Game 14: BigRezi-SmithyQ: Endgames are Harrrrrd

There are two main ways people fail to win won games.  The first is the simple blunder, where you overlook a tactic or leave a piece hanging.  Opps.  Hate when that happens.  The second is more insidious, though.  It’s where you are winning the entire game, and you know you are winning, and yet you can’t finish the endgame off.

Maybe you’re up a pawn, up an exchange, up a piece, but for whatever reason victory eludes you.  If you don’t know what you are doing, then endgames are extremely hard, and I freely admit that I don’t know what I’m doing half the time.  This game shows that perfectly: I reach a very good position, on the verge of winning, and yet the longer the endgame goes the less stable my advantage gets.

Let’s take a look at this beyond frustrating experience.  You’ll notice that my commentary is top heavy, because I can explain the early middlegame very well, but beyond that, my explanatory powers go down.

[Event “2- Chess is for fun – Round 1”]
[Site “Chess.com”]
[Date “2015.04.26”]
[Round “?”]
[White “BigRezi”]
[Black “SmithyQ”]
[Result “0-1”]
[ECO “B40”]
[WhiteElo “1511”]
[BlackElo “1812”]
[Annotator “Pettit”]
[PlyCount “88”]
[EventDate “2015.??.??”]
[TimeControl “1”]

{Though I get a great position in the early middlegame, little errors make it
less great, then merely good, and by the late endgame I’m barely better at all.
Proof positive that endgames are my weakest area.} 1. e4 c5 2. Bc4 {Serious
question, was there a YouTube video or a famous game recently where someone
used this move to win brilliantly against the Sicilian? I’ve played the
Sicilian only a handful of times since 2015, and I’ve faced this in three
games, a statistical anamoly. Let this also be exhibit A of why memorizing
openings is a waste at amateur level, as you won’t find this move in any
opening book.} ({Just for fun, we’re going to reach a similar position as the
game, but one where White doesn’t waste a move with Bc4.} 2. Nf3 Nc6 3. Bb5 e6
4. O-O d5 5. exd5 exd5 6. Re1+ Be7 {and in this position White has lots of
ideas. He can play d4, threatening dxc5 because the Be7 is pinned. He can try
Ne5, attacking c6 and maybe f7 is Black is careless. He can try Qe2 and make
it very difficult for Black to castle. White has a plethora of ideas, all of
them good, and Black will have to defend for quite a long time … all because
of one tempo. Compare that to the game.}) 2… e6 3. Nf3 d5 {As a general rule,
if Black can play d5 in the Sicilian with no major drawback, he gets a good if
not great game. Here it comes with tempo, hitting the Bishop, so it’s
obviously good.} 4. exd5 exd5 5. Qe2+ {A curious move, but not necessarily bad.
The e-file is open, and White should control it before Black, so there’s no
real danger for White’s Queen. It’s not necessary at this time, though.} (5.
Bb5+ Nc6 6. O-O Nf6 7. Re1+ Be7 {Would be similar to the game, but having the
Rook on the open file is better for White, who can now place the Queen
somewhere else if required, possibly with d4 coming.}) 5… Be7 6. Bb5+ Nc6 {
Stylistically, in such positions, I almost always play Nc6 over either Nd7 or
Bd7. The potential doubled pawns are not a real weakness, as the game will
show, and I see no reason to develop my Knight or Bishop to a passive square
when Nc6 is clearly the best spot for the Knight. That said, the Knight CAN be
a weakness, and the computer prefers Bd7 by a not inconsiderable margin.} 7.
O-O Nf6 8. d4 {It’s hard to call this natural move a mistake, as it challenges
the centre and gets ready to finish development, but White has an open e-file
and a far more thematic option.} (8. Re1 {This move takes control over the
e-file and asks Black how he is going to castle. d4 can then come next if so
desired. Play the most important moves first.} Be6 {This is the most natural
move, blocking the e-file, but White can be annoying.} (8… h6 {this is an
option, to stop the Ng5 jump and so safe-guard the soon-to-be Be6. It costs a
tempo, but because White lost a tempo with his Bc4-b5 in the opening, it
should be playable.} 9. d4 Be6 {and the position is similar to the game.}) (
8… O-O $4 {is of course a blunder.} 9. Bxc6 bxc6 10. Qxe7 $18) 9. Ng5 {
White now threatens the blocking Be6.} Bg4 {This then leads to a fairly forced
sequence.} 10. f3 Bd7 11. Bxc6 bxc6 12. d3 h6 13. Nh3 Bxh3 14. gxh3 Kf8 $15 {
White’s pawn structure is terrible and he doesn’t have much control of the
center, but Black’s King isn’t great, either, and maybe White can use the open
g-file. Black will need to waste at least two moves (Kg8-h7) to get his one
Rook into the game. The game is imbalanced, though Black may be slightly
better.}) 8… O-O 9. Be3 {On the one hand, this is White’s best move, as it
develops a piece while protecting the d-pawn and pressuing c5, but it also
shows the disconnect in White’s play. He plays Qe2 to control the e-file …
and now he places his Bishop in front, blocking the file. White is playing
individual moves without any underlying plan.} Re8 $1 {Putting the Rook on the
open file and indirectly defending c5.} 10. c4 {Again, not a bad move by
itself, but what is White’s plan? Does he really want to open up the centre
more when his Queen is in line with my Rook? After the pawns exchange, what is
the next step?} (10. dxc5 $6 Bxc5 {[%cal Ge8e2,Rd5d4,Rf6g4] and White can’t
take the Bishop because of the pin. Worse, Black is threatening d4 or Ng4, and
White is immediately on the defensive.}) 10… Bg4 {For my part, I strive to
keep the central tension and so develop my last piece. cxd4 is now a threat,
and White stops it by exchanging my Knight.} 11. Bxc6 ({A natural move runs
into the following} 11. Nc3 cxd4 12. Bxc6 {A necessary intermezzo.} (12. Bxd4
Nxd4 {[%cal Gd4e2,Gd4f3,Gd4b5,Rg4e2]}) 12… bxc6 13. Bxd4 Bd6 {[%cal Ge8e2]
Opening up an attack on the Queen, and White will either lose the c4-pawn or
have pawn structure ruined.} 14. Qd1 (14. Qd3 dxc4 15. Qxc4 Bxf3 16. gxf3 {
and White faces a tough defence.}) 14… dxc4) 11… bxc6 12. Nbd2 {All minor
pieces are developed, so the opening is basically over. Black has a small
advantage, since my pieces are slightly more active and my Rook has potential
tactics against his Queen on the e-file. As I have the advantage, I start
action in the centre.} cxd4 13. Bxd4 c5 {I’ve gotten rid of my doubled-pawns,
and I may be threatening to play d4 and get a protected passed pawn.} 14. Bxf6
Bxf6 15. Qd3 dxc4 {Up until now, I’ve played perfectly according to the
computer. This is my first slight slip, where I play only the second best move.
} (15… Bxb2 {Taking the pawn is obvious, but I couldn’t see what to do after
the simple Rb1. The computer points out the following line:} 16. Rab1 Bd4 {
Blocking the Queen’s attack on the d5-pawn.} 17. Nxd4 dxc4 $1 {Intermezzo
capture!} 18. Qxc4 Qxd4 19. Qxd4 cxd4 {Black is likely winning: a pawn up in
the endgame with a good Bishop against Knight match-up.}) 16. Qxc4 Bd4 {
Another inaccuracy, this one more severe. I was too focused on keeping my
c-pawn.} ({The computer recommends} 16… Be6 17. Qxc5 Qb6 18. Qxb6 axb6 {
Where the two Bishops on the open board give Black excellent winning chances,
and he’ll soon win back one of the Queenside pawns.}) 17. Rad1 Qa5 $6 {And
this move completely throws away my positional advantage. It’s fascinating how
I can play 14 perfect moves, and yet three small inaccuracies later it’s a
perfectly equal game. Chess can be cruel.} (17… Qd7 {is the computer’s
recommendation, which gives Black some pull. I never even considered this move,
but it makes sense in hindsight. It prepares to double on the d-file, it plays
in the center and it makes sure every Black piece is protected.}) 18. b4 $1 {
My opponent finds the best move, which exploits why Qa5 is bad.} Qxb4 19. Qxb4
cxb4 20. Rb1 $2 {This, though, is not the correct follow-up.} (20. Nxd4 Bxd1
21. Rxd1 $11 {Many beginners and even intermediate players fail to realize how
two minor pieces are better than a Rook. White prolly can’t win this position,
but it would hard to lose, as well.}) 20… Bc3 {As is, I now have two Bishops
on an open board, plus more active pieces and an extra pawn. This should be a
cakewalk … but I make it far harder than it needs to be.} 21. h3 Be6 22. Ne4
Bxa2 23. Nxc3 bxc3 {I’ve won a second pawn, which should win trivially. Should.
} 24. Rbc1 Rec8 25. Rfd1 Bb1 {I now chase the wrong plan. I thought I could
just play c2 and anchor my Bishop to b1, forcing a Rook to always defend it.
Wishful thinking on my part.} (25… a5 {I have two extra pawns, so why not
push them as far as I can? They both have Rook support, and White will
struggle to stop them.}) 26. Nd4 c2 $2 {Maybe I had chess blindness and just
didn’t notice the pawn was hanging after Black’s obvious move?} 27. Rd2 a5 28.
Nxc2 Rxc2 (28… Rab8 $1 {This is a trickier line that still gives Black pull,
as White’s forces are tied up defending each other.} 29. Kf1 a4 (29… Rb2 $2 {
This move looks great, but White has} 30. Ne1 $1 Rbb8 {Black must retreat and
repeat positions.} (30… Rxc1 $4 31. Rd8# {Opps.})) 30. Ke1 a3 31. Kd1 a2 32.
Na1 {And Black has a winning endgame as soon as he manages to exchange all
Rooks and/or advance his King.}) 29. Rdxc2 Bxc2 30. Rxc2 a4 {For all that, I’m
still up a pawn, and an outside passed pawn, but it’s not so easy. Rook
endgames are never easy.} 31. Ra2 Kf8 (31… a3 {Here’s Rook endgame tip
number One: always push pawns protected by Rooks as far as possible. This is 0.
5 better than my move.}) 32. Kf1 Ke7 33. Ke2 Kd6 34. Kd2 Kc5 35. Kc3 Kb5 {
The computer says this may not even be winning anymore. I don’t understand
endgames well enough to fully explain why, but below is 1) the computer’s
recommened move, and 2) my other candidate move (backed by computer analysis).}
(35… g6 {This is the computer analysis.} 36. Rc2 Rb8 37. Kd3+ Kd5 38. Rd2 Rc8
39. Ke3+ Ke5 40. Ra2 Rc4 {Make sense to you? Yeah, not to me, either. That’s
why computers aren’t great for learning endgames.}) (35… a3 {This was the
other move, following my tip above. It seems better.} 36. Kb3 Kb5 37. h4 (37.
Rxa3 $2 Rxa3+ 38. Kxa3 Kc4 {and the King will to the pawns before White can
protect them.} 39. Kb2 Kd3 40. Kc1 Ke2 $19) 37… Ra4 38. Kc3 f5 39. g3 g6 {
And now White must give ground with his King, as} 40. Kb3 Rb4+ 41. Kc2 (41.
Kxa3 Ra4+ 42. Kb3 Rxa2 43. Kxa2 Kc4 {Just like the other variation.}) 41… Ka4
42. Kc3 Rb3+ 43. Kc2 Rb7 44. Kc3 Rc7+ 45. Kd4 Kb3 {and the pawn soon Queens.
And no, I didn’t calculate or see or even have a glimmer of any of this.}) 36.
Kb2 $2 {White retreats for no reason.} (36. Rb2+ {repeats the position.} Kc5 (
36… Ka5 $2 {would be a huge mistake, as my King would be stuck on the a-file
forever.} 37. Rb7 {[%csl Rb1,Rb2,Rb3,Rb4,Rb5,Rb6][%cal Gc3b2,Gb2a3] and with
my King stuck in this prison, I can make no progress.}) 37. Ra2 {and I need to
find a new way forward, likely a3 like above.}) 36… Kb4 37. Ra3 Re8 {
Threatening Re2+, winning his Rook.} 38. Rf3 a3+ $6 {This move way well throw
the game away, as the computer isn’t sure if Black can make progress.} ({
Instead the following line apparently wins.} 38… Re2+ 39. Kb1 f6 40. h4 h5
41. Rg3 Rxf2 42. Rxg7 f5 43. Rg5 Kb3 44. Rg3+ Kc4 45. Rg5 Kb4 46. Rxh5 (46. Rg8
f4 47. Rg5 Kc3 48. Rg7 Rf1+ 49. Ka2 Rh1 {and Black wins another pawn. What
long analysis the computer gives, and how hard most of these moves are to find!
}) 46… Kb3 {Threatens mate, forcing the King away from the Queening square.})
39. Ka2 Re2+ 40. Ka1 f6 41. h4 {It’s extremely hard to make progress, mostly
because the White Rook stops my King from getting closer. I try to get rid of
it.} Rc2 {Though this move apparently leads to a draw!} (41… h6 42. Rg3 Rxf2
43. Rxg7 {is the only attempt at winning, says the comp, but even here it
doesn’t look obvious.}) 42. g4 $2 Rc3 {This sets a trap, which my opponent
mercifully falls into. The only move is Rf5, so that Kb3 is met by Rb5+.} 43.
Rf4+ {The most obvious move loses.} Kb3 {Threatening mate.} 44. Kb1 a2+ {
And it’s mate next move. Despite being several hundred rating points above my
opponent and having an advantage almost from move 3, I very nearly blew this
game. Endgames, they are the great equalizer.} 0-1

Conclusion

There was a Soviet saying concerning American and other Western chess players: They play the opening like masters, middlegames like experts and endgames like amateurs.  This game really showed that: I played the first 15 moves perfect according to the computer.  There was then some wobbles as we transferred deeper, and the less pieces the more wobbles there were.  In the end, if my opponent doesn’t fall for the mating trap, I’m not even sure if I win this game.

Endgames are truly the great equalizer.  I played very good chess and nearly blew it because, once the pieces started coming off, I didn’t know what to do.  My opponent kept hanging around, and he may even have hit a drawn endgame at some point.  It’s hard for me to say, because I don’t know endgames well enough.  I was rated 1800 at the time and over 2100 now, and I still don’t know these endgames.

This is why endgames are so important.  If my opponent knew endgames just a little bit better, he doesn’t lose this game.  I’m tired of that. I’m tired of drawing games I should win, of losing games I should draw, of taking 60 moves to win a game when I needed less than half.  I’m beyond tired, and once I get time to study chess again, to truly study, I’m 100% looking at endgames.

THAT is my main conclusion.  All the other moves mean nothing if I can’t convert a pawn-up endgame after a dominating advantage.

Leave a Reply

Your email address will not be published.