Unbalanced material evaluation

Why is black better or winning win white is up a minor exchange? Is there someway a human could better evaluate the position, or do engines just give black a big advantage because of a long computer line?

Sometimes the engines simply don't undersand the positions properly, after 12.g3 instead of 12.e4, you pick up the knight in the corner, and okay, black gets another pawn, but I believe white is much better despite the engine's evaluation. I wouldn't trust it so early as the 12th move (horizon effect).

Okay, after black's 13..g6, black about level material, but his position better (looking)? Probably. Is it winning like the engine says? I am not sure  yet. (i will update this later, if not ill post about it later)

Move 22, screw houdini's .21 evaluation, I know white is winning. The rooks are pathetic, and you have every reason to force black to induce a couple weaknesses and suddenly your pieces are monsters. This is my intuition talking, and what you played in the game, highlights some of black's difficulties in the position:

HE LACKS COUNTERPLAY! 

That is one thing engines suck at, positional chess, a small, quiet that helps you in the long term and down the line. I can guarrantee you, black would have lashed out for desperate counterplay later on if he stayed passive for a while. 

Another factor is the quality of the pieces really. The rooks and bishops suck donkey kong with a greater number of pawns on the board, while knights and queen thrive for such positions ---> weaknesses.

This game was no exception

I've seen queenless middlegames where white is down an exchange after 15 or 16 moves, and both sides have = pawns, but the problem is the rooks lack active play, and you can gradually improve your position.

Im trying to locate that queenless middlegame right now, its so hard to remember for me.

Here is the game, from where I remember it from. There was one amazing line in the opening where white allows nb4-nc2+ when white plays Kd2, allows Nxa1, idk about the queen exchange yet though, but somehow the engines thought the position was equal. 

IGNORE THE TRASH SIDELINE, I was trying to toy around with the position, seeing if it would trigger some of my lacking synapses. It didn't, as soon I find it, ill post it.

The point of material imbalences is who has the initiative will be likely better, or appear better and thats all that matters.

Yeah, the funny thing is this game happened right after I was discussing a line in the scotch gambit / two knights defense where black sacrifices two pieces for f2 and a rook and a nice advantage.

Whenever you have a N in the opponents corner consider it lost unless you can see a way out for it.
Therefore I would have tried something like 13... Rg8 threating to fork, g5, g4 drives away the night from f3 then you can play Bg7, Be5/Bd4 or advance the pawn to g3 and perhaps get the N out. Not sure it works point is you must make it priority to save the N.

what was he rated coz he played like a idiot also good job on becoming 1800!

He was 1800, but at his floor. And thanks for the compliments.

I'm not sure how Black defends after 2.e4. 2...Qb6 looks weird and 2...Ba6 isn't great since I don't think you'd mind swapping bishops with your pawn on d5. I think you have to just go after this move. At move 12 it looks like even material since the knight on h1 will be captured sooner or later. My only guess is that even though you have a knight and bishop for the rook it may be tricky to use them against Black's wall-o-pawns. And with the more active exchange and bishop pair you probably want to open things up instead of playing 13.c4.

I'm impressed that you understand c4 is a mistake. I didn't until much later. When playing the move I was trying to keep files closed to keep his rooks from becoming active, but I really just locked in my own light squared bishop while allowing his dark squared bishop massive influence.

Probably the worst move of the entire game, before that I was better and nearly winning, after I was nearly lost.

The other main mistake I made was Qc8+. It is too dangerous to go after the b5 pawn and being forced into a queen exchange helps the side with the rooks.