calculation – Alex Ostrovskiy

No new tournaments again so I will take another trip down memory lane. Actually, that is not exactly true as I played a Blitz tournament last weekend. The invitation only event, known as “Gelfand Cup” boasted an all master field. The 10 player round-robin event was won by GM Alex Lenderman with an 8.5/9 score. Second place was taken by GM Leonid Yudasin with 7.5/9, while I had a relatively distant 6/9, good enough for a tie for 3rd. I actually had a pretty good start at 4/5, beating Yudasin and IM Yury Lapshun, but a 2/4 score in the 2nd half wasn’t impressive. Nevertheless, blitz is blitz and blitz is fun….plus due to the generous funding I was able to snag a cash prize nonetheless.

Anyway, I wanted to share an excerpt from one of my older games.

The above position arose during the 8th round of the 2013 edition of the New York International. My opponent was the famous IM Jay Bonin. Without calculating any variations, the position above appears to be definitely better for Black. With a powerful Knight on e4, two bishops, and chances on the Kingside, Black is poised for a positive result. Furthermore, although White’s pieces are developed on decent squares, it seems difficult for him to come up with a plan. Nevertheless….

1. d5?!/!?

This move looks like the perfect move to make sense of White’s pieces. So why the assortment of exclams and question marks? When I first saw d5 played on the board, I had a moment of confusion as I thought I could simply reply with 1…c5(?!). Left unperturbed I would simply continue with Bd6 (and maybe b5). Should that happen, Black will achieve such a harmonious position that it would only occur in the most pleasant of chessplayers’ dreams. White is really struggling to create any coherent plan. Meanwhile, Black’s pieces are positioned perfectly and I can slowly prepare a strike on the Kingside. Besides obvious Qe5 ideas, I can also try some h4-g5-g4 plans. Most importantly, White is reduced to a concerned bystander. Unfortunately, 1..c5 doesn’t work! After 1. d5 c5 White has 2. d6! with ideas of Nd5 as well as Nxe4. The subsequent opening of the position will be to White’s advantage as his pieces are fully developed. For example, 2…Bxd6 3. Bxe4 fxe 4. Nxe4 (Diagram) and the difference between White’s piece activity (from the 1st diagram) is night and day.

So there it is. 1. d5 is a great positional move that aims to exploit White’s only advantage in the position (quantity of development). As any strong positional move, it is backed up by a cute tactical point. But wait! Well, at least that’s what I said to myself after spotting the pitfall that is named 1…c5. I should mention that for about the past 5 moves, I had continuously been looking to land a knockout blow on the Kingside. After all, all my pieces are taking aim at White’s king… and White’s king doesn’t even have any defenders around! So I thought and I thought; I think I may have had 30 minutes after 1. d5 but ended up only with 5 when I played my move. Fortunately, my suspicious investment ended up paying off. There is a tactical operation in this position! Actually, it takes the default engine in my ChessBase (the ancient Fritz 11) a whooping 21 seconds to give Black a favorable evaluation.

1… Ng5

2. Qe2 Qd6!

Here we go. The main points are unfolding

3. f4

Now it was my opponent’s turn to be confused. The threats have ended and Black has to deal with the hanging Knight and the repercussions of poor development after dxc6.

3… Bb6!

This is the main point. White has been forced to compromise the e3-square and now faces a dilemma. However, Black’s success is not yet obvious. Jay makes me prove my variation with the principled:

4. fxg5?

Practically, this move makes sense as it is the principled way to make Black explain the neglect of the Ng5. Objectively, this move loses.

4. Kh1! was the only way to stay in the game. Actually this powerful move puts a damper on Black’s plans. After 4…Rxe3 5. Qf1 (keeping f4 under control) Nh7, Black has an extra pawn, two bishops in an open position, and White’s king is exposed. However, Black is unlikely to be able to make use of any of these advantages if I am not able to develop properly and negate the activity of White’s pieces along the open files. The computer remains optimistic about Black’s chances, but the next 5 moves may look difficult for Black to make. Saying that, upon unraveling Black can look forward to a great game.

4… Rxe3

5. Qf2

The Queen shields the King from any nasty discovered check(mate)s. It looks like White is pretty safe right now. The immediate 5…Re1+ gives up too much material. 6. Rxe1 Bxf2+ 7. Kf2 and the Rook+2N more than compensate for the Queen deficit. Fortunately, I do not have to cash in on my pin just yet.

5… cxd5!

I was proud of this move, especially due to the idea it’s connected to and the fact that I saw it well in advance. Nevertheless even my brain-dead Fritz11 offers the alternative win 5…Qb4. But it does admit that 5…cd is stronger! Now since the Bb6 is defended, I am threatening Rxd3.

6. Na4

The final line of defense. White threatens Nxb6, but also Nc5 to plug the a7-g1 diagonal. Again, this resource had to be spotted and refuted in advance.

6… Re1+

This wasn’t good before, but now things have changed!

7. Rxe1 Bxf2+

8. Kxf2 Qf4+

Loose pieces drop off.

9. Kg1 Qxa4

and soon it was 0-1

Position After 9... Qxa4 — Position After 9… Qxa4

Hello again! With the resumption of my college semester I have began a hiatus from tournament play, although I hope this hiatus will be short lived. Instead, the next few blog posts (or at least the plan is when I have no personal recent tournaments to cover) will be devoted to some interesting games I have played in the past.

Which brings us to the a paraphrased quote from the title:

“How many moves ahead can you see?”

A funnily popular question, especially by casual enthusiasts to chess players. The question has produced equally entertaining answers, including the famous quip “Just one, but it’s always the best!” Nevertheless, this question is impossible to answer. There are endless parameters surrounding the task of calculation. I suppose a King+pawn vs. King endgame could be calculated 30 moves deep until Checkmate, but this doesn’t feel like an honest answer. I always thought that asking about calculation depth was comparable to someone asking a basketball player how many times he can do a crossover.

Nevertheless, I digress.

After all my rambling, I’m going to segue to an endgame that I played some years ago. I think this game showcases the furthest I’ve ever calculated, although once again the position is simplified and is surprisingly forcing. It’s not the most impressive bout, but at least the endgame has instructional value.

Here we have a double rook endgame that is visibly favorable for Black. White is reduced to passivity to due to the disparity between the activity of the Rooks. Aside from pressuring the f2 pawn, Black might also target the d4 or a2 pawn after Rf3 and a further rook swing. Nonetheless, despite the obvious advantages that Black possesses, it is not immediately clear how to break through. On top of this, Black needs to consider what White is going to do. How can White improve his position? First of all, White is prepared to defend all his weak pawns as the Rooks cover the 2nd rank pawns and can easily go to d2 and defend d4. Second of all, White is not looking to remain passive. With the plan of shooting the King over to e1, White will be ready to move his Rooks around to more noble goals, such as the B or A file. On the flip side, White has no other plan other bringing the King to e1, giving Black several moves to organize some kind of offensive. As mentioned, Black is doing a good job attacking the backwards f-pawn, but presently it is just not enough. Well, as the dictum goes, “You must create 2 weaknesses.” Armed with this age old advice, I played

33… g5!

Pinpointing a surprise target in White’s camp: the h3 pawn! What could be Black’s plan? In the following moves, all is revealed. White will bring his King to e1 and Black will work on pressuring the h3-pawn. Although it may seem like a stretch, I had already calculated the finish of the game from this move onward.

34. Kc1 h5!

35. Kd1 h4

36. gxh4 gxh4

37. Ke1 Rf3

Position after 37...Rf3 — Position After 37…Rf3

Both sides have achieved their plans. Black has succeeded in exposing the h3 pawn, and White has brought his King to e1 and is able to activate his rooks. Here, White obviously banks on the rooks’ newfound mobility as otherwise he will just lose a pawn.

38. Rb3 Rxb3

39. axb3 Rf3

40. Re3

White has avoided material losses and has taken off a lot of pressure from his position, but…

40… Rxe3!

It turns out that the pawn endgame is problem!

41. fxe3

41… Kb7

42. Kd2 Ka6

43. Kc3 Kb5

Position After 43...Kb5 — Position After 43…Kb5

Now this pawn endgame is converted with the help of some good old fashioned flanking with a sprinkle of waiting moves.

44. Kc2 Kb4

45. Kb2 a6!

46. Kc2 Ka3

47. Kc3 a5!

48. Kc2 Ka2

49. Kc3 Kb1

50. b4 a4

and White resigned 0-1

Naturally, the game could have continued with: 51. b5 a3 52. b6 a2 53. b7 a1=Q+, and Black gets an easy win because of one tempo.

So there we have it, from 33…g5 to 51. Resigns, Black had the whole thing planned. Without shame, this could be extended to 53…a1=Q+ (although we will refrain to further adding moves 54. Kd2 Qb2+ and such…) so we have an example of a 20 move deep calculation. Can I now answer the title question with “20 moves without fail”…? Likely not, but nonetheless this endgame brings back pleasant memories even to this day.

Until next time!

Alex Ostrovskiy

Tag: calculation

Tactics Time

How far can you calculate?