[Event "BCCC 2017"]
[Site "Mechanics Institute"]
[Date "2017-06-01"]
[White "James Watson"]
[Black "Rodney Jacobs"]
[WhiteElo "1863"]
[BlackElo "1695"]
[Result "0.5-0.5"]

1.d4 d5 2.c4 e6 3.Nc3 c5 4.cxd5 exd5 5.Nf3 Nf6 6.Bg5 c4? {Still asleep or something. For this game I had checked out some very interesting lines in which Black plays c4, and where White, in order to make something of it, has to know a precise sequence which is very hard to see at the board. However I simply had a brain fade at this point. c4 should be played later if it is to be played. (RJ)} 7.Bxf6 {Interestingly, both my programs feel e4 is the best move for White here. (RJ)} gxf6 8.g3 h5 {I was hoping James might think c4 and gf was some deep plan to open up attacking lines.(RJ)} 9.Bg2 Be6 10.O-O Nc6 11.b3 {Probably not the best. The engines give e4! (RJ)  (JW) e4 is a theme that cropped up several times throughout this opening, I did consider it but wanted something that wouldn't require such presicision} cxb3 {Bb4 was much better. (RJ)} 12.Qxb3 Qd7 13.Rfd1 Be7 14.a4 {The engines prefer Qb2 or Rab1. (RJ) (JW) Rab1 is something I considered and knew it would be objectivly better, but from a human perspective I felt sure Rod would castle queenside} O-O-O {Castling into it. But the only way I was going to get some counterplay was with my rooks on the g and/or h files. I felt I had to risk it. (RJ)} 15.a5 h4 16.Nxh4 Rxh4? {At the time I knew this was dodgy at best, and the engines confirm it. But I was still completely disheartened by my blunder on move 6. I felt I was going to go down so I may as well do it flailing.(RJ)} 17.gxh4 Rg8 {Stockfish 7 says this is best. But it gives White an advantage of 3 here. (RJ)} 18.Rd3 Nxd4? {Hoping that if 19. Rxd4  Bh6  winning the Whjite bishop with some play. (RJ) (JW) As soon as I saw this I doubted it could have any degree of soundness} 19.Qa4? {I had thought this would have been the best response. But White can happily take the rook ....19. Rxd4  Bh3  20. Qxd5! with a massive advantage. (RJ) (JW) On the other hand I knew that Rxd4 was bound to be crushing for white. I saw 19.Rxd4 Bh3 20.Qxd4 easily since I'd considered the theme a bunch in ealier lines. But even though I knew it should be easily winning I felt like this might be a controlled way to win where there wouldn't be much room to mess up} Nc6 20.Nxd5? {Rg3 with a winning game. (RJ) The programs now say it's about even. (RJ) (JW) This deserves ?? it goes against the point of avoiding Rxd4 a couple of moves prior. Instead white has plenty of ways to simply win in a manor that doesnt allow counterplay} Rxg2+ {I was pretty happy with this. But simply 20......Bxd5 is better. (RJ)} 21.Kxg2 Bxd5+ 22.e4? {Kf1. (RJ) (JW) or f3} Qg4+ 23.Rg3 Bxe4+ 24.Kg1 Qxh4 25.a6 b6? {After the game James told me I should have played simply Bd6 here. He was right. (RJ)} 26.f3! {Now White's on top again. (RJ)} Bc5+ 27.Kg2? {Neither of us realised it then, but the programs say this is a critical error. The correct move is Kh1! (RJ) (JW) it's a subtle point but playing Kh1 prevents the rook fork in the line starting Bd4. It's easy to miss because Kg2 seems natural} Bd4! {After agonising over this position for 13 minutes I was delighted to find this move. It effectively keeps the White queen from the defence of the kingside. The fact that it attacks the White rook was just a side benefit. (RJ) (JW) I'd almost be inclined to give this !!. Infact it's the only defense in the position and Rod did very well to spot it.} 28.Rc1 Bxf3+! 29.Kxf3 Qh5+ 30.Kg2 Qd5+ 31.Kh3 Qh5+ 32.Kg2 {Overall, an appalling swindle. But I was very happy to be able to stage something of a fightback after making a beginner's blunder on move 6. I was simply lucky at many moments when James made an error or  there just happened to be a saving resource. At the end, were the perpetual check idea not there, Black was plainly crushed.(RJ)} {#d}