I haven't analysed with an engine because I want to learn from the mistakes I made but can anyone give any constructive criticism from this one?
At many points sacs and tactics seemed to not quite work. I feel like my play generally must have been pretty good though.
You came out of the opening with a big advantage.
11 Nb5 was not accurate, as the threat Nc7+ is easy to parry and as he will drive your knight back with 14...a6. Better develop Bc1 instead of moving the already developed Nc3 a 2nd time.
17 b4 weakens your squares on the c-file, where he already has his rook. Better move your queen away from the d-file, creating vis-a-vis of your rook and his queen.
18 Nxf6+ just hands him a knight for nothing. Before this 'sacrifice' you still held a winning advantage. Why did you sacrifice? What were you thinking?
Qapla ! The opening played by Black is very provocative and totally refuted by your glorious development !
I disagree with Stockfish on the assessment of 11.Nb5. I think it is the strongest move in the position, with a consistent follow-up : Bb3 to evacuate the c-file and Be3 to win the queen (with Bb6) if Black plays a6. The center is already semi-open, you need to bring your rooks while Black's king is still unsecure.
Komodo's line runs as follows :
11.Nb5 Rc8 12.Bb3 Ne5 13.a4 a6 14.Be3 Nc4 15.Bxc4 Rxc4 16.Bb6 Qc8 17.Ndc7+ and Black should resign. I hope this is self-explanatory. Black has nothing better on the 13th move, e.g. 13...h5 14.Be3 Rc6 15.Nxa7 Ra6 16.Nb5 Rc6 17.Nbc7+ Rxc7 18.Bb6 Qc8 19.Nxc7+ Ke7 20.a5 with the plan Qd4-Rad1-Qxd6.
The real mistake comes one move later : 12.Bf4? does not coordinate with the knight on b5. The one-move threat on d6 is neutralized by the move that Black wants to play anyway (Ne5, attacking the bishop on c4). I don't like 13.Bxe5 but the harm is done anyway.
Komodo again disagrees with Stockfish's blame on 14.Qd3. This move is Komodo's second choice after 14.Be2 and their assessment are not very different.
Black digs a deeper grave for himself with the subsequent moves. 17.b4? misses the opportunity to play 17.f4!, further opening the center on Black's uncastled king. 17.Qe3 is also good, setting up an array of threats, but it's a computer move that needs a lot of calculation (no less than six branches on the first move : Ne7, Bf8, Bxd5, Bd7, b5 and Nh6). I think that Komodo's first choice 17.f4! is much clearer and more purposeful. 17...gxf3? 18.Qxf3 (on its way to h5) strengthens the attack, and 17...exf4 18.Rxf4 b5 19.Rxg4 bxc4 20.Qg3 is a nice pseudo-sacrifice where White recovers the piece with interest immediately. Any other move will make fxe5 possible, with more open lines.
Let's admit that you play b4 on move 17, forget about f4 and Qe3. If you play b4, you have a plan of pushing it to b5. It's the only possible idea behind 17.b4. Indeed, 18.b5 would preserve White's winning advantage. So 17.b4 is a move that cannot be criticized for itself, but merely because better opportunities were missed. Yet White has to be consistent after that.
On top of the natural b4-b5 push, Komodo actually prefers a line that looks like an improvement on the game : 17.b4 Ne7 18.Bb3 h6 19.Nxf6! Bxf6 20.Qc4! with the dual threat of Rxd8 and Qf7#. The knight sacrifice would be correct if the square c4 were available for White's queen, and it isn't in the game, causing White's demise.
After 18.Nxf6??, not only is White a piece down, but Black finds some good square for his pieces, like g6 for the knight. White could save the game (not more than that, don't believe Stockfish's optimistic assessment) with 22.Nxb5 that forces further concessions by Black for the piece. Komodo's line is 22.Nxb5 Bxb5 23.Bxb5+ Ke7 24.Rd7+ Qxd7 25.Bxd7 Kxd7 26.Rb1Rxc2 27.Rxb7+ Kc8= ; it's very similar to Stockfish's line, only the assessment is different, and I trust Komodo in presence of unusual material imbalances (Q+2p vs R+B+N).
I hope this clarifies where White's play could be improved and why the sacrifice Nxf6 could have worked in a different position, but definitely not this one.
Sorry I don't mean to sound hostile but I didn't really want someone to barrel in with 'the engine says'. Never mind.
It's not what the engine says. It's what we can understand about the recommendation of a very strong player (who turns out to be an engine). Sometimes Komodo's line is useless because it's based on calculations that appear random (e.g. 17.Qe3). Sometimes it's very instructive (e.g. the justification of Nxf6 with just a couple of moves inserted), especially when one compares Stockfish (quick assessment, brute force) and Komodo.
There is a big difference between letting the computer do the analysis and using a computer to guarantee the quality of an analysis.
I totally agree with
@A-Cielbleu.
When analyzing and learning from our mistakes, we want to be as objective as possible. We need variants and understanding of those variants.
@petaQ, you have a great feedback, constructive criticism.
@A-Cielbleu's post is not about recommending some computer moves, it is all about understanding the position.
If you want, could you explain, why you didn't want posts like this?
I would really like to understand your thinking.
The
@A-Cielbleu is very profound as always.
So in short, you were better and played well until you gave it away with 18 Nxf6?. Even then you did not have to lose and you probably could save to a draw with 22 Nxb5.
All that is very insightful.