Decision time

Every “weekend warrior” knows that the faster pace of modern chess is affecting the quality of play. In a recent team match we had 90 minutes for all our moves, which is an especially insidious time control. Game/30 is obviously a form of speed chess, and everyone knows to adjust their decision-making accordingly. But Game/90 is quite a different matter. It plays like classical tournament chess for quite awhile, but of course it too inevitably deteriorates into a time scramble without any safe harbour on the horizon. When this turning point is reached the position may still be quite complicated, or it may be relatively simplified (which is not to say simple.) Either way the effects are unpredictable, and that, my friend, is one of the charming ideas behind “sudden death” time controls. Unless someone goes wrong, neither player can win.

In my game from this match I reached the diagrammed position as White and with an important decision to make. In the preceding complications I had managed to win the exchange (or as my opponent said afterwards, “I had to sacrifice the exchange.”) This has come at a price: Black has an unopposed dark-squared bishop, active pieces and an extra pawn. Nevertheless I believed then (and still believe) that White must be better here.

How to proceed? The answer isn't obvious. My thoughts ran along a narrow track for awhile, and when I eventually found the right idea I didn't execute it properly. I blame the clock.

When you have won the exchange for a pawn, winning back the pawn is often a good idea because then you will simply be up the exchange. As everyone has known since the days of Reuben Fine, when you are up the exchange you can sacrifice it back to win a pawn and create good winning chances. With this philosophy in mind my eye naturally fell on the weak pawn at e6.

There are two ways to get at this pawn. One is to play 1.Ng5 and threaten to take it, while the other is to play 1.Ne5 and threaten to take the bishop that is defending it. The first option has the added virtue of putting pressure on Black's king position and creating possibilities such as Nxh7 followed by Qh5+ and Bxg6. The second option has the added virtue of restricting Black's unopposed bishop. Maybe there are sacrificial possibilities on g6 as well, but they're not as incisive.

1.Ng5 is obviously more complicated, so I gave it some priority. Black can react in several ways:
a) 1...e5. This I did not like for Black because it weakens the a2-g8 diagonal. White can take advantage with 2.Rd1! threatening the forced manoeuvre Rxd6, Qc4+ and Nf7+ which trades off two sets of pieces and reduces the bishop on g7 to pawn status. Black is curiously short of remedies. Perhaps 2...Rf4 is necessary, but I still like White after 3.Nge4. (In the post-mortem my opponent told me he was intending to meet 1.Ng5 this way. When I showed him 2.Rd1! he had to agree he had dodged a bullet.)
b) 1...Re8. This protects the pawn but allows the sacrifice 2.Nxh7!? After 2...Kxh7 3.Qh5+ Kg8 4.Bxg6 Re7!? the game is still rather complicated, but in new ways. White has a rook and pawn against two minor pieces, kingside pressure, and the weak pawn on e6 to attack. But Black is not without his own chances. Verdict: unclear, but probably equal. And in this line, Black does not have to take the knight. For example, 2...Nf5!? is interesting.
c) 1...Bd4!? The reason I eventually rejected 1.Ng5. White has to play 2.Rf1 because 2.Nxe6? fails to 2...Rxf2! (not 2...Bxf2+ 3.Qxf2 Bxe6 4.Qg3 with advantage) 3.Nxd4 Rxe2 4.Ndxe2 and now 4...c4! brings all of Black's pieces to life. Despite having two rooks for the queen White is in a bad way since they are quite uncoordinated. But my assessment of 1.Ng5 Bd4 2.Rf1 was perhaps hasty and not accurate. White threatens 3.Nxe6 as before, and both 2...Qc8 and 2...e5 are refuted by 3.Nxh7! Black probably has to play 2...Rf6, but this allows 3.Nf3, exchanging off the bishop on d4 and improving White's chances somewhat. I blame the clock here too.

1.Ne5 is the quieter method. It didn't take me too long to decide that 1...Bc8 was the best reply. Now it is hard to see a good follow-up. Black is threatening the manoeuvre 2...Nf5-d4. Depending on how White reacts he will be threatening ...Nbc2 or ...Bxe5. Because 2...Bc8 has cleared the second rank, sacrifices on g6 are unlikely to succeed.

Time to look at the clock. My goodness! Already 30 minutes behind... if it goes on like this you'll really be in time trouble! Are there any other ideas for White? Well, in the previous analysis it transpired that the knight on d6 was a bit of a danger piece. When Black gets time for ...Nf5-d4, the effect on White's game is not pleasant. So let's see if we can trade him off. How about 1.Nb5? If 1...Bxb5 2.axb5 (stronger than the immediate 2.Qxe6+) 2...Qd7 3.Bd3! Nxd3 4.Qxd3 Bxb2 5.Rxa7 and things are going quite badly for Black. And after 1...Nxb5 2.axb5 it's mission accomplished!

Another quick look around... nothing else comes to mind... so it's 1.Nb5, press the clock, and let him start thinking! Unfortunately, the game continuation revealed that 1.Nb5 did not improve White's chances. The queens came off after 1...Nxb5 2.axb5 Qb6 3.Ra3 Bxb5 4.Qxe6+ and my pawn on b2 proved to be a nagging weakness. When my opponent eventually offered me a draw, I had 15 minutes to his 40 minutes on the clock and only a symbolic advantage on the board. But it was an enjoyable game!

Back at home I set up diagrammed position again. With no clock ticking I saw the right idea right away: 1.Ne5 Bc8 and only now 2.Nb5! After 2...Nxb5 3.axb5 b6 4.Ra3 Nd5 5.g3! White is clearly on top. How could I have missed that?!

I learned three things from this game:
1. When the opponent has a pawn and active pieces for the exchange, winning back the pawn might be a good idea, but exchanging or restricting his active pieces might be a better idea.
2. Keep a flexible outlook. An idea from one variation might be the key to another variation.
3. Try to maintain objectivity, even when behind on the clock.