As many readers know, I have been a long-time advocate of the Bird Variation of the Ruy Lopez for Black (1. e4 e5 2. Nf3 Nc6 3. Bb5 Nd4). Recently I’ve started experimenting with Henry Bird’s other major opening legacy, the Bird Opening (1. f4).
I’m not planning to go off the deep end with this — I am only considering it as an occasional diversion from 1. e4 and 1. d4. For one thing, it’s hard to argue with statistics. According to ChessBase, White scores about 53% or 54% with those moves. With 1. f4, White scores only about 48%.
There are two times in life when you don’t want to score 48%: in an election (sorry, Mitt Romney!) or when playing White in chess.
On the other hand, statistics don’t really provide us understanding, and that’s really the goal in chess. If Bird’s Opening is bad, why is it bad? And are there lines that are more or less okay?
Now I’m no expert, having played (to date) only three tournament games with 1. f4. The first two games quickly transposed into something else. The first one went 1. f4 e5 2. e4, and we were in the familiar ground of the King’s Gambit (and I eventually won). The second went 1. f4 d5 2. Nf3 c5 3. e4!? and we were in the familiar (for me!) territory of the Bryntse Gambit (a game I eventually drew).
So it wasn’t until last month’s Western States Open that I got to play an authentic Bird’s Opening against IM Ed Formanek. He beat me in very instructive fashion, and in the process gave me some idea how White can get into trouble.
I’ll get to the Formanek game in a little while, but I’ll start by saying that one problem with the Bird is that it does not achieve what it’s supposed to achieve. If there is any point at all to playing f4 on the first move, it’s that you want to take control over the e5 square. And yet there are a variety of ways in which Black can frustrate this goal and get in the pawn break … e5.
First of all, Black can play 1. … e5 right away. As we know, that is From’s Gambit. But as noted above, it transposes into a King’s Gambit after2. e4, and I am willing to live with that.
A second, less well-known way in which Black can defeat White’s objective is my own pet line against the Bird, which I came up with back in 2000 or so. It could be called the “From Gambit Deferred.” After 1. f4 Nf6, many White players will continue 2. b3, intending to continue the plan of controlling the dark squares. I would then play 2. … d6 and then, after 3. Bb2, I would surprise White with 3. … e5!
FEN: rnbqkb1r/ppp2ppp/3p1n2/4p3/5P2/1P6/PBPPP1PP/RN1QKBNR w KQkq – 0 4
Unlike its better-known relative, which is a true gambit, the From Gambit Deferred is a gambit in name only. If White actually accepts the pawn, then Black is pretty close to winning by force. The position after 4. fe de 5. Bxe5? has only been seen three times in ChessBase. Interestingly, one of them was a game Szpisjak-Brooks from 1982, where Black was Michael Brooks, well-known American IM. So I’m not the first person to invent this line; perhaps it should be called the Brooks Gambit. After 5. … Ng4 Szpisjak continued 6. Bg3 Bd6 7. Nc3 Bxg3+ 8. hg Qd6 9. Ne4 Qd4 10. d3 f5 and Black wins material.
I played this idea twice against a class B player named Reuben Lenz and won both. He accepted the gambit the second time and the game went 4. fe de 5. Bxe5? Ng4 6. Bb2 Bd6 7. Nf3 Nxh2!
FEN: rnbqk2r/ppp2ppp/3b4/8/8/1P3N2/PBPPP1Pn/RN1QKB1R w KQkq – 0 8
If White takes either way, he gets mated with … Bg3+. Lenz soldiered on with 8. Kf2 Ng4+ 9. Kg1 Bg3 10. Rh5 Nf2 11. Qc1 (White’s pieces are fleeing to the far corners of the board!) 11. … f6 12. e3 Bg4 13. Rb5 a6 14. Rxb7 Qd5!
FEN: rn2k2r/1Rp3pp/p4p2/3q4/6b1/1P2PNb1/PBPP1nP1/RNQ2BK1 w kq – 0 15
The hunt for White’s rook is only a sideshow. The real target is White’s king, which would perish after 15. Rb4 Qh5 16. Rxg4 Qh1 mate. This explains Lenz’s increasingly desperate moves, as the game concluded: 15. Rxb8+ Rxb8 16. Ne5 fe 17. Bc4 Qd6 resigns.
Well, this is great fun. In any case, it should convince you that 1. f4, 2. b3 and 3. Bb2 simply does not work as a way of preventing … e5.
For that reason, I made up my mind that after 1. f4 Nf6 I would not play 2. b3 but would instead play 2. Nf3. Only in case of 2. … d5 (or the reversed move order, 1. f4 d5 2. Nf3 Nf6) would I continue with 3. b3. In that case I feel pretty good about White’s position. It’s basically a Queen’s Indian reversed where White has played a useful move, f4, with his extra tempo. I went to ChessBase and saw that GM Pavel Blatny has played this way a number of times.
But in my game against Formanek, I ran into the problem: What does White do if Black doesn’t cooperate by playing an early … d5? Mackenzie-Formanek went as follows:
1. f4 Nf6 2. Nf3 g6 3. b3?! (Playing this before Black has committed himself to … d5 is not recommended) Bg7 4. Bb2 O-O 5. e3 d6! (Sorry! No … d5 today!) 6. Be2 e5!
FEN: rnbq1rk1/ppp2pbp/3p1np1/4p3/5P2/1P2PN2/PBPPB1PP/RN1QK2R w KQ – 0 7
And once again Black has achieved his thematic break, … e5, with the utmost of ease. In this case it’s not a gambit, and it’s not even a pseudo-gambit. After 7. fe Ng4 the pawn on e5 is pinned, and so Black wins it back. Probably White should play this anyway, but I don’t see any problems for Black in this position. I played 7. O-O? ef 8. ef and suffered for a long time in a position where I was never able to generate any serious threats.
Interestingly, when I showed the game to Gjon Feinstein he knew all about this line and said it’s just good for Black. He didn’t think I made any serious mistakes after this, it was just a case of IM Formanek slowly and methodically increasing his positional bind.
But the picture is not all bleak for White. Gjon said that after 2. … g6 I just shouldn’t play 3. b3, the move that keeps getting White into trouble. After 3. g3 Bg7 4. Bg2 O-O 5. O-O we get to a position where White is playing what Jonathan Rowson calls “zebra chess” — he’s waiting for Black to commit to either … d6 or … d5. If Black plays 6. … d6 now, we can play 7. d3 and work towards e4. If Black plays 6. … d5, we can either play 7. d3 or we can play a move that I think looks intriguing, 7. c4.
But that will be a subject of another post (maybe).