The Queen vs Rook Endgame

Why choose to study a game that occurs about 1 in every 1200 games?

After searching for something that wouldn’t take long I realized that this endgame cannot be mastered quickly. In my pursuit, I found which has a fifteen chapter course on the Queen vs. Rook endgame!!

Again, why would one want to go through such a course for a 1 in 1200 chance at using it? While I just have a thirst to master any and everything in chess, I think there is valuable lessons in simply how the rook and queen can battle each other across the board.

For the rest of you out there who may like the smallest tidbit of knowledge about this game, I would recommend learning the Philidor position with black to move and also how to get to the Philidor position with black to move if it is white to move instead.

Philodor’s Position

That is covered in chapter 2 and is essential.

If it white to move and you want this position with black to move, you can do this in a few ways. Here is one.

  1. Qe5+ Ka7 or Ka8 (since 1 … Kc8 2 Qe8# and 1 … Rc7 2 Qxc7+ with mate to follow) 2. Qa1+ Kb8 (since 2 … Ra7 3 Qh8#) 3. Qa5

Now, you’re in the above position with black to move. There are several options, some lose quicker than others. All are worth studying, the hardest ones being those when the rook moves to b3, b1, f7, or h7.

Anyway, the main purpose of the post is to introduce you to the valuable resource which has several courses that may go into much more detail than what you can find on or

Real Endgames

Last week I paid attention to an interesting endgame in which I felt that the game was a draw. After analyzing it further, with white to move, it was white to win. Study the game board below and see if you can find the winning move.

Only one move will win for white

There are several possible moves to consider. Let’s throw out a few right away. Pawn to g3 or g4 will allow black to capture (regular or en-passant), with no way for white to stop the promotion. Forget those.

White’s king capturing the pawn on f4 leads to … stalemate!  Black has no legal moves and the game is a draw.  We will forego analyzing king to e6 or king to f6 as we’ll see those moves would lead to the same consequences as other moves. Let’s look at king to e5, king to e4, and king to g6 in that order.

If 1. Ke4, then black has two options. Let’s only look at the threatening 1 … Kg3. The white king will have to step away from the pawn on f4, as the e3 square is protected, or maintain pressure on that pawn by retreating to e5 or f5. Either of which will lead to black taking the pawn on g2. The line could continue 2. Ke5 Kxg2 3. Kxf4 Kf2 and now white has to struggle to maintain defense of the f3 pawn while dealing with the advancement of the h5 pawn. This looks like a stalemate. 

If 1. Ke5, then the same sequence of moves above can occur and stalemate is likely without further huge blunders.

How can there be any hope in moving the king to g6? Well, before we give up for a stalemate, let’s take the patience to analyze before offering that white flag up.

With 1. Kg6 black is forced to move 1 … Kg3, which seems fine with him, as he is attacking the g2 pawn. Now consider what happens after 2. Kxh5 Kxg2.  White has the very effective move 3. Kg4! attacking the black pawn and protecting the f3 pawn! With black having no way of protecting its pawn, white will capture on the next move and carefully bring it down the file for promotioon and the win.

But what if 2 … Kf2 for black?  He opts not to capture the g2 pawn to provide protection to his f4 pawn after 3. Kg4 Ke3. Now white has the weapon of advancing his g pawn.  After 4. g3 fxg3 5. Kxg3, there is no way to stop white from advancing his f pawn for promotion (unless white makes a blunder). 

A Way to Get Better

When I was training jiu-jitsu in Kansas, one of my mentors encouraged me to record myself as I rolled (what jiu-jitsu practitioners call sparring) and then go over it later. You can discover small adjustments that can vastly improve the game. Likewise in chess. If we want to get over an improvement plateau, we should record our games. Obviously, the act of recording by itself isn’t going to help. We need to go over it later.

A way I get better is to use the free Stockfish engine on the analysis board provided by Take last Wednesday, for example. I played a game in which I offered a draw that was accepted after the 31st move. I was white.

Analyzing a Game

I like to quickly play out the first several moves and then begin thinking about any move that may have been slightly better and then toggle the engine on/off to see what move was better. Moves that don’t move the advantage by 3 or more points I shrug off at our level of play. If I were a national master or something, I may pay more heed to 1 point swings. Here was how the game started.

1. e4 e6 2. d4 d5 3. Nc3 c5 4. Nf3 Nf6 5. Bg5 Be7 6. dxc5 Bxc5 7. exd5 exd5

White has an advantage in this position

It was at this point I made my first blunder by taking the d5 pawn with my knight. In my mind, I was thinking how he wouldn’t take it with his knight because he would lose his queen. Why I didn’t consider that he would just take my knight with his queen was the blunder.

8. Nxd5?? Qxd5

The game went from white having about a 2.5 pawn advantage to black having a 4.5 pawn advantage here. So, what was the better move to maintain the advantage. In hindsight, it is easy to see that I needed to remove the defending knight with 8. Bxf6 first. Whether black followed with 8 … gxf6 or Qxf6, we could then safely take the pawn with 9. Nxd5 and be a pawn up in material with about a 3 pawn advantage (according to Stockfish). The game continues:

9. Qe2+ Qe6 10. Bxf6 gxf6 11. Qxe6 fxe6 12. g3 Nc6 13. Bh3 Nd4 14. Nxd4 Bxd4 15. c3 Bb6 16. 0-0 Bd7 17. Rfe1 f5

I want to pause here because it was at this point, I thought I found a clever move after the f5 pawn push. The point of black’s f5 move was to provide a layer of protection on his g6 pawn, as white had both the rook and bishop attacking with only his bishop defending. With the g6 pawn pinned, I played 18. Bxf5. It turns out that this was not that great of a move. Black needed to respond with18 …Rf8 with the idea of 19… Bxf2 to follow. Instead, black has a small blunder with a queenside castle 18. …0-0-0. After 19. Bxe6 Bxe6 20. Rxe6 I (white) found myself coming back.

White has a fighting chance!

Black’s best move here to stay on top is 20… Rd2 with a double attack on that f2 pawn along with an attack on the b2 pawn. White will definitely be losing a pawn with whatever move would follow. Instead, black decides on the next small blunder 20…Rhe8 allowing white to get to an even game.

For some reason, I did not see that it was best to exchange the rook right away, which would have been in my best interests. Instead, I opted for the 21. Rae1. The game continued

21… Kd7 22. Rxe8 Rxe8 23. Rxe8 Kxe8 and the game was even.

My three extra pawns compensate for black’s bishop

We played this even game until I offered a draw. It turns out, however, that my last move was terrible and black should not have accepted! Take a look at the position after the accepted draw and see if you can find out why.

24. h4 Ke7 25. Kg2 Kf6 26. g4 Kg6 27. f4 h5 28. f5+ Kh6 29. g5+ Kg7 30. Kf3 Bc7 31. b4?? b5.

Black has great opportunities

By opting for b5 instead of something like a5, I establish my left side pawns on the dark squares! The bishop has attacking opportunities on the weak pawns on the third rank. With whatever I decide on my next move, black can begin attacking either the c3 pawn or the h4 pawn or threaten to attack both, and white’s king would be overloaded.

Closed Sicilian

March is a great month for chess as it will bring 5 Wednesdays on which our wonderful club can meet and play and learn some chess. Come on out to The Landing anytime in the 6-9pm window tonight (3/23/22) for some chess games and fun.

In the lastest American Chess Magaizine (Issue #25), the closed Sicilian is explored. The Sicilian is signified by the moves 1. e4 c5, which is initiated by black after white’s king pawn opening. An “Open” Sicilian is signified by the next move 2. Nf3, while the “Closed” Sicilian is signified by 2. Nc3.

Closed Sicilian

While the Sicilian is intiated by black, white can then initiate an open or a closed Sicilian with the second move. Here is one such opening line for the closed Sicilian.

Main Line for a Closed Sicilian

Although the main line is given at the bottom of the picture above, it can also be arrived at through the following:

1. e4 c5 2. Nc3 d6 3. g3 Nc6 4. Bg2 g6 5. d3 Bg7

Yet another alternative:

1. e4 c5 2. Nc3 Nc6 3. d3 d6 4. g3 g6 5. Bg2 Bg7

As you’ve noticed, we can interchange several moves to arrive at this closed Sicilian position.

So you might be asking what the next best move for white might be. Several options have been explored, with 6. Be3 being the most popular.  Other options include 6. f4 or 6. Nh3.. However, a chess engine would rather see 6. Ngce2 or 6. Nf6 rather than the previous two even though some chess experts disagree.

Saturday Special

Hello chess enthusiasts! Come out and join us for a special chess day on Saturday, 2/26/2022, beginning at 2pm at Impact Coffee. Some DCC members will be playing a few games and promoting the club. We’ll even request the fancy chess table they have for tomorrow.

As the club continues to grow, we’re hopeful to have more weekends for chess!

The big chess news lately was how the 16 year old Indian Rameshbabu Praggnanandhaa defeated Magnus Carlsen. If you’re interested in the game that they played, I found it on and produced it below using the lichess app.

Magnus Carlsen vs Rameshbabu Praggnanandhaa

Their game began with a queen pawn opening, with RP answering with the King’s Indian. The game eventually fell into what is called the Tarrasch Defense: Symmetrical Variation. The first five moves indicate why the symmetry:

1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3 c5 5. e3 Nc6

Nice symmetry!

The game continued…

6. cxd6 exd5  7. Bb5 cxd4  8. exd4 Bd6  9. 0-0 0-0  10. h3 h6  11. Re1 Qb6  12. Ne5 Rc8  13. Bf4 Re8  14. Rc1 Qb6  15. Bxc6 bxc6  16. Na4 Qa6  17. Re3 Ne4  18. f3 Ng5  19. Rec3?

This move brough them to the following position. According to a chess engine, this was MC’s first minor mistake. Better was 19. Bxg5.

The advantage began to swing to black

Only a few moves later, we would see both players make mistakes!

19. … Qb7  20. Bh2 Ne6  21. Nxc6?? Better, was 21. g4.

First significant mistake by MC

The only move to maintain the advantage for RP was 21. … Bxh2. Then, 22. Kxh2 Nf4.  However, RP decided on Nf4 first to bring the game back to even.

Play continued 21. … Nf4??  22. Ne5 Bxe5  23. dxe5 Nd3  24. Rxc8 Rxc8  25. Rxc8+ Qxc8  26. Bg3 d4  27. b3 Qc6  28. Qd2 Kh7  29. Kh2 Bg6  30. Qa5 Qc1  31. Qxa7 Qe3  32. Nc3??

Second significant mistake by MC

The better move to keep things close would have been 32. Qa5. At this point RP took advantage and never faltered.

32. … Nf4  33. Nd1 Qd2  34. Nf2 Ne2  35.  h4??

Third significant mistake my MC

To keep a somewhat better game going, white has some options after 35. Ng4 h5  36. Qe7. Again, RP capitalizes on the mistake, and play continues with

35. … Qe1  36. Qd7 Nxg3  37. Qxd4 Nf1+  38. Kh3 Ne3

Two pawn advantage for white here is an illusion.

Although white (MC) has a two pawn advantage at this point, the chess engine shows a significant disadvantage for white. To prolong the inevitible, any of h5, Nd3, Ne4, Ng4, or even Qxe3 was better than MC’s next move.

39. Qb2??

The chess engine indicated checkmate in 8 moves. Here is a possible line for that: 39. … Qg1  40. Ne4 Nf1  41. Kg4 Qh2  42. g3 Ne3+  44. Kg4 Bf5+  45. Kxf5 Qh3+  46. g4 Qxf3#. Rameshbabu’s final move was actually 39. … Bc2.  Although this maintained the mating opportunity, it wasn’t as efficient as the Qg1 route. Magnus resigned.

The Skewer

This week we’ll have a sign-up sheet for Decorah Chess Club that will gather just some basic information from each of you (Name, email, phone number) so that we can keep in contact with you about Decorah Chess Club meetings and events. With the increased turnout, let’s try and follow this schedule tonight (2/16/22, 6pm at The Landing):

  • 6 – 6:15 PM : Arrive, sign-up, and set round schedule
  • 6:15 – 7 PM : First round of games
  • 7 – 7:45 PM : Second round of games
  • 7:45-8:30 PM : Third round of games
  • 8:30-9 PM : Final fast round

After signing up, we’ll get the first few rounds scheduled and alter as needed with late-comers. We’ll have 45 minutes for each game (some pairs may be able to get 2 or more games in), so the use of a timer is encouraged using 20 minutes each with a 2 second delay.  For those that finish early, we’ll have some puzzles that you can explore printed out while you wait for the next round.

If you haven’t already, please sign up with an email at the bottom of this site as well to have a weekly update delivered to your inbox. You can unsubscribe from that list at any time.

The Skewer

Yesterday’s puzzle of the day on (named Fever Pitch) was a good lesson in using a skewer! Let’s look at the puzzle to begin thinking about a good move for white.

“Fever Pitch” – 02/15/2022 puzzle on

Reading the board, we can see that white is two pawns down in the game and the rook is under attack by the d4 pawn. While the black queen has an attack on white’s knight, the rook on a1 is guarding it so that is not the issue.

What to do with the rook then? If this were a real game, you may consider Rd3 or Re5 as some moves to put your rook into safety. Or, maybe an exchange of knight for bishop with Nxb7. It turns out that all of these moves are not very good for white, and would miss out on a great opportunity! If you began to consider the rook taking the bishop with Rxe7 good job!  You may think that this is bad, exchanging a rook for a bishop, but look at what it sets up!

The skewer! After the queen takes your rook with 1. … Qxe7 we find the beautiful move 2. Ba3 which skewers the queen to the king. The only thing that can get in the way is the pawn on b5 (which is something you would have wanted to think through before playing the initial move in the first place).  If black moves the pawn on b5 to b6 to block, the bishop can simply take it since white has the protective pawn on c3.

The best move for black is to take the bishop with the queen immediately, and save a pawn. In summary, the best line for each color is the following:

  1. Rxe7 Qxe7
  2. Ba3 Qxa3
  3. Rxa3

Although difficult to see, black’s best response after white’s first move is as displayed: to take the rook.  Not taking the rook leads to gaining more material and a much better position.

New Location!

Word on the street says that The Landing is now open until 9pm on weekdays! As much as I love Pulpit Rock Brewing Company for a venue, I think it is in our best interests as the Decorah Chess Club to change to this location so that it is more open for younger people to join as well as offers more choices for those that don’t drink beer.

So, on Wednesday, 1/26/2022 at 6:00pm, we’ll be at The Landing! Hope to see you there.

King Pawn Openings – King’s Gambit

During a game at chess club last week I decided to try the King’s Gambit opening. After a small analysis afterwards, James and I noticed some mistakes we both made. I initially made a mistake in the early King’s Gambit line. Not seeing how to capitalize, it was followed by a mistake. But we learned something and wanted to share. 

If you’re willing and able, please come out and join us Wednesday, 1/5/2022, for some games & lessons in chess at Pulpit Rock Brewing Company from 6-8:30pm.

King’s Gambit

The King’s Gambit begins with 1. e4 e5 2. f4

King’s Gambit

In 24 of the 36 lines that are explored in Modern Chess Openings 15ed by Nick de Firmian, the gambit is accepted. When accepted, it seems best to follow that by bringing the king side knight to f3. Let’s see why.

First, the accepted King’s Gambit: 2. … exf4. Instead of 3. Nf3, what if we instead try to take total control of the center with 3. d4, like in the following diagram? The logic being not only to try and dominate the center, but to also open up a path to retake the pawn on f4 with your bishop!

d4 is a mistake

Why doesn’t this work? See if you can find the response for black. If you see 3. … Qh4+ good job. What should white do in that case? It turns out that 4. g3 is a huge blunder and that the best move is actually 4. Ke2 which doesn’t seem like a good move at all, as it gives up the right to castle.

Instead, white should anticipate the Qh4+ move and develop a piece to prevent such a move. Hence, the popular line is 3. Nf3. The most popular responses for black are g5 or d6, where the free Stockfish engine on favors d6.

Since the rabbit hole is quite deep for the King’s Gambit, I’ll stop with any further analysis on King’s Gambit accepted and offer a few moves from black instead of accepting.

  • Falkbeer Counter Gambit 2. … d5
  • Gambit Declined 2. … Bc5

Other answers are not as good, and should be explored by the reader if interested.

Chess Wednesdays – My Three Game Experience

We’ll be learning and playing chess at 6:30pm Wednesday 10/13 at Pulpit Rock Brewing Company. Come join us!

Last week, we played a round-robin style tournament, although we couldn’t quite play everyone. I won two of my games and lost one, and wanted to try and analyze those here. Although I don’t think the players would care if I gave their names, I’ll keep it anonymous. These games are purely amateur, and the purpose of the write up is for the experience of writing and analyzing chess matches.

First Game

My first opponent, Fifty, drew white. In my last few games against Fifty, I noticed he’s been bringing out his queen early. This can be a bit aggressive and is not advised against good players, as you find yourself needing to move the queen around a lot. Our game started 1. e4 e5 2. Qh5 Nc6 3. Nf3 g6 4. Qg4.


I’m not sure if it was the 15 minute timer ticking away while I recorded the moves, but I neglected to see the easy double attack on the queen and e4 pawns with 4. …d4. This theme would continue as I never did see Black’s light squared bishop attacking the queen! Play continued with 4. …Nf6 5. Qh3 Nxe4 6. d3 Nf6 7. Bh6 Bxh6 8. Nc3??. This was Fifty’s major blunder. I was expecting Fifty to take my bishop with his queen to keep things close, but he gave me an out.

I quickly retreated my bishop out of harm’s way and play continued. 8. … Bg7 9. Nb5 0-0 10. c3 d5 11. Be2?? This was Fifty’s next blunder leaving his queen open for the taking by the light squared bishop.

But guess what? I didn’t see it! WTF?? 11. … Bf5?? I have no idea how I missed this. 12. g4 Bxg4 13. Qg3 Re8 14. 0-0 a6 15. h3?? Here again, Fifty didn’t calculate the exchange. Instead of retreating his knight, he attacks my bishop, which just forces me to exchange his knight for my bishop before stealing his knight.

15. … Bxf3 16. Bxf3 axf3 17. d4 exd4 18. cxd4 Nxd4 19. Re1 Nxf3+ 20. Qxf3 Rxe1 21. Rxe1 Rxa2? The pawn was too tempting. Better play would have been 21. … Ne4.

22. Qb3 Qa8 23. f4?? Fifty should have taken the pawn I left hanging with 23. Qxb5.

It didn’t take much longer before Fifty abdicated. 23. … Ne4 24. h4 Rxb2 25. Qxb2 Bxb2 (I DID see this bishop attack) 26. Kg2 Qa3 27. Rf1 Qg3+ 28. Kh1 Qxg4+. White resigned.

There was a mate in 3 from this position with 29. Kg2 Qg3+ 30. Kh1 Bd4 31. f4 Qh3# [31. Rf3 Qg8#] [31. Rf2 Nxf2#]. In fact, if 28. … Bd4, I had mate on the next move.

Second Game

My second game was against The Advocate, who plays quite a bit more than the rest of us. He is very good at timed games! The Advocate drew black, but was able to take advantage of my early blunders using a French variation of the Sicilian. The game began 1. e4 c5 2. Nf3 e6 3. Nc3 d5 4. Bb5+ Bd7 5. Qe2??

Much better was 5. Bxd7 as this would have maintained a small advantage for white. However, The Advocate played 5. … a6 which gave me a good chance to get back in the game with 6. Bxd7. Alas, I didn’t want to take the bishop for some reason. 6. Ba4 Nf6 7. d3??

This was my biggest mistake of the game and I could not recover from here. If I had finally decided to trade bishops at this point it would have remained a somewhat even game.

The computer suggested that taking my bishop at this point was best, as the exchange would win the knight by 7. … Bxa4 8. Nxa4 Qa5+ 9. c3 Qxa4. However, The Advocate gave me a fighting chance.

7. … d4?? Here again, I could have made it a close game by exchanging bishops before moving my knight. But, no.  8. Nd1, which I wish I could say was my worst blunder of the game. The moves 8. … Qa5+ 9. c3 Qxa4 quickly won my bishop. After a few pawn exchanges 10. cxd4 cxd4 I quickly made the second worst blunder of the evening with 11. Ne5. I guess I didn’t learn from my previous mistake. Much better would have been 11. b3.

The Advocate took advantage to win my knight. 11. … Qa5+. Instead of 12. Bc2, I played 12. Qc2 which could have been really bad if answered with 12. … Bb4! I would have lost much more material that way. The knight was too tempting, 12. … Qxe5 13. 0-0 Bd6 14. g3 Nc6 15. f4 Qb5 16. e5 Bc7. With white to move, isn’t it obvious? I should take the knight, right? Maybe I recorded the game incorrectly?

We were not really down in time. I had much less time than The Advocate, but I still had time. Why I missed this for several moves is beyond me. Instead, I was more interested in getting my knight activated by giving it a spot to go to with 17. b3 Ne7 18. a4 Qd5 19. Nb2 Ba5? Eventually The Advocate would see that Bc6 was the better move. I had to defend my queen, but should have went for the trade at this point with Qg2. Instead, 20. Qf2 Bb4? 21. Nc4 Bc6. Focusing on the f3 and g2 squares as possible attacks for black, I did not see the mate square for the queen on h1. Thus, the game ended with the biggest blunder of them all… not seeing the mate. 22. Nxa5 Qh1#. While I may not have learned a great deal while playing that game, I certainly learned some things writing and analyzing it.

Third Game

My third game was against Double Oh Seven (007). He is a great player when time isn’t an issue. I won on time, and am anxious to analyze the game to see who may have had the advantage. I played black while 007 played white.

1. e4 e5 2. Nc3 Nf6 3. Nf3 d5? This game is referred to as a Three Knights variation of the Russian game. Apparently, 3. … d5 was not a good move as this gave white a +1 advantage using Stockfish 10+ on the lichess analysis board. My better move was 3. … Nc6.

4. d4 dxe4 5. Nxe5 Nc6 6. Nxc6 bxc6 7. Bc4 c5 8. 0-0 Qxd4?  I made a series of bad moves which kept giving white an advantage, but 007 continued to answer back with mediocre moves of his own that kept driving it back toward even. Using the queen to take the d4 pawn was a mistake. A good answer to this would be 9. Nb5 as it attacks my queen, and if I were to decide on taking the bishop, white wins the c7 pawn and my rook after the check. Stockfish gives white a +2 advantage at this point. However, the next series of moves it to black advantage with -1.

9. Qxd4 cxd4 10. Bb5+ Bd7 11. Bxd7 For some reason, I neglected to see the protection the knight was offering my e4 pawn with my bigger desire to maintain the ability to castle. 11. … Nxd7 quickly evened up the game.

12. Nxe4 Be7 13. Bf4 Rc8 14. Re1 0-0 15. c4 Re8 After this series of moves, 007 and myself are even-Steven. Instead of getting his h file rook into the game, 007 wastes a few moves with his e file rook.

16. Red1 c5 17. b4 Bh4?? A huge blunder on my part! The better move is to simply take the b4 pawn with 17. …cxb4.  While it allows the rook to gain ground with 18. Rxd4, the game remains somewhat even. Now that I’m in this bad situation, the best move is for 007 to fork my rooks with 18. Nd6.

18. Re1?? He didn’t see it in time! This was lucky on my part, which gave me a huge out. If he goes forward with the fork now, I can simply exchage rooks after a check and readjust my own rook. Right now, I should just take his b4 pawn!!

18. … Rc6?? This was a much bigger blunder than I had realized. I was not thinking at all of my back rank weakness with this move. If 007 answers with 19. Nxc5, he is attacking my knight as the back rank check mate must be addressed! After 19. … Rxe1 20 Rxe1 saving my knight would be futile. He gave me a fighting chance by missing this move.

19. Nd6 Re6 20. Rxe6 fxe6 21. Re1?? This was a game changing blunder on 007’s part. By applying pressure on my rook with 21. b5, he would have maintained the advantage. This move took the advantage from +1.2 to -2.4 by Stockfish’s standards!

21. … e5 22. Bxe5 Nxe5 23. Rxe5 Rxd6 24. bxc5 Rc6 007 didn’t calculate that exchange correctly and vocally said so. Time was beginning to be an issue as he was down around 2 minutes left. I was sitting comfortably around 5 minutes. White’s next best move is not obvious. Stockfish suggested that it should be 25. g4.

25. Re4?? This should have been answered with 25. … Bf6 to offer the pawn protection while providing protection for my bishop. I miscalculated however.

25. … d3 26. Rxh4 I did not calculate this correctly, or if I had, I lost it. I neglected to see I couldn’t protect my d3 pawn as the d6 square was guarded. However, I also did not see 007’s back rank weakness and how the the move 26. … Re6 would eventually win his rook (or checkmate him if he didn’t see it) by 27. f3 d2 28. Rd4 Re1+ 29. Kf2 d1=Q 30. Rxd1 Rxd1. I didn’t think enough on my next move because I wanted to keep the time advantage. My next move 26. … Rxc5 eliminated all advantage, and in fact, gave white the advantage.

27. Rd4 Kf7 28. Rxd3 Rxc4 29. f4 Rxf4 30. g3 Ra4 31. Rd7+ Kf6 32. Rd6+ Ke5 33. Rd7 d5 34. Rxg7 Rxa2 35. Rxh7 Ra1+ 36. Kg2 a4 37. g4 a3 38. g5 Kf5 39. h4 Kg6 40. Ra7 Kh5 41. Kh3 a2 42. Rh7+ Kg6 43. Rd7 Rh1+ and white (007) was out of time.