Fortunately I found something to complicate things and eventually give me the full point.
perrypawnpusher - obmanovichhh
blitz, FICS, 2010
1.e4 e5 2.Nf3 Nf6 3.Bc4 Bc5 4.Nc3 Nc6
The Italian Four Knights Game.
5.Bxf7+
The Italian Four Knights Jerome Gambit.
5...Kxf7 6.Nxe5+ Nxe5 7.d4 Bb4
An unusual but certainly playable defense. Including this game, the updated New Year's Database has 26 examples of this position, with White scoring 65%.
Stronger for Black, though, was 7...Bd6 8.0-0 Nc6 9.f4 Be7.
8.dxe5 Bxc3+ 9.bxc3
If Black keeps his wits about him, he can now play 9...Re8, giving back the piece but reaching a safe, even position where he has play against White's weakened pawns: 10.exf6 Rxe4+ 11.Be3 Qxf6
9...Nxe4 10.Qd5+ Kf8 11.Qxe4 d6
White is a pawn up – a situation that is always psychologically uneasy for me, as I feel like I've "escaped" in another Jerome Gambit, and I tend to play vapid moves that lead to even (or worse) positions. (Brian Wall may have a valid insight into this.)
12.0-0 dxe5
I am convinced that my opponent was already thinking about reaching a drawn endgame here. I was willing to go along with him, and hoped to out-play him – a notion that obmanovichhh would probably have scoffed at.
The fact is that we both missed 13.Ba3+ here, when the position becomes anything-but-drawish: 13...Kf7 14.f4 (to open more lines for attack) Be6 15.fxe5+ Kg8 and White wins Black's Queen with 16.Rf8+ Qxf8 17.Bxf8 Rxf8.
analysis diagram
White's Queen can now go munching on Black's pawns.
13.Qxe5 Qe7 14.Bf4
Again, too accomodating of Black's plan – I thought that I knew what I was doing – when there was a stronger option available.
After the game Stockfish showed that White could have shaken things up with the piece sacrifice 14.Ba3!? as after 14...Qxa3 15.Rfe1 Bd7 16.Qxc7 Bc6 17.Rad1 he would have had a very strong position, with the possible continuation 17...h6 18.Rd8+ Rxd8 19.Qxd8+ Kf7 20.Qxh8 when White is up the exchange.
14...Qxe5 15.Bxe5 c6
For the next 15 or so moves, White and Black play cooperative chess: we both aim for a Bishops-of-opposite-colors endgame that I think that I can win, and that my opponent thinks that he can draw.
There is another factor in play, however – we'll get to that later.
16.Rad1 Bf5 17.Rd2 Kf7 18.Rfd1 Rhe8 19.f4 Re6 20.Rd8 Rxd8 21.Rxd8 Re7 22.Rd2 g5 23.g3 gxf4 24.gxf4 Be6
Each side will put pawns on squares the same color as their Bishop, the Kings and Rooks will help in the blockade of the enemy pawns, and all will be peaceful in the world...
25.a3 Bc4 26.Kf2 b5 27.Ke3 a5 28.Rd6 Bd5 29.Rh6 Kg8 30.Kd4 a4 31.Kc5 Be4
Black has reached the position that he has been aiming at for a long time. So have I.
32.Rxc6
Surprise!
Truth be told, after this sacrifice the game is still drawn – but it is tough work for Black, made tougher by the blitz time control.
32...Bxc6 33.Kxc6 Re6+
Instead, according to Rybka and Stockfish, if Black centralizes his King to blockade White's passed f-pawn, and maneuvers his Rook to pin the Black King to the a-file in front of the passed a-pawn, then it doesn't matter if he loses all but one of his pawns: 33...Kf7 34.Kxb5 Rb7+ 35.Kxa4 Ke6.
analysis diagram
White will not be able to make progress.
34.Kxb5 Rh6
This also leads to a draw, but not an easily seen one.
35.c4 Kf7 36.c5
Now if Black can scoop up the two lagging pawns with 36...Rxh2 37.c6 Rxc2 and keep his Rook on the c-file to capture the White pawn when it Queens, he actually would have winning chances after 38.c7? Ke6 39. Kb5 h5! as White cannot stop the Black h-pawn.
analysis diagram
Instead of 38.c7?, White has to find 38.f5! when his Bishop will be free to stop the Black h-pawn and his f-pawn does not allow Black's King to wander too far from the center.
A reasonable continuation might look like 38...Rc1 (keeping an eye on the passer) 39.Bd6 (threatening to block the Rook's access to White's c-pawn, and possibly able to block checks along the b-file) Ke8 40.f6 Rf1 (Black's King and Rook are working well together in frantic times) 41.Be7 (safeguarding the f-pawn) h5 (to the rescue!) 42.Kb6 Rb1+ 43.Bb4 Rf1 44.Kb7 Rc1 45.c7 Kd7 46.Ba5 Ke6 (Black abandons his Rook in favor of capturing White's f-pawn) 47.c8/Q+ Rxc8 48.Kxc8 Kxf6
analysis diagram
– and suddenly, miracle of miracles, Black realizes that it no longer matters if he loses his last two pawns, as the game will enter the notorious "Rook pawn plus wrong Bishop" endgame, which is a draw. (Of course, if Black greedily tries to win the Bishop for his h-pawn, White will queen his a-pawn and win.)
I didn't realize that the endgame was that complicated, but I did think it would "set problems" for my opponent.
36...Ke6
37.c6 Ke7
Black can no longer count on 37...Rxh2, as White has 38.c4!? to help guard his advanced (and advancing) c-pawn.
Black can try the swindle 38...Rh3!? hoping for a King move by White, but 39.f5! wins for the first player, as in 39...Ke7 40.f6+ when one of the passers will Queen, for example: 40...Kd8 41.Kc5 Rd3 42.Bd6 Rf3 43.Be7+ Kc8 44.Kd6 Rf1 45.Ke6 Re1+ 46.Kf7 Kc7 47.Ke8 Kxc6 48.f7 Rf1 49.f8/Q+ Rxf8 50.Bxf8 and White's extra pawn makes all the difference in the world compared to the above analysis.
analysis diagram
38.Kb6
My turn to slip: this activates 38...Rxh2 as a drawing strategy, much as in the analysis after White's 37th move (for one thing, c2-c4 doesn't work as a shield as White's King is too far advanced) – ending up as before with a Rook Pawn plus wrong Bishop endgame for White.
White's winning move was therefore 38.h4!? which cannot be answered by 38...Rxh4?! because of 39.Bf6+! Kxf6 40.c7 and the pawn will Queen. Black defends better with 38...Kd8, but White can slowly advance his forces and win, e.g. 39.Kc5 Rh5 40.c4 Kc8 41.Kd5 Kd8 42.Ke4!? Rh6 43.f5 and either 43...Rxc6 or 43...Rxh4+ will be answered by 44.Kd5 and the Bishop pawns will continue to advance...
38...Kd8
Missing his last chance (38...Rxh2).
39.Kb7
At this point my opponent disconnected and was forfeited. He has to give up his Rook for the advanced c-pawn, and the ending thereafter would not be difficult for me to win.
