Computer says ?! but... why?

Hi everyone,

I've been learning chess for just a few months and this is something I'm struggling with when analyzing my openings and middle game: the engine tells me I messed up, but the why seems VERY unclear to me. I'm sure this will get better with time, but here's a good illustration of this problem:

I played Ngf3 with the goal of preparing O-O. I'm guessing this is an inaccuracy because black can go Bd3 and block my castling? The engine suggests Nb3 instead but WHY? How can I come up with that move during the game? What would your thought process have been in this position?

Thank you!

well i think you have already answered your own question! Nb3 prevents black from Bd3 which won't let you castle.

Yes! That is exactly right!

Great! Now on to the next part of this position then: How do you come up with Nb3 as the best play here? 

Thanks again, all.

Nb3 prevents Bd3 which allows castling and looks at c5. That’s what I see, at least. However, Ngf3 is quite natural, but wrong. 

Nb3 does more than prevent Bd3.

A question you might consider as you are evaluating this position is "Where do you want your knights to end up?"  Not just right now, but where do you think they will be able to exert long-term pressure against black's position.

The places I see are e5 and c5, with c5 being the better one (there is no pawn that might later kick the knight away).

So, absent immediate tactics, I would be looking at the following maneuvers:

Ngf3 -> Ne5 and Nb3 -> Nc5, with a potential Ne5 -> Nd3 to add additional support to c5 if the e5 knight should get kicked away.

Now, with that in mind, both Ngf3 and Nb3 seem ok.  Yes, Ngf3 allows Bd3, after which you can do Nb3 to attack the bishop.  However, this might allow the black bishop to maneuver over to the queenside and have long-term influence over the a6-f1 diagonal.  Given that, Nb3 first is indicated.

Another consideration that might lead you to this is that there is no need for a quick castle in this position.  Yes, castling is nice and would allow you to eventually double up your rooks on the a-file, but black doesn't have an immediate way to attack your still-centralized king.  Taking a move to prevent Bd3 before preparing to castle will help to ensure that you can get castled.

Im not sure

There's nothing wrong with Ne2, either.

That's typical computer nonsense, you shouldn't pay attention.

A reasonably working engine (which is not the cloud chess dot com stockfish stuff) should have no trouble suggesting Ngf3 as a strong move.

1.Ngf3 is just fine, because 1...Bd3 does not work: 2.Ne5 and now either the bishop goes back and allows castling, or after 2...Bb5 3.Ra2 risks being trapped with b2-b3 and c3-c4.

I would play 1.Ngf3 without much thought, as the conservative 1.Ne2/Nb3 does nothing more than meeting a  possible move (Bd3) which is not dangerous at all.

What matters here is that you are a juicy pawn up with a structurally better position (weak pawn at a7), and you should get a winning position by just playing simple chess.