An experiment in imbalances

Ordinarily, when I play chess, I favor relatively balanced positions, in which the result of the game is more likely to hinge on positional ideas, rather than tactics. Hence, I do not sacrifice material often, unless I have calculated that doing so is guaranteed to work. However, this past weekend, I played a game in which I was inspired to play more freely, sacrificing material for evident compensation, but no clear win.


Simone Liao (2202) vs. Simon Rubinstein-Salzedo (2026)

My opponent is yet another of the endless string of very strong kids to play chess in the Bay Area. It’s actually a rather unusual occurrence to play against someone who has graduated from high school around here, even in the master section of a tournament. This particular kid has the distinction of being the top woman under age 21 in the country, despite being only 14!

1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 c5 5. a3 Bxc3+ 6. bxc3 Ne7 7. Qg4 Qc7 8. Qxg7 Rg8 9. Qxh7 cxd4 10. Ne2 Nbc6 11. f4 dxc3 12. Qd3

Up to here, everything is completely standard, except that perhaps 11…Bd7 is slightly more common than 11…dxc3. With 12…Bd7, we would transpose back into the most common line, but I wanted to play something different.


I was still well within my book knowledge at this point, but my opponent started to think for the first time after this move. Before this, we had each only taken a few seconds to play all of our moves. The most testing line is 13. Nxd4, but she opted for what I consider to be a less dangerous variation.

13. Ng3

This is still a known move, but I forgot what the textbook response is. Fortunately, it isn’t very hard to find: white wants to put her knight on either d6 or f6 after Ne4 (or Nh5), so I have to make sure that that isn’t particularly dangerous.

13…Bd7 14. Ne4 O-O-O

White can win a pawn with 15. Nd6+ Kb8 16. Rb1 b6 17. Nxf7, but after 17…Rdf8 18. Nd6 Nf5, it may well be black who benefits more from the loss of the pawn, due to the files available for the rooks. Thus, it makes sense to play something else. In my understanding of this opening, white suffers a bit by having an exposed king and somewhat awkward piece play, but the compensation is a passed h-pawn that can become dangerous quickly if black is slow to react. Therefore, my opponent’s next move seems very reasonable.

15. h4

Okay, so white’s plan is clear. As for black’s plan, well, nothing makes a French player happier than acquiring mobility for the central pawns. One way of obtaining that is by playing f6 and trying to trade them off. However, I favored a more dramatic way of accomplishing this task: blowing the center open with a piece sacrifice. I didn’t want to play 15…Nxe5 immediately though, because I was concerned about 16. Qxd4. It turns out that black is doing fine, but I preferred to defend the d4-pawn first and then see what would happen.

15…Nf5 16. Rh3



Definitely not the way I usually play chess, but it felt like such a logical thing to do. Her king is stuck in the center, so I have an attack that isn’t going away any time soon, and there isn’t any clear way for white to clarify the position. I believe that I get full compensation for the knight, but no more. (I’m certainly not winning, which is reasonable, because white hasn’t done anything wrong.)

The main line I calculated before playing 16…Nxe5 was 17. fxe5 Qxe5 18. Qe2 Bc6 19. Nf2, after which I get my material back. But of course 19. Nf2 isn’t the strongest line.

17. fxe5 Qxe5 18. Qe2 Bc6 19. Ng5!

Definitely the right move. Before I played 18…Bc6, I realized that there were plenty of reasonable options other than Nf2, and I had to work out that in particular 19. Nxc3? loses (to 19…Qa5!). I also had to work out what happened after Ng5, calculating the line played through move 23.

19…Qxe2+ 20. Bxe2

20. Kxe2 f6 and now 21. Nxe6? is impossible due to 21…Rde8, winning the knight, with a big edge for black.

20…f6 21. Nxe6 Rde8 22. Nf4


And now I have two ways of recovering some material: 22…Bxg2 and 22…Ng3. I chose wrong, unfortunately.

22…Ng3 23. Rxg3

Forced. Black threatens 23…Nxe2 24. Nxe2 Rxg2. And after 23. Rh2, 23…Bb5 wins (although not my planned 23…Rxe2+? 24. Nxe2 Re8 25. Kf2 Nxe2 26. h5, and white is better).

23…Rxg3 24. Kf2 Reg8 25. Bf1

Now is a good time to take stock of the position. I felt that I was the one playing for a win here, although it’s not necessarily clear what to do. My logic, though, is that white’s pieces are rather awkwardly placed. She has to keep a bunch of pieces defending g2, and in the meantime it’s not so easy to get the queenside pieces into the game. However, it’s not so easy for me to improve my position and increase pressure. So, I decided to eliminate Ne6 possibilities.

25…Kd7? 26. h5 Be4 27. Ra2?

This seems logical enough, and it’s what I expected, but actually 27. Rb1! is much stronger, and white is doing very well. Because of this possibility, it was better to switch my 25th and 26th moves, as 25…Be4 more or less forces 26. Ra2. (The computer initially considers 26. h5 a viable option for tactical reasons, but at deeper levels of analysis, it turns out not to work very well.)

27…R3g4 28. Bd3 f5

28…Bxd3 is actually better, since white quickly gets in zugzwang: 29. cxd3 Rh4 30. Rc2 Re8, and there is no way to save the h5-pawn. However, white can still draw with 31. Ne2 Rxh5 32. Nxd4 Rd5 33. Be3 Rde5 34. Bc1 Rd5, repeating moves. During the game, though, I thought I would do well to encourage a bishop trade if doing so would allow me to strengthen my central pawn mass. After 29. Bxe4 fxe4, the central pawns are rather dangerous.

29. Bf1 Kd6 30. Ng6



I felt that playing this move was the height of frivolity, as there is really no need to sacrifice the exchange here. However, I couldn’t imagine that I might be worse after doing so, and the sacrifice eliminates one of white’s main sources of play. (Note how most of the white pieces are rather tied down.)

31. hxg6 Rxg6 32. Bf4+ Kc5 33. a4

This was a move I was hoping to be able to prevent, but then there it was. I was about to play 33…a5, but then I noticed something interesting.


With the threat of 34…Rb2. There is only one way to defend.

34. Bc1 Rb1 35. Ba3+ Kb6

And now I threaten 36…Bd5. Again there is only one defense (although she can throw in 36. a5+ first).

36. Bc4 Rd1

And now white loses at least a pawn.

37. a5+ Kc6

Not 37…Kxa5?? 38. Bc5#, but typically players at this level don’t do things quite that stupid.

38. Bb3 f4

The g-pawn will still be available later, so I wanted to prevent her from getting her king activated.

39. Be7 Rd2+ 40. Kf1?

So far, she has defended fairly well, and after 40. Ke1 Rxg2 41. Bf6, the game is close to equal. One possible line is 41…Bd5 42. Ra4 Bxb3 43. cxb3 c2 44. Rc4+ Kd5 45. Bxd4 c1Q+ 46. Rxc1 Kxd4, which white can hold easily. But after 40. Kf1, black is winning.

40…Bxg2+ 41. Kg1 Bd5 42. Ba4+ Kc7 43. Ra1 Rg2+ 44. Kf1 f3


Threatening 45…Bc4+, so white has to abandon defense of the c-pawn.

45. Bb5 Rxc2 46. Bf6 Rd2?!

The zwischenzug 46…a6! wins immediately: 47. Be5+ Kd8! (not Kc8, after which white gets an extra check on a light square and can therefore put up stiffer resistance) 48. Bd3 Rd2 49. Bf6+ Kc7, and the light squared bishop cannot move off the diagonal and must therefore be sacrificed.

After 46…Rd2, black is still winning, but I started losing the thread, as the bishops became increasingly active.

47. Bg5 Rb2 48. Bd3 Kc6 49. Bc1 Rh2 50. Kg1 Rg2+ 51. Kf1 Kc7?

I wanted to bring my bishop around to h3 without having something nasty happen on e4, but this is the wrong way to play. Instead, 51…f2! with the idea of running white out of moves wins comfortably. For example, 52. Rb1 c2 53. Ra1 (53. Rb2 Rg1+ 54. Kxf2 Rxc1 55. Bxc2 Ra1, and winning with three extra pawns is trivial) Rh2 54. Be2 Be6 55. Bf3+ Kc5 56. Bg2 Bd5 57. Ba3+ Kc4 58. Bxd5+ Kxd5, and the pawns win.

After this, there are still some chances for black, but we were reaching the time trouble stage after nearly five hours of play, and so there wasn’t sufficient time to make sure that our moves made much sense.

52. Rb1 a6 53. Bf4+ Kd7 54. Rb6 Bc6 55. Rb4 Ra2 56. Rxd4+ Ke7 57. Bd6+ Kf6 58. Rf4+ Kg5 59. Rf5+ Kg4 60. Rf4+ Kg5 61. Rf5+ Kg4 1/2-1/2

Neither one of us knew during the game whether this position was equal, or if not, who was better, but with 2 minutes each, we were both happy not to do anything really stupid and split the point.

I found it quite freeing to sacrifice material in this game. It has been several years at least since I last sacrificed material for any reason other than a forcing line or out of desperation in a bad position, but it might be worth trying it out again in the future.

I’d say that the opening line I played in this game was a great success, because my opponent (with the white pieces) did nothing wrong in the opening, but I was still able to sacrifice a piece and get full compensation for it. From a philosophical point of view, it’s hard for me to understand how that is possible, but perhaps the philosophy is irrelevant, and I simply ought to enjoy the games I can get out of it.


About Simon

Hi. I'm Simon Rubinstein-Salzedo. I'm a mathematics postdoc at Dartmouth College. I'm also a musician; I play piano and cello, and I also sometimes compose music and study musicology. I also like to play chess and write calligraphy. This blog is a catalogue of some of my thoughts. I write them down so that I understand them better. But sometimes other people find them interesting as well, so I happily share them with my small corner of the world.
This entry was posted in chess tournament. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s