Category Archives: Endgames

The Wrong Rook (2)

“We are our choices”
Jean-Paul Sartre

More than a year ago I wrote an article on the same subject. You can review it here and that could help you figure out the solution to this puzzle as well.


You could say “But this one has 2 extra pawns in it”, so let’s look into why those pawns are on the board. The position has equal material. Re3, Re2 and Rb3 are in a standoff, all being under attack one way or another. An exchange leads to a simple draw since both pawns can either do damage or be captured as shown. In the same time Black’s rook battery along the 3rd rank protects its king from being checkmated and keeps an eye on the f6-pawn.

You might get the feeling in the beginning those pawns are important. Both of them are passed and on the 6th rank. The White king is not in the a3-pawn square, while the Black king is in the f6-pawn square (please review lesson 26, level 3 of our app). The a3-pawn cannot advance at the moment; the f6-pawn can and Black could catch it by moving either the rook or its king. If Black wants to catch it with its rook (1. f7 Rf3), it has to consider Rb3 is under attack and would be lost. That means the only move it really has is 1. f7 Kg7 Next we should look at what White can do about its pawn. Defending it 1. f7 Kg7 2. Rf2 Kf8 leads nowhere fast, so what about promoting it?


Now we have reached a similar situation with the other puzzle. White has sacrificed its pawn and all it has left is to attack the king. Should it do it with 1. Rf2+ …, 1. Rf1+ … or it does not matter? If the king goes toward the h8-corner, White wins no matter what because like in the other puzzle Black loses a rook. We also see in the process why the a3-pawn is needed, as the White king uses it to hide from checks (see line A). Now we look at what happens if the Black king goes in the center and we could observe quickly the difference between having a rook on e2 or not (see line B). Going back to the main line, we conclude it matters which rook is used to check with; one move leads to a draw and the other one to a win. Hope you have enjoyed it.

Valer Eugen Demian

Chickening Out

By now the league season had finished but we were still in the cup, facing Surbiton in the semi-finals. I had yet another white, against former RJCC member Jasper Tambini, who was graded 185 at the time, but is now 202.

I wheeled out my trusty QGD Exchange. Here’s what happened.

1. d4 d5
2. c4 e6
3. Nc3 Nf6
4. cxd5 exd5
5. Bg5 Be7
6. e3 h6

An unusual move order. Black usually plays c6 or O-O here.

7. Bh4 c6
8. Qc2 O-O
9. Bd3 Re8
10. Nf3 Nbd7
11. O-O Ne4
12. Bxe7 Qxe7
13. Nd2

White usually heads for the minority attack with Rab1 here. Bxe4 is another popular choice. My plan of trading knights on e4 shouldn’t give me anything.

13… Ndf6
14. Ndxe4 dxe4
15. Be2 Nd5
16. Nxd5 cxd5
17. Rac1 Qg5
18. Qc7 Re6

He could have played Be6 here, intending to meet 19. Qxb7 with Bh3. Tactical points like this are always important. Calculation in chess is more about spotting this sort of idea than ‘sac sac mate’. Now I might have tried 19. f4, but instead, predictably, head for the ending.

19. Qg3 Qxg3
20. hxg3 Rb6
21. b3 Bd7
22. Rc5 Bc6
23. Rfc1 a5
24. f3 exf3
25. Bxf3 a4
26. bxa4 Rxa4
27. R1c2 Ra3

White’s attacking the black d-pawn while Black in turn targets the white a-pawn. It’s still equal.

28. Kf2 Rba6
29. Bxd5 Bxd5
30. Rxd5 Rxa2

An alternative was 30… Rf6+ 31. Ke2 Re6, switching his attention to the e-pawn.

31. Rd8+ Kh7
32. Rxa2 Rxa2+
33. Kf3 Kg6
34. Rd6+ f6
35. Rb6 Ra7

This is clearly a mistake. Black should give up the a-pawn to remain active rather than moving his rook to this poor square. 35… h5 36. Rxb7 f5 and Black is holding. One idea is Kf6 followed by g5, g4+ and Rf2#, although of course White isn’t going to allow this!

36. e4 Kf7
37. Kf4 h5
38. e5

Giving Black some counterplay. 39. d5 should have been preferred.

38… Ra4
39. Rxb7+ Ke6
40. Rb6+ Kf7

Losing a vital tempo. 40… Ke7 still offered drawing chances.

41. e6+ Ke7
42. Ke4 g6

At this point Jasper unexpectedly offered a draw. My emotions were conflicted. Regular readers, as well as anyone who knows me in real life, will be aware that I’m almost always happy to agree a draw, regardless of the position. As my opponent is a former RJCC member and we’ve always been very big on cultivating sportsmanship, I’d assume he would only offer a draw if he thought he could hold the position. Offering a draw in a position you know is lost when your opponent has enough time on the clock is, to say the least, bad manners. It seemed to me like a position in which, whether or not I was winning, I could press without any danger of losing. But then I became tormented by negative thoughts. Perhaps I would freeze and end up losing on time. Perhaps he’d capture my g-pawns and his pawns would start advancing towards promotion. Perhaps, perhaps, perhaps.

In fact the position’s an easy win for White, as long as I find some fairly accurate king moves to escape the black rook’s attention. For example: 43.Rc6 Rb4 44.Kd5 Rb5+ 45.Kc4 Rb2 46.d5 Rc2+ (or 46…Rxg2 47.Rc7+ Kd8 48.e7+ Kxc7 49.e8Q) 47.Kb4 Rb2+ (or 47…Rxc6 48.dxc6 Kxe6 49.Kc5 with a winning pawn ending) 48.Kc3 Re2 49.Kd3 Re1 50.Rc7+ Kd6 51.Rd7+ Kc5 52.e7 Re5 53.d6 and the e-pawn will eventually promote.

But of course I agreed the draw. Meanwhile, although we were heavily outgraded on all but the top board, a couple of the other games went in our favour. We lost the match 3½-2½, but if I’d played on and won, we’d have drawn 3-3 and gone through to the finals on board count.

What else could I say? A lot, actually, but not now.

Richard James

The Heffalump’s Escape

For my penultimate game of the season I was paired against a formidable opponent in Alan Perkins, joint British U16 Boys Champion in 1965 and student international in the 1970s. In recent years he’s preferred to ply his trade in the calmer waters of the local chess leagues.

My archives remind me that we first met 40 years previously in a weekend congress when I managed to draw. I was slightly worse in the final position but quite probably ahead on the clock. We met again in 2010, in another match between Richmond B and Ealing A, when I lost.

In both games I had white and faced the King’s Indian Defence, trying the Saemisch Variation in 1977 and the Smyslov Variation in 2010. In 2017 I was again White, and it was another Smyslov Variation.

1. d4 Nf6
2. c4 g6
3. Nc3 Bg7
4. Nf3 O-O
5. Bg5 c5
6. d5

I’d learnt from my earlier game against Mike Singleton to play d5 here.

6… d6
7. e4

Looks natural, but the stats suggest that e3 is to be preferred. It’s slightly more popular, the choice of stronger players and has a much better percentage. Maybe next time.

7… h6
8. Bh4

Again natural, but the stronger players – and the stats, prefer Bf4.

8… e6
9. Be2

Again, Nd2 is the expert move.

9… exd5
10. exd5

Choosing a King’s Indian rather than Benoni formation. An unpopular decision which also scores poorly for White. If you’re playing a 2200 strength opponent it helps if you know some theory!

10… Qb6

A new move. 10… g5 11. Bg3 Nh5 is the recommended plan.

11. Qd2 Bf5
12. O-O Nbd7
13. h3 Rfe8
14. Bd3 Ne4
15. Bxe4 Bxe4
16. Nxe4 Rxe4
17. Rac1

White has problems with the long diagonal but could defend tactically: 17. Qc2 Rae8 18. Qa4.

17… Rae8

Here there was no reason why Black couldn’t have taken the pawn: 17… Qxb2 18. Qxb2 Bxb2 19. Rb1 Bg7 20. Rxb7 Nb6 and c4 will fall while Black can hold d6.

18. Rfe1

There was no reason not to play 18. b3 here, and no reason for Black not to trade rooks and then take on b3. We’d both misjudged the position.

18… f5
19. Bg3

Again, I could have played 19. b3 and he could have captured the pawn.

19… g5
20. Rxe4 Rxe4
21. Nxg5

21. b3 was still about equal. I was concerned about my bishop being buried alive after 21… f4 but I can always play g3 at a convenient time.

Instead I lash out with a ridiculous sacrifice, hoping to get three pawns against a piece. Or perhaps, aware of Alan’s tendency to get into time trouble, trying to lure the heffalump into a swamp in a deep dark forest. You decide.

21… hxg5
22. Qxg5 Qxb2
23. Bxd6 Qf6

I’d missed this simple defence. I might have played 24. Qg3 to keep the queens on, but instead traded.

24. Qxf6 Bxf6
25. Rc2 Be5
26. Bxe5 Nxe5
27. f3 Rxc4

Now I only have one pawn for the piece. Time to resign?

28. Re2 Nf7
29. Re8+ Kg7
30. Re7 Rc2

30… b5 was the easiest way to win.

31. Rxb7 Rxa2
32. d6 Kf6
33. d7

My passed pawn reaches the seventh rank. Black will have to be a bit careful.

33… c4
34. Rc7 Ra4
35. g4 fxg4

35… Ke7 was the way to go: 36. d8Q+ Kxd8 37. Rxf7 c3 and the white rook can’t get back.

36. fxg4 Nd8

Now Ke7 doesn’t work because the white rook can return via the f-file to stop the c-pawn.

37. g5+ Ke7
38. g6 Ra6
39. g7 Rg6+
40. Kf2 Rxg7
41. Rxa7 Kd6
42. Ke3 Kc5
43. Rc7+ Kd5

The territory’s becoming swampy for Black now as he only has one pawn left and the white d-pawn is surviving. The only path to victory here was 43… Kb4, but it’s not so easy in the quickplay finish.

44. h4 Rh7
45. Ra7

The wrong plan. The way to hold was to get the white rook to the eighth rank to have access to the b-file. So: 45. h5 Rxh5 46. Rc8 Rh8 47. Rb8/Ra8 and there doesn’t seem to be any way for Black to make progress.

Now 45… Kc5 would have put Black back on track, but instead he captured the h-pawn.

45… Rxh4
46. Ra8 Rh3+
47. Kd2 c3+
48. Kc2 Kc4

48… Rh8 was a simple draw, and even Kd4 was good enough. But instead the heffalump tumbled head first into the swamp. Will the tiger put the boot in and score an unlikely and, frankly, undeserved victory?

49. Rc8+

Sadly not. All I had to do to win the game from here was to play one of the most obvious moves in the history of chess: 49.Rxd8 Rh2+ 50.Kb1 Rh1+ 51.Ka2 Rd1 52.Rc8+ Kb4 53.d8Q Rd2+ 54.Kb1 c2+ 55.Kc1 Rxd8 56.Rxd8 and wins. For some reason (or for no reason at all other than having to blitz during a mutual time scramble) I had a brainstorm and decided I needed to check before rather than after capturing the knight.

49… Kd4
50. Rxd8 Rh2+
51. Kb3

Moving up the board because I was scared of mate threats. This is fine but Kb1 and Kc1 also draw, although Kd1 loses. Ironically, without the white pawn on d7 the draw would be automatic.

51… Rb2+
52. Ka3 Rb7
53. Rh8 Rxd7
54. Rh4+

54. Kb3 was again an automatic draw. He could only prevent Kc2 by playing Kd3 when I can just play Rh3+.

54… Kd3
55. Rh3+ Kc2
56. Rh2+ Rd2

At this point I stopped recording my moves. I’m not sure what I played here but it certainly wasn’t Rxd2. The position’s still drawn but it’s easy for White to go wrong now, which is what happened, and Alan just about had enough time to force checkmate.

An exciting ending which I certainly should have drawn, and was, for just one half-move, winning. I really shouldn’t have been allowed to get that close. Perhaps randomising the position on move 21 was justified even though it was an awful move.

You might also think that trying to play a proper game of chess in 2½ or even 3 hours is ridiculous. I agree, but I also think both adjournments and adjudications are ridiculous.

Richard James

Inexplicable Endgame Play

“If you are weak in the endgame, you must spend more time analyzing studies; in your training games you must aim at transposing to endgames, which will help you to acquire the requisite experience”
Mikhail Botvinnik

This week’s endgame comes from a voting match we played as part of one Canadian team during an 8 months period. The team componence (46 players for us versus 6 players for them) seemed to favor us by quite a bit, still getting things organized as a team with so many players is not easy to do. We are getting better at it as time goes on. We have far less “drive-by” players (those who just vote for any move they think of, even moves never discussed) and we have managed to prove to our regular team members that discussing our options before we start voting actually pays off. In this particular game we managed to overcome a so-so opening and shaky middle game play into the following endgame position (White to move):

The general consensus here was that despite the extra pawn, we had no chance to win at correct play. I was one of the members interested to offer a draw, but the team decided to play on. It turned out to be a very interesting experience. Do you agree the position should lead to a draw at correct play? Here are a few reasons for it:

  • The extra pawn is doubled and even if they are center pawns, as long as they stay doubled they are of little use
  • The double rook endgames are far more tactical because of the existing fire power and both kings need to be protected
  • The important h4/h5 pawn moves have already been played, establishing clear boundaries on what those pawns can do
  • White’s plan should be very simple here: take control of the 2nd rank and put pressure on the e5-pawn with both rooks to impede its advancement

Instead of the above White chose firstly to bring his rook onto the 7th rank. Of course an (un)written rule says the best position for any rook is on the 7th rank. We actually have the opportunity to see how any of these rules cannot be applied without making sure the situation on the chessboard warrant them.

The above mistake was important but not decisive. Letting us take control of the 2nd rank, the same idea they tried at the wrong time, made absolutely no sense. That also meant we now had a clear path toward winning. Some may say this second mistake allowed us to win it; in reality they were both connected. The remaining of the endgame was more or less technical. Enjoy the winning line and hope you will learn a bit from it. You never know when your opponents might offer you the opportunity to punish their endgame mistakes in inexplicable fashion.

Valer Eugen Demian

“What say you?” The 1 minute challenge (7)

“A wise man can learn more from a foolish question than a fool can learn from a wise answer”
Bruce Lee

A quick reminder about how to do it:

  • Have a look at the position for 1 minute (watch the clock)
  • Think about the choices in front of you and pick the one you feel it is right
  • Verify it in your mind the best you can
  • Compare it with the solution

Are you ready? Here it is: which of the three possible king moves wins the game for White?

This is a very difficult endgame; if you are able to figure out the guiding ideas and guess the correct move, consider yourself a strong player. My take on it:

  • Material is equal
  • White has a chance to win because of the better position of its king
  • It is easy to see White could win the a-pawn, bringing the position into a basic king and pawn versus king endgame
  • If White captures Black’s pawn with its own pawn, key is to hold control of the critical b7-square: if White has it, it is a win; however if Black has it, it is a draw (lesson 19, level 2 of our chess app)
  • If White captures Black’s pawn with its king, it has to place the king in front of the b-pawn in order to win (lesson 19, level 2 of our chess app)
  • On the Black side of it its king should either consider taking control of the b7-square (see above) or attack and capture White’s b-pawn to reach a draw

With the above in mind, we have to choose a move. Which one did you pick? Have a look at the solution and go over it with care to understand all the twists and turns.
The highlight: white uses the opposition to force the black King all the way to the h2-square. That saves the b-pawn, allows white to capture the a-pawn and reach a won endgame. It is kind of amazing how two simple concepts combined together can create such a complicated puzzle, isn’t it? Enjoy!

Valer Eugen Demian

A Successful Wager

My next game involved a trip along the motorway to Maidenhead, the Thames Valley League’s furthest outpost, where I had black against their top board, John Wager, a strong and experienced player graded nearly 30 points above me.

He chose the Colle System, a popular opening in these post-theory days, just getting your pieces out and setting up a flexible pawn formation ready for action in the middle game.

1. d4 Nf6
2. Nf3 e6
3. e3 c5
4. c3 b6
5. Nbd2 Bb7
6. Bd3 d5
7. O-O Nbd7
8. b3 Bd6
9. Bb2 O-O
10. Qc2 Rc8
11. Rac1 e5
12. dxe5 Nxe5
13. Nxe5 Bxe5
14. Nf3 Bb8
15. Bf5 Rc7
16. Rfd1 Qe7
17. c4

This is the critical period of the game. Something I wanted to write about at some point, because I find it difficult myself, is the whole idea of compensation. As a naturally cautious player myself I tend to be very materialist. Here, I might have considered a pawn sacrifice for attacking chances. I start with 17… d4 18. exd4 Bxf3 19. gxf3 Rc6 20. d5 Qd6 21. dxc6 Qxh2+ 22. Kf1 Re8 23. Be4 Nxe4 24. fxe4 Qh1+ 25. Ke2 Rxe4+ 26. Qxe4 Qxe4+ 27. Kf1 when Black has queen and pawn for two rooks, and, with care, will eventually be able to pick up the c6 pawn. White can do better by not taking the rook: 21. f4 Qxf4 22. f3 is equal according to the engines. But this, at my level, is very much a computer line. Would a grandmaster have played d4 here, and how much would they see? I’m not sure.

17… g6

I chose this natural alternative, which leaves Black, rather than White, with doubled f-pawns.

18. Qc3 gxf5
19. Qxf6

He didn’t have to take this immediately: it wasn’t going anywhere in a hurry. Instead simply 19. cxd5 and White has an extra pawn, but Black might want to claim some compensation in the shape of the two bishops. Enough? I don’t understand chess well enough to tell you.

19… Qxf6
20. Bxf6 dxc4
21. bxc4

You might think Rxc4 looks more natural here. There again you might not…

21… Rc6
22. Ba1

So the bad news is Black has doubled isolated pawns, while the good news is that he has two raking bishops.

22… Rg6

This is where things start to go wrong for me. This seems to me, at least superficially, a very obvious move, setting up a pin and planning a later f4 to undouble my pawns. But, as you’ll see, it’s not correct. My computer tells me 22… Re6 followed by f4 was correct, with perhaps a slight advantage.

23. Nh4 Rg4

Continuing along the wrong path. I had to play Rg5 here but I’d missed a simple tactical point.

24. g3

At this point I realised that my intended f4 would be met by Nf5 with an immediate win for White. It was still possible to swallow my pride and play Rg5 to keep the pawn. Re8 was a better try for compensation than my choice.

24… Bc8
25. Rd5 Be6
26. Nxf5 Bxf5
27. Rxf5 Rd8
28. Rd5

Trading when you’re ahead, but the computer is not impressed. Now I can tie his rook down to defending the a-pawn.

28… Rxd5
29. cxd5 Ra4
30. Rc2 Bd6
31. f4 Re4

Not a good idea. 31… Kf8 followed by Ke8 gives drawing chances.

32. Kf2 b5
33. Kf3 Ra4
34. e4 Ra3+
35. Kg4 b4

The final mistake. 35… c4, giving my bishop some room, was the only way to stay in the game. Now White’s centre pawns go through.

36. Bf6 Rd3
37. e5 Bf8
38. d6 Bxd6
39. exd6 Rxd6
40. Be7 1-0

Tell me, why did I lose this game? At one level I was just beaten by a stronger player. Although it wasn’t technically the losing move, my problems started with 22… Rg6, which I played because I hadn’t foreseen the knight’s journey to h4, f5 and h6.

Richard James

Mednis Principles (2)

“With major pieces (queen or rook) on the board, having bishops on opposite colors favors the side with an attack.”
Edmar Mednis

A couple of nice articles about these principles can be reviewed HERE and HERE
SIM Michael R Freeman is a very strong ICCF player from Darwin, New Zealand. The fact he has been able to perform at around 2500 correspondence chess rating since 2009 is a high accomplishment not many are capable of. He is also FIDE-CM over the board and this summer he had the opportunity to play in the 8th IGB International Seniors Open Chess Championship 2017, Malaysia. Michael was kind to share interesting positions from his games along the way and I liked one in particular. The position was extremely interesting and the additional thoughts and comments by Michael caught my attention and made me take a closer look. We also had a very instructive online discussion about it; in the end can say for sure I learned more about opposite bishops endgames. Here it is with comments as indicated:

Of course Mednis principles apply here perfectly. White was pressing all along and that opened the door for a nice ending. Personally I think this is also an excellent example of how we need to pay attention to what is going on until the opponent has signed the scoresheet. Michael could have been rattled by the missed chances or by the tough defence he had to face up to that point; also he might have thought this was a done deal with those 2 passed pawns ready to promote. Any of us in his shoes would have had to consider 1… Bc5 as the best reply and gather our last drops of energy to figure out the winning idea with Bb5 hanging. He did it and was rewarded for it. Below is the full game score. Thank you Michael for sharing it with us!

Valer Eugen Demian

“What Say You?” The 1 Minute Challenge (4)

“A wise man can learn more from a foolish question than a fool can learn from a wise answer”
Bruce Lee

A quick reminder about how to do it:

  • Have a look at the position for 1 minute (watch the clock)
  • Think about the choices in front of you and pick the one you feel it is right
  • Verify it in your mind the best you can
  • Compare it with the solution

Are you ready? Below is this week’s position asking you to choose the next move for White. What is the most likely result based on your choices?

Here are my thoughts:

  • Material is equal
  • Each side has a passed pawn and both kings are within reach (see rule of the square in our app level 3, lesson 26) with a plus for the Black king being closer
  • Pushing the pawn forward 1. c5 … gives Black time to activate its king (1… Ke6 for example). White will have to capture the a-pawn, while Black will do the same with the c-pawn; after those pawns come off the board, Black’s king will be closer to capture the g4-pawn and promote its remaining f6-pawn (see basic pawn endgames in our app level 2, lesson 19). Black could win in this case
  • Based on the above idea 1. Kd4 … does not look like a good idea either because it also allows 1… Ke6. The difference in this case might be the fact the White king stays close to its c4-passer, it could capture Black’s a-pawn and come back in time to defend the passer; hmm, this is an interesting thought after all
  • Continuing along this line of thought 1. Kd5 … looks the best since it is keeping Kf7 away. The problem here is that after 1… a4 there is no other response but the forced 2. Kd4 … or the a-pawn promotes. We are now back to the previous line with the a-pawn farther down the board. This will draw the White king away and allow the Black one to activate; after the simple moves 2… Ke6 3. Kc3 Ke5 4. Kb4 Kd4 there is nothing better for each side than pushing pawns down toward promotion and a draw
  • Going back to 1. Kd4 Ke6 and using the information gathered in the 1. Kd5 … line, this must be the move to play. The a5-pawn has not moved yet and black must choose between moving it or bringing its King closer; in both cases this is good news for white

Conclusion: 1. Kd4 … gives white the best practical chances and should be played. You might be out of time by now to be able to determine if white can win this or not. The endgame has one more nice wrinkle white must consider in order to win and you can see it looking at the solution below:

What can we conclude out of it? Sometimes we might have to go ahead and play the most promising line even if we don’t see the final result for various reasons. We must keep our focus and apply our thought process along the way to uncover opportunities and achieve the best possible result. Start with the simple stuff first and build on it based on your knowledge; wherever your knowledge stops, mark it down and make sure you focus on expanding it during your home preparation. Good luck!

Valer Eugen Demian

King Power in the Endgame

Following up my earlier column about king centralization, here’s a great example of a strong king being decisive in the endgame. It was also a huge upset, with Vishwanathan Anand losing to a player 259 Elo points lower than himself!

Sam Davies

Knowing the Endgame for Better Trades

Beginners only tend to be interested in learning endgames if you can show them how useful they are. And for successful coaching it is highly desirable that your students are interested in learning!

Accordingly this article is aimed at beginners only. I will demonstrate how knowing or learning an endgame can help in making the right exchanges.

Black to play and win
This position is taken from one of my internet games. This position looks even at a first glance but it is completely winning for Black.

Hint: The king and pawn endgame is the best way to realize the material advantage.

Solution:

1…f5

Forcing the knight to c5.

2.Nc5

The only move. Now what?

2…Nxf2

The simplest solution that forces exchanges.

If 2…Bxc5 then 3. Bxc5! drops a pawn but retains drawing chances for White. The same is true of 3.Nxc5. Note that 3.bxc5 leads to a winning king and pawn endgame for Black after exchanges on f2 because White’s king can’t protect the c5 pawn and the pawn is within the reach of Black’s king (Rule of Square).

3.Rxf2 Rxc5!!

The point, winning a pawn and forcing exchanges. My opponent resigned here in view of 4.bxc5 Bxc5 when the rook is pinned. Black can take the rook on next move and resulted position will be an easy win because of the extra pawn in a king and pawn endgame.

Ashvin Chauhan