Interesting endgame

Marc Benford View Profile

Only Chess

08 Jul '13 08:20

Marc Benford
Joined
06 May '13
Moves
0
08 Jul '13 08:20
Hello.

I just played a blitz game with someone that had roughly the same rating as me, and although it was kind of poorly played the endgame interested me.
I really wonder if the first (after black’s 22th move) and the second (after black’s 43th move) phases of the endgame were a win or a draw with best play, and what I should have done better.
If you could give a look to the game, and perhaps give me some advices, I’ll appreciate. Thanks in advance.
Marc Benford
Joined
06 May '13
Moves
0
08 Jul '13 08:21
Here’s the game (I was black) :

EVENT Rated game, 18m+8s
DATE 2013.07.08
WHITE Caradesmen
BLACK Me
RESULT 1/2-1/2
WHITEELO 1691
BLACKELO 1688
ECO B50
Invert Board PGN Headers Comments

[Event "Rated game, 18m+8s"]
[Date "2013.07.08"]
[White "Caradesmen"]
[Black "Me"]
[Result "1/2-1/2"]
[ECO "B50"]
[WhiteElo "1691"]
[BlackElo "1688"]
[PlyCount "134"]

1. e4 c5 2. Nf3 d6 3. Nc3 Nc6 4. Bc4 Nf6 5. d3 Bg4 {I always love to pin this knight to the queen in the sicilian, but I don't know if it's really effective.} 6. O-O Nd4 {Not falling for 6. Ne5?? 7. Nxe5 8. Bxd1 Bxf7#.} 7. Bb5+ Nc6 {A second after I played this move I realized that I could have just blocked with the other knight...} 8. Bg5 a6 9. Bxc6+ bxc6 10. h3 Bh5 11. g4 Bg6 12. e5 Nd5 13. Nxd5 cxd5 14. c4 d4 {I thought it would have been wise to completely lock the center because I wanted to attack his king with h5 and leave my king in the center, but I completely missed that he had the possibility of trading queens with Qa4+.} 15. exd6 Qxd6 16. Qa4+ Qd7 17. Qxd7+ Kxd7 18. Ne5+ Kc7 19. b3 {He wants to put all his pawns on light squares to have a good bishop. But unfortunately, this move is a mistake that gives me a pawn.} f6 20. Nxg6 hxg6 21. Bf4+ e5 22. Bg3 Rxh3 23. Kg2 Rh8 {I have a pawn advantage, but my bishop is terrible while his bishop is very good, and also my pawn structure is not so good.} 24. Rae1 Bd6 25. f3 a5 {My plan was to exchange this isolated pawn and then to attack the remaining backwarded b3 pawn with my rooks. Houdini points out that I could have first played the strong ...g5 to stop f4.} 26. Re2 a4 27. Rb2 axb3 28. axb3 Ra3 {I didn't want to play ...Rhb8 because I didn't want him to take the h file, and I didn't want to play ...Rab8 because I didn't want him to take the a file. So I chose ...Ra3.} 29. Re1 Rb8 30. Reb1 Kb6 {I thought I could infiltrate his position with my king...} 31. Rh1 Ka5 32. Rhb1 Kb6 {But then I realized that he has Be1, so I turned back.} 33. Be1 Kc6 34. Bf2 Bc7 35. Bg3 Ba5 36. f4 exf4 37. Bxf4 Rb7 38. Bg3 Kd7 39. Bf4 Bc3 40. Re2 Raxb3 41. Rxb3 Rxb3 42. Ra2 Rb2+ {Apparently Houdini says that I could have played ...Bb4 which is way stronger, but I was low on time and was just happy to completely simplify the position.} 43. Rxb2 Bxb2 44. Kf3 Ke6 45. Bb8 f5 {I think that this could have been a mistake, even though Houdini says it's a good move. Maybe I should have tried first ...g5, then ...g6, and only then ...f5. Then if my opponent captures I'm good since I have two connected passed pawns, and if he doesn't capture I could perhaps try ...f4.} 46. Bf4 Ba3 {I wanted to get my bishop on d6 (so that he could protect my c5 pawn while at the same time being active), but I found it to be very difficult, and my bishop was a big dummy trapped on the corner of the board for most of the rest of the game.} 47. Bb8 Bb4 48. Ba7 g5 {Houdini says it's a blunder, and evaluates the resulting position as -0.31. Yes, but what else could have I tried? Houdini says that I have only three good moves : ...Kd7, ...Ke7, and ...Ba3 (all are evaluated as -1.64). Every other move are draws. And seriously, these three moves don't seem to do much.} 49. gxf5+ Kxf5 50. Bb8 g4+ 51. Kf2 g6 52. Bd6 Ke6 53. Bf4 Ba5 54. Bb8 Bd8 55. Kg3 {Houdini says ...Kd7 apparently was winning, but I didn't have enough time to see that.} Kf5 56. Bd6 Bb6 57. Be7 Bc7+ 58. Kf2 {And now the winning move was ...Kf4. I considered it for a millisecond, but was too afraid that his bishop would eat all my pawns. However, I would have undoubtedly found that move if I had two or three more minutes on my clock…} g3+ 59. Kf3 Bb6 60. Kxg3 Ke6 61. Bg5 Bc7+ 62. Kg4 Bd6 63. Bd2 Kf6 64. Bh6 Be7 65. Bd2 Bd6 66. Bg5+ Ke6 67. Bd2 Kf6 {Draw by threefold repetition.} 1/2-1/2
greenpawn34
e4
Joined
06 May '08
Moves
42492
08 Jul '13 13:111 edit
Hi Marc

If you have gone over the game with Houdini I doubt if there is much
we can add, don't these things have tablebases that kick in when there
are so few bits left?

Threads entitled "......Endgame" are always worth a look as I'm of the
opinion that at our level 90% of the games should never have been an ending.

What usualy happens is a player enters a lost ending and asks for help in saving it.
Out come 110 suggestions on what endgame books/Vids/Sites are best
whilst the real problem was in the middle game.

There was an interesting moment here when the game could taken a
more dramtic ccourse.

FEN r3kb1r/4pppp/p2q2b1/2p3B1/2Pp2P1/3P1N1P/PP3P2/R2Q1RK1 w kq - 0 1
Invert Board PGN Headers Comments

[FEN "r3kb1r/4pppp/p2q2b1/2p3B1/2Pp2P1/3P1N1P/PP3P2/R2Q1RK1 w kq - 0 1"]

1. Re1 {Much better than tamely taking the Queens off with Qa4+ (what a miserable move). Black is seriously lagging in the big D and is in danger of getting scudded.} 1... h5 {As Hinted at in the notes. (now we know what is coming f6 is possibly best.)} 2. Bf4 {If Black tries to stop the coming Qa4+ with 2...Qd7 or 2...Qc6 then 3.Ne5 does him.} 2... Qxf4 3. Qa4+ Kd8 4. Ne5 {An attempt to give the bit back with 4...Bf5 meets 5.Re4! Qd2 (else the Black Queen is lost.) 6.Qc6 hitting the a8 Rook and threatening mate on f7.} 4... Ra7 5. Nc6+ Kc7 6. Nxa7 Qd6 {To stop Qc6+} 7. b4 Qb6 {7....cxb4 8.c5.} 8. b5 Qxa7 9. Rab1 {Black is still in trouble, this won't make an ending and the moves I gave looks plausible (quite a few wee snares for Black to avoid) and defending in blitz is very hard.}
SwissGambit
Caninus Interruptus
2014.05.01
Joined
11 Apr '07
Moves
92274
08 Jul '13 18:411 edit
Originally posted by Marc Benford
Here’s the game (I was black) :

[pgn]
[Event "Rated game, 18m+8s"]
[Date "2013.07.08"]
[White "Caradesmen"]
[Black "Me"]
[Result "1/2-1/2"]
[ECO "B50"]
[WhiteElo "1691"]
[BlackElo "1688"]
[PlyCount "134"]

1. e4 c5 2. Nf3 d6 3. Nc3 Nc6 4. Bc4 Nf6 5. d3 Bg4 {I always love to pin this knight to the queen in the sicilian, but I don't know if 5+ Ke6 67. Bd2 Kf6 {Draw by threefold repetition.} 1/2-1/2
[/pgn]
I don't think you can win after the last rook trade. If he just holds his K on f3 and P on g4, where does your K break in? Not a4, because he has Ke2, assuming he can't just take some pawns with the Bishop.
DeepThought
Losing the Thread
Quarantined World
Joined
27 Oct '04
Moves
87415
09 Jul '13 00:113 edits
Originally posted by greenpawn34
Hi Marc

If you have gone over the game with Houdini I doubt if there is much
we can add, don't these things have tablebases that kick in when there
are so few bits left?

Threads entitled "......Endgame" are always worth a look as I'm of the
opinion that at our level 90% of the games should never have been an ending.

What usualy happens is a (quite a few wee snares for Black to avoid) and defending in blitz is very hard.} [/pgn]
Hi greenpawn, yes endgame table bases (EGTB) can help in post game analysis, but not there. Five piece table bases (this includes the kings) use up 4Gb of hard drive. By the time you're up to 6 pieces that hits the terabyte mark. There is no way on earth that an (affordable) engine equipped with EGTB is going to cope with that position, they'll just moronically test the table-base at the end of their 20 move line and find most positions score the same. To the extent that some engines will come out with better moves if EGTB access is turned off...
greenpawn34
e4
Joined
06 May '08
Moves
42492
09 Jul '13 01:56
Cheers Deep Thought.

I know these things are out there and they were limited to a certain
number of pieces. Not sure how many though, last I heard it was 7.
So it' seven counting ALL the men I thought it was 7 a side.

So eventually they are going to solve Chess backwards.
Give me a shout when they have a 32 piece TB. Be interested what the
first White move will be.....be funny if they discover no matter what White
plays he's losing...White's in Zugzwang!

I read an artilce in NIC years ago when these things were making an
appearance that a 'clever' computer in a position say KRPP v KR would
just toss away it's extra pawn so it could then access it's TB and play
the ending 100% accurately with zero time used.

Just finished a nice unbeaten blitz session at Gameknot.
Played approx 20 - won the lot.
Tonight I'm hot and lucky - other nights I'm cold and Mr Blunderman.

Of course it helps you see the Mouse Slip to end all Mouse Slips.

EVENT GameKnot Blitz
SITE http://gameknot.com/
DATE 2013.07.09
ROUND -
WHITE hugh-jwang
BLACK kidzebra
RESULT 0-1
Invert Board PGN Headers Comments

[Event "GameKnot Blitz"]
[Site "http://gameknot.com/"]
[Date "2013.07.09"]
[Round "-"]
[White "hugh-jwang"]
[Black "kidzebra"]
[Result "0-1"]

1. e4 e5 2. Nf3 f5 {Guess now you know what colour I was.}3. Nxe5 Qf6 4. Nf3 fxe4 5. Ke2 {Obviously the lad intended 5.Qe2}5....exf3+ {White eventually unplugged...pity I had intended} 6.gxf3 Qe5+ 7.Kd3 d5 {When the coming Bf5+ mates him. Even giving up the Queen does not appear to stop it. } 8.Qe2 Bf5+ 9.Qe4 Bxe4+ 10. fxe4 Qxe4+ 11. Kc3 Qb4+ 12. Kd3 Qc4+ 13.Ke3 Qe4
ptobler
Patzer
Canberra
Joined
16 Oct '06
Moves
12006
09 Jul '13 05:35
38...Bc3 springs to mind,... looks strong
ptobler
Patzer
Canberra
Joined
16 Oct '06
Moves
12006
09 Jul '13 05:411 edit
In fact 38...g5 is even stronger - winning, according to Komodo (-1.61) - whereas 38...Bc3 only gives you an advantage of -1.05 or so (after a minute or so of running it)
ptobler
Patzer
Canberra
Joined
16 Oct '06
Moves
12006
09 Jul '13 05:48
-1.67

38...g5 39 Kf3 Re7 40 Bf2 Re8 41 Rh1 Bc3 42 Rc2 Rxb3 43 Ra2 Re7 44 Ra6+ Kb7 45 Ra2 Bb4 4 Rd1 Kc6
greenpawn34
e4
Joined
06 May '08
Moves
42492
09 Jul '13 14:19
Hi Ptobler

Honestly mate without a diagram or PGN moving thing you effort is wasted.
The lads have to go back to original game, copy it, paste it into a PGN
mover, then flick back to RHP and enter your line move by move to see
what you are saying.

That is too much to ask from the lads on here.
Chess players are famously lazy. That is why many Chess players wanting
to start a family marry pregnant women because they cannot be bothered
making love and would rather have someone else do it for them.

I'll do the cutting and pasting this time but it would have been easier if you
had done it because you had line already to go up.

Perhaps all you work and time will get a response.
(if the line is sound then I want ½ the credit.) 🙂

Personally I think the lad found his own answer when he said here:

45...g5 was better than 45....f5 the plan with g6 and then f5 creating two
connected passed pawns looks very good.

Your line:
(or your computers....'Komodo' what is that? another chess program or your pet name for Fritz?)
Starts here.

Black to play.
(I would not class this an ending whilst both sides have two Rooks.)

FEN 8/1r4p1/2k2pp1/b1p5/2Pp2P1/rP1P2B1/1R4K1/1R6 b - - 0 38
Invert Board PGN Headers Comments

[FEN "8/1r4p1/2k2pp1/b1p5/2Pp2P1/rP1P2B1/1R4K1/1R6 b - - 0 38"]

38... g5 39. Kf3 Re7 40. Bf2 Re8 41. Rh1 Bc3 42. Rc2 Rxb3 43. Ra2 Re7 44. Ra6+ Kb7 45. Ra2 Bb4 46. Rd1 Kc6 {End of variation.}

EVENT	Rated game, 18m+8s
DATE	2013.07.08
WHITE	Caradesmen
BLACK	Me
RESULT	1/2-1/2
WHITEELO	1691
BLACKELO	1688
ECO	B50

EVENT	GameKnot Blitz
SITE	http://gameknot.com/
DATE	2013.07.09
ROUND	-
WHITE	hugh-jwang
BLACK	kidzebra
RESULT	0-1

Only Chess Forum

Interesting endgame