Help understanding a move and game feedback

Here is the game:

Now my main question is this: Why on earth does the computer like 17..Qe7 instead of the move played. I happen to think that this may be a very computer like move but would like to hear some thoughts. 

To me this move is begging for trouble saying "I don't care that my bishop is lose and while I'm at it I might as well put my queen in front of my king and urge white to put their rook on the e-file."

I did consider h6 to be better right after I played 17..Bf6 though. Funny how that happens.

Other comments about the game:

1.Nc3 is not an opening move that I would ever play. I definitely wasn't expecting it and was out of book.

4..Bxe4 was a bummer once I realized that my b-pawn would be hanging after the exchange. I think black is still fine though and I would probably play it again. I don't really like the 4..Qd5 idea.

I thought 10..Ne5 would be interesting but it would be a classic example of hope chess I think. What appealed to me was the chance to win some material if say something like 11.Nxe5 Qxe5+ 12. O-O Bxa3 trying to highlight white's loose rook but obviously 13. Qxa3 would be a rude awakening. It is fun to keep track of thoughts like this as examples for my calculation abilities.

10...Ne5 11.Nxe5 Qxe5 12.Qe2 only helps white it seems since white gets to move their queen out of the way of their d-pawn without losing time.

White's bishop sac was rather interesting but I didn't think that it would work. To my surprise, the computer says that it was the best move. It was hard for me to play the passive 14.Nf8 but keeping cool under pressure can save your hide.

Thanks for your time! Feedback welcomed. 

Well, it doesn't make that much difference to the engine in a highly tactical position, so trying to read the computer's mind is going to be hard and probably is a waste of time.  That said, qe7 allows meeting qh5 check with ng6.  And qh5 is the engine's top move after your bishop move.  

I traded because 4.Qf3 seems like an awkward move and I think blacks position is easier to play.

Is Qh5+ that big of a threat though? I don’t see it. I guess the main point would be a positional one since after Qe7 black still has prospects of king protection and castling?! That would make sense. And then something like Bf6 can always be played after.

I’ve noticed that it’s a common theme with higher rated players to have a more concrete approach with “I can meet this move with that” kind of thinking. I am working on being more like that.

Qh5+ is not that big of a threat, but the engine isn't showing that big of a difference in the moves.  After qh5, g6, qf3, the engine thinks bxn is a further step in the wrong direction, and if that is true, what value did you get out of the bishop move?  

This is really the wrong type of position to be studying the computer's non-tactical evaluation differences so I'm going to check out.  Best,  

"I traded because 4.Qf3 seems like an awkward move and I think blacks position is easier to play."

Unexpected, perhaps, but awkward? The fact is that the move created a single, easily handled threat (Qxf5), There are at least three good ways to meet that threat while improving your position (4...Qd5, 4....e6, 4....Bg6) all of which left you with a slight advantage. 4...Bxe4 trades a fine bishop for a knight and gets rid of your only developed piece. It's a poor move. Trading a piece is the easiest way to resolve the tension in a position, but you should resist that temptation Try and make trading your last option, after you've looked at everything else

double post

"11.Nxe5 Qxe5+ 12. O-O Bxa3"

you have white castling out of check here. it probably involves a queen trade in game.

That makes sense. I’ve heard “to take is a mistake” be thrown around but I didn’t see that the b7 pawn was hanging at the time I played the move and I had illusions of getting ahead in developement.

Thanks for your reply! Something else that I need to work on.

I noticed that later as well. I gave 12.Qe2 later and failed to go back and correct the previous. Thanks for pointing it out.

....a comedy of blunders.

I never claimed to be good but hey, at least you got something out of it.

Very good points. Thanks for your help.

Some moves don't aim to generate threats but to prevent the rival from developing his. In this specific situation, all Black needs is moving his King into a safe position. Otherwise, his pieces will have to remain nearby for protection... meaning no active play for Black until his King is safe.

So, the engine suggests the plan Qc7–e7, Nf8–g6, Bd4–f6 and Nd5–c7, followed by 0-0 (in between or at the after those moves), as the fastest and bulletproof way to achieve: Black King in safety.

17...h6 works but it has the drawbacks of losing the g6–square as a safe spot for the Nf8, and weakens the light squares in the vicinity of the own King. Should White not be a piece down, 17...h6 wouldn't even be considered.

And, what's wrong with 17...Bf6 straight away? Chess is about precision, and precision usually involves the order in which a player moves his pieces into the squares dictated by the plan he's following. In this case, having the c7–square free for Nd5–c7 should White play c2–c4.

Amazing insight as always. This is the reason I post questions like this. Freeing the c7 square for the knight in case of c4 is very instructive. Thank you very much for your post.

I could of asked myself “If c4, where exactly is my knight going?” At first glance, the b6 square seems like a really bad answer because of b4 and black can ask themselves why they allowed this as c5 and re1 are both threats. I think re1 would have to come after b4 though.. The other options also seem to misplace the knight as it feels like the knight wants to be on c7 because of blacks e pawn. Nf6 would put the knight where the bishop wants to be and Ne7 cuts the queen off from the kingside.

Knowing what questions I should ask about a position is helpful in my opinion and I look forward to using this as an example in the future.

@K_Brown: Thanks but the reasoning is very simple, once you acknowledge that "piece activity" is the most important particularity when assessing a position.

If "asking the right questions" is the matter, you may start with: Are the Black pieces directed towards anything in White's position? And soon you notice that no, they aren't. Then you ask: Are the White pieces directed towards anything in Black's position? And you notice the pressure against the e6–pawn. You'll also notice that White can increase the pressure "soon" with Rf1–e1. Now, the pawn is protected by the Nf8, but posted at that square, the Knight interferes with 0-0. Furthermore, the e6–pawn isn't important because of its material value, but because if shelters the King on the e–file; therefore, once castled, you can let go the pawn if you get an attack in exchange (remember, piece activity).

So the Nf8 has to move but at the same time the e6–pawn has to be protected at least twice since White can use both a Knight and Rook against it. That's how you notice Qc7–e7 and Nd5–c7. Also, you should beware of tying down high value (the Q in this case) into the defense of a pawn, and only accept it if of temporal nature... and there you get the reason behind posting the Bishop and Knight on f6 and g6: They both support e6–e5, ending White's N+R coordination on e6 and supporting your Knights should they deem profitable to go Ng6–f4 or Nc7–e6–d4.

Choosing a plan (where to post the pieces and pawns, in what order, as to increase the own threats and, or deny the opponent's) should be natural for every chess player, no matter if 1700 or 2700. What varies is the precision when choosing a plan or following it according to the ever-changing circumstances. That's the main difference between peasants and royalty. Take the following annotated game as an example:

Thanks for the post.

Is that annotated game from a book or site? I enjoyed it and those particular annotations were well written and instructive.

I think you are asking the wrong question and btw the engine evaluates 17...Qe7 and 17...h6 almost the same if you let it run long enough.

The engine thinks VERY tactical. The engine doesn't play it with the idea of playing the passive Nc7, that move is still fine though. Instead it likes to answer a move like Qh3 with Bxb2 undermining the knight or Bf6.

There are several ways of playing this and the computer's suggestions aren't always best for humans, because they assume perfect play. You want to reach a position which you can easily convert in a win without complications. I like the following one:

And black will bring the king to safety by playing Kd8-Kc8.

This stuff bothers me...

"1.Nc3 is not an opening move that I would ever play. I definitely wasn't expecting it and was out of book."

How can you be out of book on move 1???  Follow opening principles:

Control the center.

Develop toward the center.

Castle.

Your over complicating the first move of the game.

All I was saying is that I don’t believe in or follow 1.Nc3 theory. I can just play normal moves and be fine I feel. 

1.Nc3 blocks the c-pawn for no reason. I can’t see how anyone could advise it. How am I overcomplicating?

If someone plays 1.a4 or the like then I am out of book as well. I haven’t studied the book continuations for that since it seems like a waste. Being out of book just means that I think for myself from that point on. Nothing more.