[Event "Teters 2016"]
[Site "BCC"]
[Date "2016-08-18"]
[White "Rodney Jacobs"]
[Black "Bas van Riel"]
[WhiteElo "1660"]
[BlackElo "1880"]
[Result "0-1"]

{584MB, Fritz11.ctg} 1.e4 d6 2.d4 g6 3.f4 Bg7 4.Nc3 c5 5.d5 {(BvR) Fritz indicates a strong initiative for white after the following variant, in contrast to my own assessment when I played the move. (RJ) Yes, this is the most favoured move on Chessbase. Interestingly, both Deep Shredder and Stockfish 7 prefer dc!} (5.dxc5 Qa5 6.cxd6 Bxc3+ 7.bxc3 Qxc3+ 8.Bd2 Qc6) 5...a6 {(BvR) 5...Bxc3 would have made for an easier game to play. As it happened, I got into more trouble and complications than I had bargained for. Most of the game I had to try and survive...(RJ) Chessbase gives Nf6, Both my engines in fact prefer Bxc3.} 6.a4 Qa5 {(RJ) Not an alternative on Chessbase. I was quite happy for my bishop to be forced to d2,} 7.Bd2 Qc7 {(RJ) Interestingly, both my engines far prefer Nf6.} 8.Nf3 Nf6 9.Be2 {(RJ) Recommended by Stockfish 7,which gives White a .46 edge here. I was really wanting to go for e5 here, pretty much on the basis that the opening up of the position and resulting wild play might lead to my advantage, with Black's Queenside undeveloped. But I wasn't game. Stockfish 7 gives the line 9. e5 de 10. fe Ng4 11. d6 ed 12. Nd5 Qc6 13. Ba5 Kf8 14. Nb6 Bf5 15. Bd3 Bxd3 16. Qxd3 Ra7 17. ed, changing its assessment to a 1.39 plus for White.} O-O 10.O-O b6 {(BvR) to prevent 11.a5 with space advantage for white. (RJ) My engines prefer Re8.} 11.Qe1 {(RJ) Recommended by both engines.} e6? {(BvR) a serious tactical mistake, which could have been punished by Rod (see added variation)} 12.dxe6 {(RJ) Yes, this inaccuracy dissipates most of White's advantage.} (12.Qh4 exd5 13.e5!) 12...Bxe6 {(RJ) The engines indicate this gives White back his advantage. fe is necessary.} 13.Qh4 {(RJ) It's fascinating that Deep Shredder (rated about 2800) says this is the best move, giving White an advantage of 2.2. But the far stronger new Stockfish 7 (around 3350!) says that it only gives White a .95 pull, and that f5! is better.} Nc6 14.Ng5 {(RJ) Both engines get excited about f5, giving it a 2.7 and 1.9 advantage respectively.} h6? {(BvR) Overlooking the following variation. After the game we analysed the perceived threat for checkmate involving the rook sacrifice on f6, not spotting the interposed 17....h6 defence. (RJ) No, I'm not sure it's a mistake. One of my engines recommends it, the other Qd7.} (14...Nd4 15.e5 {(RJ) No, this would be a mistake. f5! is winning.} dxe5 16.fxe5 Qxe5 17.Rxf6 h6) 15.Nxe6 fxe6 16.Bc4! {(BvR) of course (RJ) I thought so too, and was very happy at this point. But Deep Shredder feels Qg3 is stronger and Stockfish 7 recommends f5!} Nd4 {(RJ) Best.} 17.f5 {(RJ) The computers prefer Qh3} g5? {(BvR) Now am going to suffer until right at the end, where I miraculously escape defeat.} (17...gxf5 18.exf5 d5) 18.Qh3 {(RJ) The programs like Qg3 better,} b5!? {(RJ) d5 was better.} 19.axb5 axb5 {(RJ) It's fascinating that the engines both say this is a mistake, calling for d5! An indication as to how complex the position is.} 20.Rxa8 Rxa8 21.Bxe6+ {(RJ) I agonised over this for a long time, including looking at the sac 21. Nxb5 Nxb5 22. Bxd6ch. I was right not to try it. As Bas later pointed out, White has nothing after 22....Kf8. But I did not even contemplate the line the computers say is best, the winning 21. Nxb5 Nxb5 22. fe!} Nxe6 22.fxe6 {(RJ) The engines say White has a 1.5 edge here.} Qb7 {(RJ) Best} 23.Qf5 {(RJ) Also best.} Rf8!? {(BvR) Complicating matters in a precarious position for black, but with white in some time trouble} 24.Qg6 {(BvR) a very strong alternative would have been 24.e5!} (24.e5) 24...Qe7 {(BvR) 24..is black's best option according to Fritz (see variation) but that did not appeal to me at all. Unbeknown to me when I played it, 24..Qe7 is even worse, as I thought (very wrongly) to follow 25.Nd5 with 25...Qxe6 (see the next variation). By now Rodney had only 6 minutes left.} (24...b4 25.Nd5 Nxd5 26.Rxf8+ Kxf8 27.exd5 Bd4+ 28.Kh1 Qg7) 25.Qf5? (25.Nd5 Qxe6 26.Bc3 Qxe4 27.Nxf6+) 25...Ng4! {(BvR) This move took 16 minutes, leaving me with only 3! However, it was worth it, more or less equalising.(RJ) This did throw me.However, weirdly, it's an inaccuracy! My engines give Nh7! with equality. See following notes.} 26.Nd5 {(RJ) Best.} Bd4+ 27.Kh1 Qg7 28.e7 {(RJ) Again best.} Rxf5 29.e8=Q+? {(RJ) A key mistake, in time trouble. Rxf5 and White still has an edge.} Rf8 {(RJ) Now it's even.} 30.Qe6+ {(RJ) As so often happens, following one mistake with another.} Kh8! {(RJ) Calm, also under time pressure!} 31.Rxf8+?? {(BvR) we were both down to less than 2 minutes on the clock, and Rodney made the last mistake in a equal position. See 31. Rf3! (variation)} (31.Rf3 Nf2+ 32.Kg1 Rxf3 33.gxf3 Nxe4+ 34.Be3 Bxe3+ 35.Nxe3 Nd2) 31...Qxf8 32.Qf5? {(RJ) White is so badly lost now that the best move is Bf4.} (32.Bf4 Nf2+ 33.Kg1 Nd3+ 34.Kf1 Nxf4) 32...Nf2+ 33.Kg1 Nxe4+ {(RJ) A wild, interesting game, with Bas keeping his cool when things got heated! The position around moves 12-13-14 is particularly fascinating and complex.} {#R}