: 852
Location: Netherlands
Joined: 09/07/05
Gender: Male
Re: Rechel-Hebden endgame
Reply #2 - 07/06/08 at 15:17:02
Post Tools
pawn endgame second plan: white king invasion

after 43....g5 

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

The invasion-plan is bound with some risk. White can no longer catch up with black kingside pawns in case of a pawn race.
As we see later, this is a good reason for keeping the knights on the board. 


44.Kb6, f5
45.Kxa6

(It is not to late to repent:  45.Kc5 still holds the draw)
45......., h5
Now black is ready for the pawn race. White has two choices:
variant 1: promote the b-pawn
variant 2: promote the a-pawn


variant 1 white promotes the a-pawn
46.b4,g4
47.fxg4!


47.b5 is no good: 47...,gxf3 48.K6,f2 49.b7,f1+ 
Again an example how important it is to head promotion with check in your calculations.
The text move asks black a not to simple question:
Which pawn are you going to promote?

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

The answer to the question above should be simple, if black asks himself another:
Which pawn can I promote with check?
The answer then becomes clear
47........f4!
Note that the natural reply 47....hxg4 doenst win
47....hxg4? 48.b5! (position before material: after 48.fxg4??,f4! white again is lost) 48...gxh3 49.b6,h2 50.b7,h1Q (no check!) 51.b8Q=


48.b5
48.gxh4,f3 doenst help: black's king is in the square of the h-pawn
48..........,  f3
49. b6     , f2
50. b7     , f1Q+ -+

Black has achieved his goal, he promotes first and with check.
The rest is technique: to bring first his queen and than his king as close to the b-pawn as possible:
51.Ka7, Qf2+ 52.Ka8,Qf3! (pinning the b-pawn and threatening Kd7-c7-Qxb7#) 53.Ka7,Qe3+ etc.


variation 2 white promotes the a-pawn
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*
46.a5,g4



Now we get to three subvariations:
variation 2.a 47.fxg4
variation 2.b 47.hxg4
variation 2.c 47. various king-moves.

variation 2.a. 46.a5,g4 47.fxg4

47.fxg4,
Black should again think about which pawn he wants to promote.
White's king is blocking his pawn and has to move to the b-file.
If he should play Kb7, the answer is obvious: the h-pawn promotes with check.
 
But what if white plays Kb6/b5?
The answer remains the same: the h-pawn. The reason is that if the h-pawn promotes first, the new queen will cover the promotion field of the a-pawn!. 
Note that this reason is unique for a pair of a- and h-pawns.
47.....,fxg
Preserving the h-pawn! 
Instead 47....hxg4 would throw away the win:
47...hxg4 48.fxg4,fxg4 49.Kb7,g3 50.a6,g2 51.a7,g1Q 52.a8Q=
48. hxg4
White has no choice: after 48.K?, gxh3 black has two h-pawns!
48......., h4!
Again 48...hxg4 would have thrown away the win.
49.Kb7, h3
50.a6,h2
51.a7,h1Q+ -+

And black wins by bringing his queen close to the a-pawn:
52.Kb8 (52.Kb6?,Qa8!), Qh2+ 53.Kb7,Qg2+ 54.Kb8,Qg3+ 55.Kb7,Qf3+ 56.Kb8,Qf4+ 57.kb7,Qb5+ 58.Kc7,Qa6 59.Kb8,Qb6+ 60.Ka8,Kd7 etc.


variation 2.b white promotes the a-pawn 46.a5,g4 57.hxg4

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*
Now black must take care, because one condition isnt fulfilled.
At the moment the f-pawn blocks the a8-h1 diagonal so the h-pawn doesnt promote with check, which makes a difference!

47.hxg4, fxg4!
After 47...h4? black doesnt win 
47...h4 48.Kb6!,h3 49.a6,h2 50.a7,h1Q (no check!) 51.a8Q=

48.Kb7
After all, the h-pawn isnt close enough, is it?
48.fxg4, (now the diagonal a8-h1 is cleared, so) 48...h4! 49.Kb7,h3 50.a6,h2 51.a7,h1Q+ and wins.
Note that if white had played 49.Kb6 (iso Kb7) white cant promote after 51...h1Q since 52.a8Q,Qxa8!
48.Kb6 (which pawn is black going to promote now?) 48...g3! (now this pawn promotes with check) 49.a6,g2 50.a7,g1Q+ and black wins.

48.......gxf3!
Black still clears the a8-h1 diagonal, with a specific purpose in mind.
49.a6,f2
50.a7,f1Q
51.a8Q

It seems that white has achieved his goal, both sides have promoted their pawns
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*
Unfortunately for white
51.....Qh1+ !
Still wins for black, after the queens come off, white loses the pawn endgame.

variation 2.c white king invasion 47.K?

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

If white clears the way for his a-pawn at once, the position of the king determines blacks reply. If you worked through the previous variations you should have no difficulty to understand the variations:

47.Kb6, g3! (the g-pawn promotes with check)
47.Kb7, gxf3! (black will exchange queens after both sides promote their pawn, leaving him with a won pawn endgame):
47.Kb7,gxf3 48.a6,f2 50.a7,f1Q 51.a8Q,Qh1+ 52.K?,Qxa8 53.Kxa8, Kd6! etc.

This completes my analysis of the pawn-endgame from Rechel-Hebden
I hope you agree with me that there was something to learn from it.


I intend to post  the analysis of the Knight endgames during the next week.
  
Back to top
 
IP Logged
 
micawber
God Member
*****
Offline


like many sneaks and skunks
in history he's a poet

Posts: 852
Location: Netherlands
Joined: 09/07/05
Gender: Male
Re: Rechel-Hebden endgame
Reply #1 - 07/05/08 at 08:54:33
Post Tools
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

Lets examine this position again:
[b]The task of both black and white is clear: create passed pawn
s
White has two ways of doing so:
I. To advance his b-pawn
II. To capture the black a-pawn with his king.
Plan I
44. b4, h5
45. b5, axb5
46. axb5

we will examine 46.a5 separately, but this draws as well
46.........., f5!
Blacks last preparations for a pawn break, with g4.

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

White has achieved his goal, he has a passed pawn. But there is a catch: 
It is an isolated passed pawn, and it cant be promoted without the help of the white king. By the square rule, blacks king is in time to prevent unaided promotion.
But what if white is going to use his king to aid the promotion with say Kc6. Then the white king, by the same square rule will be unable to prevent promotion of the black pawns.

This leaves white with three strategies to pursue:
1. Use his king to promote his pawn.
2. Advance his pawn without help.
3. Use his king to stop the advance of the black kingside pawns.


variation 1  White uses his king to promote the pawns
47.Kc6, g4
Is is not usefull for black to try to hinder the advance with his king by playing Kd8?
True, this will force white to obstruct his pawn initially by Kb7, but he will regain his tempo because he now promotes with check!
47....Kd8 48.Kb7, g4 49.fxg4,fxg4 50.hxg4,hxg4 51.Ka7=
Rule: in a pawnrace always calculate if one of the promotions occurs with check.

48.b6, g4!
49.fxg4,fxg4
50.hxg4,hxg4

Now black is unable to stop whites pawn while white is unable to stop blacks pawn.
The logical resut is a draw.



variation 2. White tries to advance his pawn unaided

47.b6,  Kd7!
Black keeps his king in the square of whites b-pawn!
48.b7,  Kc7
Now the b-pawn is doomed and white must hurry to neutralize black's kingside pawns.
49.Kd4,  h4
Black signals his intention to promote the h-pawn rather than the g-pawn. White's king is not yet inside the square of this pawn.
50.Ke5, g4!
51.fxg,

Forced, as 51.Kf4,g3 52.Ke3,f4+ is lost.
51......, fxg4

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

52.Kf4!!
The only move as 52.hxg4,h3 is lost
52......., g3
Black is no better off playing 52...gxh3, as white picks up the h-pawns after Kf3-f2-g1-h2
Isnt black winning here, with his protected passed pawn?
No he isnt, this is the fortress show in the first diagram in this thread.

variant 3 White attends at once to the defence of his kingside

47.Kd4, Kd6?!
Trying to win the b-pawn
Instead 47...g4! 48.fxg,fxg 49.hxg,hxg is a clean draw.
48.b6, h4
49.Ke3,Kc6
50.f4!!


* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

Black can still draw with 50.....gxf4! 51.Kxf4,Kxb6 52.Kxf5,Kc6 53.Kg4,Kd6 54.Kxh4,Ke7 and black reaches the critical square f8 in time.

More instructive is what happens instead if black persists in taking risks:
50.......g4? 51.hxg,fxg 52.f5!!
It turns out that white is now winning:
a. He easily holds up blacks g and h pawn with his king
b. After 52....Kxb6 53.f6 it turns out that black is to late to stop the promotion of the f-pawn. In fact one of whites remaining pawns will promote without his kings aid.


This principle is worth remembering:
Two disconnected passed pawns can be stronger than two connected passed pawns
This reason is that a king may well stop the advance of two connected passed pawns, but cant stop the advance of two disconnected passed pawns if after capturing one of them, he will not be able to catch the other (by the square rule).


The second plan will be illustrated in the next post
« Last Edit: 07/05/08 at 10:08:43 by micawber »  
Back to top
 
IP Logged
 
micawber
God Member
*****
Offline


like many sneaks and skunks
in history he's a poet

Posts: 852
Location: Netherlands
Joined: 09/07/05
Gender: Male
Rechel-Hebden endgame
07/04/08 at 20:09:36
Post Tools
I examined the main lines of this endgame in another thread:

http://www.chesspub.com/cgi-bin/yabb2/YaBB.pl?num=1210712688/0/#13

There I allready remarked that there was more to say about this endgame. First I will return to the pawn endgame that might have occured after:
38.Ke3,Nd5 39.Kd4,Nc7 40.Nb7,Ne6 41.Kd5,Ke7 42.Nc5 and now
42....Nxc5! (iso Nf4?) 43.Kxc5

But before I do so I want to introduce a basic endgame that is very practical and explains some of the variations presented later.

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*
This position is a fortress, black can make no progress.
1.Kg1,g2 2.Kh2,Kf2 stalemate.
1.Kg1,Ke3 2.Kh1,Kf2 stalemate.
With black to play it also draws:
1....g2+ 2.Kg1,Kg3 stalemate.

With this position in mind, you shoud quickly find the easiest method to 
draw in the next diagram. 
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*
This postion and its solutions is in the first (German) edition of Awerbach's monograph on pawn endings. And this is what he has to say about it:
"We encounter this type of endgame quite often in practice and therefore it deserves our attention" Wink

Now that we are warmed up lets look at the pawn endgame from Rechel-Hebden:

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*
It's black's move. And it should be obvious that he must find some active defence. If black remains passive white captures the a6-pawn and wins easily with his connected passed pawns. Thus he must distract white and create a passed pawn of his own. But it matters greatly how he sets his own pawn majority on the kingside in motion.
The 'golden' rule is to move the potential passed pawn first.
It is the pawn that is not opposed by an enemy pawn, in this case the g-pawn. According to this rule, 43....g5! is the right move.

To prove this I will first show that neither 43...f5 nor 43...h5 are insufficient to draw. 
variation 1a.
43.....,f5 
44.f4!,      

(white makes it hard for black to create a passed pawn)
Note that moves like 44.Kb6 or 44.b4 would throw away the win, as black can correct his error playing 44.....g5!
44.... , a5  
(black tries to returns the favour)
45.b4!,       
45....., axb4
46. Kxb4, Kd6

White immediately creates a passed pawn. Black can stop this pawn with his king, and hold off whites king from the critical b7 square. But now white's king goes pawnhunting on the other side of the board.

* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*
47. Kc4!
The white monarch starts his journey to the kingside (47.a5 wins as well).
47........, Kc6
Still blocking whites king  (47.....g6 48.a5! and black's king must make way as he has to stop the runaway a-pawn)
48. Kd4, Kc6
49. Ke5, g6
50. Kf6 , Kb5


* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

51. h4!
Note that white can still throw the win away with the greedy and impatient 51.Kg7?? when black draws with 51....g5! creating a passed pawn of his own.
After 51.h4 its all over:
51.h4,Kxa5 52.Kg7,Kb5 53.Kxh7 and the black king is too late to rescue the pawns he deserted.

I have explained this example at some length, so that you should have no difficulty to comprehend the other side variations:
From the starting postion
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
* * * * * * * *
*

variation 1b.
43...f5?  44.f4,h6 (black tries to activate his kingside first iso improving the position of his king with Kd6 as in the variations above) 45.h4! (a preemptive move to render g5 harmless), g5 (after any king move white plays h5!) 46.fxg5!,hxg5 47.hxg5, a5! 48.Kd5! (and white picks up the f-pawn; note that 48.b4 is a blunder: 48...f4!=)



variation 2.
43....h5 44.f4!,a5 45.b4,axb4 46.Kxb4, Kd6 47.Kc4 (is very similiar to variation 1a.
Just note that 47..., f5 48.h4! +- and 47....,f6 48.f5 +-

In the next post in this thread I will examine 43...g5 in detail.
The remaining two posts will be devoted to various knight endgame variations.





« Last Edit: 07/04/08 at 23:30:47 by micawber »  
Back to top
 
IP Logged
 
Bookmarks: del.icio.us Digg Facebook Google Google+ Linked in reddit StumbleUpon Twitter Yahoo