Chris Ross v Andrew Greet
Bedfordshire v Cornwall, 2014/15 (Bd.1)
A welcome return to Cornwall colours for IM Andrew Greet.
White avoids the main line of 10.Nxd4 in favour of an early queen exchange. This isn't considered especially challenging for Black, but it's not a bad option, and leads to a balanced position where both sides can play for a win.
I spent nearly twenty minutes deciding on this move and anticipating the likely follow-up.
I was expecting 14.a4 a6 15.Bf3 0-0-0 16.a5 Ba7 17.Re1 Bc6 18.Bxc6 bxc6 when the position seems about equal; Black's queenside pawns have been split, but the king is well placed to defend them while Black's pieces are nicely centralized.
This leads to a similar type of situation as the previous note. However, with the a-pawns on their original squares, my queenside is more stable.
The position is close to equal, but Black has some small advantages. The extra central pawn is a useful asset, enabling Black to gain space in the future. Black is well placed to contest the d-file, and can consider meeting a future Rad1 with ...Rd5, when a rook exchange may allow Black to transform the pawn structure favourably. And finally, Black's king can advance more easily. White should be able to hold, but his position is slightly unpleasant and there is no obvious way to simplify to a clear draw.
When playing for a win in such positions, it's important not to put pressure on yourself to prove anything, but rather to keep playing good moves and improving the position. Eventually, if you make enough small improvements, they will add up to something big.
Otherwise ...Nb4 could be annoying.
A good decision. 19...Rd6? 20.Nd2! is rather awkward, as the knight is heading for c4 and e5 with tempo. Black is actually slightly worse here.; I had been planning 19...Rd7 , but then I noticed that White could improve his knight with 20.Nd4! Kb7 21.Nf3 , and once again Ne5 will be annoying. I therefore elected to improve my king while defending c6, ensuring that a future Nd4 will not come with tempo.
The tension along the d-file is slightly awkward for White, as he cannot really move either rook, whereas Black always has the ...Rd5 possibility.
The knight was not well placed on b3. If White does not improve it, then Black may consider ...c5 and ...Nc6.
Diagram Now I have control over the d-file. There's no way to invade, but it will take White a few tempos to neutralize the active rook, which will give me time to improve the position in some other way.
I considered allowing him to take on d5, when ...cxd5 would give me a strong central majority that could be supported by my king. However, I thought that the further simplification would increase the likelihood of a draw. Moreover, White has the potentially troublesome possibility of creating an outside passed pawn on the a-file. By the way, I decided at this point that 25...Rd8 was the move I wanted to play, but took the opportunity to repeat the position as I was down on the clock.
No draw thank you!
Here's another important point regarding playing for a win from an almost equal position. Piece exchanges are fine up to a point, especially if they improve your pawn structure, as was the case with the bishop exchange on b6 for example. However, any further exchanges from this point in the game would lead to a pure knight or pure rook endgame; when you only have a single piece left, then your active possibilities become much more limited. However, when you have both a rook and a knight, there are many more ways to put the opponent under pressure, by using the two pieces together to attack a weakness. Additionally, in this particular position, my main plan for the next few moves is to start advancing my pawn majority and perhaps even attack my opponent's king! Clearly, in such a situation, two pieces working together are worth more than the sum of their individual parts.
Preparing to contest the d-file, but blocking the e-file, thus facilitating my next move. If 29.Ng5 Nd6 30.Nxh7 Rh8 31.Ng5 Rxh2 Black maintains a slight initiative, especially considering that 32.Nh3 (intending to force a rook exchange) runs into 32...Rh1+ 33.Ke2 Rxe1+ 34.Kxe1 Nc4 and Black wins a pawn.
Diagram Once again, no more exchanges thank you!
Here's another simple point regarding piece simplifications. To simplify from three pieces to two (not including the king) is not necessarily a big deal. However, to simplify from two pieces to one reduces your active possibilities more drastically, as you lose all potential for the two pieces to cooperate in attacking a weakness. Obviously there are certain situations where it makes sense to do it, but just be aware of the significance of swapping down to just a single piece plus your king.
Quite a committal move, but I wanted to make my opponent worry about ...e4 followed by ...Nd4+. 32...h5 is a good alternative, gaining space without fixing the queenside structure.
White decides to gain some space and force me to make a decision.
Sending the knight to a5 (with tempo) followed by c6 and later d4 seems obvious, but it is inaccurate. 34...h5!? is one good option.; Alternatively, if Black wants to improve the knight, then 34...Kc6 followed by ...Nf7-d8-e6 avoids the plan mentioned in the next note.
35.Nh4! would have enabled White to equalize: 35...Na5 (35...g6 36.Ng2 Na5 37.Ne3 followed by Nd5+ is fine for White) 36.Nf5 Nxb3 37.Nxg7 Rg8 38.Ne6+ Kc6 39.Nd8+ Kc7 40.Ne6+ with perpetual check is one possible finish.
The time control has been reached, and White is under pressure. Black's knight is ready to jump to d4 or b4, and his flexible pawn majority is clearly superior to White's static queenside constellation.
At the time I thought this was forced, otherwise the rook would be trapped after ...Nb4. However, it is too passive. [White needed to resort to active defence with 37.Ne4! Nd4+ 38.Kd3 Nxb3 39.Nd6 followed by Nb5+, when his active counterplay may well be enough to draw.
38...g6 is also good, but opening a file for the rook seems best.
The pawn endgame is winning for Black, but a good deal of accuracy is required in some lines. I would be lying if I said I calculated all of them before entering this endgame; I needed to keep some spare time on the clock, so I calculated as much as I could and followed my instinct, which told me that I would at least have good winning chances.
This fails to offer any resistance at all. White could have made my task harder with either of the following options:
b) 46.Kc3 Kb7 47.b4 cxb4+! (I must confess that during the game I was intending 47...Ka6?, when 48.g5! reaches the same mutual zugzwang as above, with Black to play and draw. I'd like to think I may have spotted this over the board, but we'll never know.) 48.Kxb4 Kc6 Black is winning. The main line continues: 49.Kb3 Kc5 50.Kc3 fxg4 51.fxg4 e4 52.g5 e3 53.Kd3 e2 54.Kxe2 Kxc4 55.Ke3 Kb4 56.Ke4 Kxa4 57.Ke5 b5 58.Kf6 b4 59.Kxg6 b3 60.Kf7 b2 61.g6 b1Q 62.g7 Qf5+ Black has a textbook winning endgame.
It's all over, as White can do nothing to stop the king from invading.
Notes by Andrew Greet