Last season, long-standing readers may recall, I switched from playing the Sicilian to 1… e5 in reply to e4.
Just as last season, I’ve had the black pieces in most of my games so I’ve had several more opportunities to imitate my opponent’s e-pawn advance.
My first 1. e4 e5 game this season was against Alfie Onslow, a recent member of Richmond Junior Club who has outgrown the Saturday group and is now about my strength. I’d expected something like a Catalan or an English but discovered he’d switched to 1. e4.
Let’s look at the game.
1. e4 e5
2. Nf3 Nc6
3. Bc4 Nf6
4. d3 Bc5
This is the way most stronger players choose to handle the Italian these days. White avoids the theory and tactics of 4. Ng5 or 4. d4 as well as the boring 5. Nc3 so popular in kiddie chess, heading for a strategically rich middle game.
6. Nbd2 Bb6
7. Bb3 a6
This looks rather artificial. White’s planning to leave his king in the centre for the time being.
9. h3 h6
10. Nf1 Be6
11. Ng3 Qd7
12. Nh4 Ne7
Starting a king-side attack which we perhaps both over-estimated. This sort of thing looks tempting from the white side and scary from the black side. A stronger or more confident player than me wouldn’t have panicked, though.
14. Qxh5 Bxb3
15. axb3 Qe6
16. Nf5 Nxf5
17. exf5 Qf6
18. h4 g6
By this point I was getting worried about a potential g4 followed by g5 but, as usual, I was fearing phantoms. I can always meet g5 with Qxf5 when his g-pawn is pinned so I should just continue with a move like 18… Rfe8 or 18… d5. Instead I panicked and sought a tactical solution which only gave Alfie some genuine attacking chances.
19. Qxh6 Qxf5
Suddenly both players have king-side attacks. I guess it takes a certain amount of courage to ignore Black’s threat and press on regardless with h5, but perhaps that’s what Alfie should have done. We can look first at 20. h5 Qxf2+ 21. Kd1 when White’s king is safe and Black has to deal with the threats on the h-file. Her Majesty has to scuttle back with 21… Qf6 22. hxg6 Qg7 when White can win the exchange by trading queens followed by Bh6+ or, even stronger, continue the attack with 23. Qh3, with the idea of Ra4, which gives White a winning attack. So instead Black must play 20… Bxf2+ 21. Ke2 Bg3 (best) 22. Ra4 g5 (best) 23. Bxg5 f6 (best) 24. Be3 when Stockfish gives White a slight advantage (don’t ask me why).
Back in the real world, though, most of us would, as Alfie does, stop and defend f2. But now White’s position is not so easy to handle and I gradually outplay him over the next few moves. The computer, of course, suggests various improvements which need not detain us here.
20. Be3 Bxe3
21. Qxe3 Kg7
22. Rh3 Rh8
23. Ra4 d5
24. g4 Qf6
25. g5 Qe7
26. Qf3 c6
27. Kf1 Raf8
28. Qg4 f5
29. gxf6+ Qxf6
30. Qg3 Rh5
This should have been the losing move.
Or 32. h5 Rxf2+ 33. Kg1 Rf1+ 34. Kh2 Qf2+ 35. Qxf2 R8xf2+ 36. Kg3 Rf6
37. Rxg6+ Rxg6+ 38. hxg6 Kxg6 with a winning rook ending.
A desperate shot which, because I don’t stop to think, pays off. I’d assumed he had to play 33. Qg2 when I’d seen that 33… Rxd3 could be met by 34. h5, keeping White in the game, so had planned, correctly, to play Qf5 instead, which is indeed winning. But when Alfie played 33. Qg1 instead I went into autopilot and played what I was going to play against the move I’d expected without any further consideration.
Now, with a skewer coming up, 33… Rxd3 is winning very easily, but there’s a significant difference after…
… because g2 is available for his rook so White has the tactic, which of course I’d completely missed…
… which was accompanied by a draw offer.
There are quite a few variations to consider, and, running towards the end of the session, I used up too much time trying to work them out so had little choice but to accept.
We’d both considered the pawn ending after 34. Rxg6+ Qxg6 35. Rg2 Rxf2+ 36. Qxf2 Qxg2+ 37. Kxg2 Rxf2+ 38. Kxf2. Yes, it’s another OPP ending: I was wondering if I had some sort of sacrificial breakthrough on the queen side but I don’t and the position is, according to the engines, drawn after either 38… a5 or 38… c5. After anything else White plays 39. b4 when his OPP apparently wins.
In this line White also has the option of 36. Rxf2 Qxg1+ 37. Kxg1 when Black can choose to keep the rooks on the board by playing, say, 37… Rh8, but that also appears to be equal.
Another try for Black is to head for a RR v Q ending after 35… Qxg2+ 36. Qxg2+ Kh7, again with probable equality.
There’s also yet another option for Black, which neither of us had considered at all. Instead of taking the rook I could play 34… Kh7 when Black’s attack looks, superficially, stronger. Stockfish analyses 34…Kh7 35.Rg7+ Kh8 36.Rg5 Qxd3+ 37.Kg2 Qe4 38.Kf1 Qb1+ 39.Kg2 Rxf2+ 40.Qxf2 Rxf2+ 41.Kxf2 Qxb2+ 42.Kg1 Qc1+ 43.Kg2 d4 (43…Qxc3 44.Rh3) 44.cxd4 exd4 45.Rh3 when Black has queen and some extra pawns against two rooks. At first it thinks Black’s winning but, after further consideration, doesn’t seem at all convinced that he can do much about White’s plan of Rhg3 followed by a perpetual along the g-file.
So perhaps a draw was the correct result in the final position but my carelessness on the previous move threw away the full point.