Calculation

The Eastern Washington Open was my worst tournament performance in the past several years. I lost one game to a much lower player because I blundered, overlooking simple checkmate threats. Earlier in the game, I twice rejected moves that were probably best because I perceived dangers that I neglected to calculate. Effort assessing the perceived dangers might have revealed that they were not credible threats. That loss gave my opponent the biggest upset prize for the event. My other loss stemmed from lack of opening preparation. Although I had White, my position was technically lost by move ten. Even so, it was my longest game.

Despite my poor performance, I was pleased with my play in round four. My longest think of the game was thirteen minutes considering my opponent's responses to the only move I examined seriously, and which I played. It was the best move in the position. Most gratifying was that my calculation of the critical line consistently found good moves for both sides. About half of these were best or only moves, and where I did not look at the best move, the best was only slightly better than ones in my line. My assessment that I would stand better in the resulting position eight moves deep also was accurate.

My deep calculation begins with this position.

Black to move

16...e5! is best. It makes sense positionally, but I had to verify that it was correct tactically.

I had spent some time preparing this move. This moment appeared the correct time to make it. If not now, White might play Ne5, which would then necessitate more preparation.

Calculation was simple and straight-forward if White replied 17.Nxe5 or if he replied 17.dxe5.

However, the zwischenzug 17.Bb5 creates complications. It was this line that I analyzed for thirteen minutes.

I briefly considered a) 17...Bd7, and spent a few minutes looking at b) 17...e4. I considered the latter as an unattractive, but safe option. Here, my analysis concluded in a position where I thought that I was better, but Stockfish evaluates as equal. However, I missed the best move half-way through this line.

The line that took the longest to calculate and assess begins with c) 17...exd4! It is the best move.

a) 17...Bd7 is Stockfish 4's second choice. I gave it only cursory examination, but realized that some of the ideas in the lines I did examine would be present.

b) 17...e4 is Stockfish 4's third choice.
I looked at 18.Ne5 Bxe5
     Unexamined was the correct 18...Ne5! 19.Bxe8 Nd3-+.
My analysis continued 19.dxe5 Rxe5

White to move

I assessed this position as better for Black. Black is two pawns ahead, but Stockfish sees the position as equal. Black can generate threats against h2, but White has sufficient defensive resources. Black's pawn center is fragile, and thus his material advantage is temporary.

My analysis did not extend to 20.c4, which is White's only move. Otherwise, Black has secured an advantage with subpar moves.

c) 17...exd4 occupied the longest portion of my thirteen minute think. It is the best move.

I did not look at 18.Qd1, 18.Qf1, nor 18.Qd3. 18.Qxe8 struck me as the critical line, as it is a direct effort to refute my idea. Black's back rank has weaknesses. These are not fatal, but White has several forcing moves. Moreover, there are some material imbalances to consider.

18...Nxe8 19.Rxe8+ Bf8 and now more branching occurs.

Examining branching in one of several possible lines four moves from the current position has not been the norm in my over the board play. It must become so. In correspondence chess, such analysis is easier to perform thanks to such tools as analysis boards, database software for recording and observing possible positions, and the old fashioned paper and pencil with a physical board. Doing this analysis under tournament conditions with the clock running while looking at a static board requires imagination. I closed my eyes several times in order to imagine the board.

c1) Bxc6 Qxc6 21.Ba3 (the computer prefers 1.Ne5, which I do not recall considering) 21...Bb7 22.Rxa8 Bxa8 23.Nxd4 Qxc3

White to move

Black's advantage is clear, but I thought that White had a more dangerous line.

c2) 20.cxd4 seemed less dangerous, and I did not go deeper.

c3) 20.Nxd4 seemed critical. Stockfish vacillates between this move and 20.cxd4 as best for White.

20...Bb7 (I do not recall looking at 20...Nxd4, which the engine considers best) 21.Rxa8 Bxa8 22.Nxc6 (Stockfish prefers 22.N2f3) 22...Bxc6 23.Bxc6 (23.Bd3 may be better) 23...Qxc6 24.Rxa7.

Black to move

Although I was not certain my advantage was overwhelming in this position, I thought that I would probably choose this line. I knew that I could spend more time examining it after playing 16...e5.

Our City Champion was watching our postgame analysis, and chimed in that he thought he might like White's position. Black could be in danger if White's bishop and rook could both be brought to bear against f8. We discussed these dangers briefly. White's problem is that Black's queen controls the rook's access to the back rank.

White has three pieces to Black's two, but the queen is a powerful piece and White's queenside pawns are vulnerable once Black shores up the dangers on f7.

Stockfish concurs that Black is winning.

After 16...e5!, my opponent replied 17.dxe5. His move is simplest and best, although it saved me from the dangerous and complicated line. I was relieved. The rest of the game presented one difficulty when my opponent found some tactical tricks to expose my king. Initially, I marched forward, only to see that he threatened a draw by repetition, so my king went back and I had to retreat a knight that I wanted to use in the final attack.

My opponent then missed a tactic that gave me checkmate in two.