Beginner game - where did I go wrong?

This is a game I lost today. I seem to have played a lot carelessly in the last few moves, but where did I go wrong in the beginning? I think by the 25th move or so the game was already in my opponent's favor. 

I've tried to annotate how I thought during the game as much as I can. This is my first post so please point out I can improve the annotation. Thanks!

I would have castled for move 6.

12.Nxb3 13.axb3 doesn't create a weakness at all. You should know that doubled pawns aren't automatically weak.

On move 6 you just castle. You don't play 5...d6 and then 5...d5. Costs a tempo. Maybe you could still fight for it with 8...Be6? But already it is bad. 

13.cxb is a terrible move on white's part. Black is actually not so bad with the pair of bishops.

You don't look so bad after you win a pawn back, but it remains the case it is an isolated pawn and the bishops need to do some flywork there.

Now, I can't assess that endgame just like that either. You might have picked the knight up maybe. If not, 28...Be5 might have been an active try. Giving up the bishop was a shame anyway.

After that, the e-pawn is lost anyway. Even Bg6 And Re7 cannot change that because of the knight and f3. 

He might have considered Nxb7 there too... 

After that it is even in material, but I worry somewhat about the black pawns on the queenside. You had 36...b6 or b5. Your rook is still attacking pawns also, so I dont see how that is lost immediately. it's at least worth a try.

Yeah if you can castling makes the Ng5 move worse than useless.

On 6...O-O was much better.

On 8...Be6 was an improvement, but it was already too late.

On 11.Bxg5, you cannot say he missed Bxc6. It may have been slightly better, but was clearly not an "oversight."

On 11...Nd4, you played a good move, but you over-stress the importance of doubled pawns.

On 12...Nxb3, you did well by gaining the Bishop pair, but his structure is actually improved after 13.axb3 gaining a semi-open file.

On 13...Bf5 14. Rd1 O-O may have been better, with pressure on f2 and d3.

On 14...Bd6, not exchanging because of material deficit was pointless, since it made your piece less active. Rather, I suggest 14...Bb6! with maintained pressure on d4. If 15.Bxb6?! then 15...axb6 leaves you with a good position.

I've got to go now, I'll look at the rest of it if I have time. Keep up the good work! Analyzing your games is one of the best ways to improve! :)

7.d5 is terrible... use the game explorer and look at some different ways to handle these 1.e4 white openings that target your f7 with Bc4, and Nf3-g5. It's standard stuff and blocking with d5 is generally not going to work.  

In your game, after 7...Nxd5 8.Nxf7 Kxf7 9.Qf3+ and Black has nothing but problems.

After 8....f6? (better was 8...Be6) 9.Bxd5 would have simply won a piece (9...fxg5?? 10.Qf7#)

Well... I agree with all that has already been said about white opening the a-file with the pawn recapture... it was quite a blunder not to.

Rather than exchanging the knight for bishop the move I like is (12...Rf8! It's a tricky little move targetting f2 with a little tactical edge. Suppose he makes the visually appealling 13.Be3?? Ne2+ 14. Kg1 Bxe3 winning the piece. 

completely agree about Bb6 preferable to Bd6

19...Bc7 better than Bb4 

... I'll look at the rest of the game in a bit...

People have already mentioned a couple of improvements (such as the simple 6... 0-0).

However, despite the inaccuracies, by far your greatest mistake was the sudden blunder 36... Bxb3?? which was a clear moment of oversight to the immediate danger to your king.

I haven't analysed the game deeply, but you seemed to have gradually improved your position after the error in the beginning, reaching an equal position, possibly even superrior with your two bishops in an open board, and the closed c file (18... c6 was a called for, yet a great move). I was also slightly saddened to see 27... Rf7 instead of the natural and clearly better 27... Rd8, siezing the open file. However you are aware of this. The natural moves are, more often than not, the best.

At a glance, on the last move 36... Rxb3 looks drawish, but I think only you can win.

part II 

22...Bc4 targetting f2 again, it looks dangerous for Black, but I think it's strong. If Nxe5 Bxf2+ Kh1 Be4 ... and black threatens Rf5 among other things.

I think 23...Ba5 was interesting. Any difference between N and B is overridden by the advanced pawn. That is the key feature of this position. It's also really hard for a patzer like me to calculate correctly. after 23...Ba5 24.Nxd3?! exd3 ... Stopping that pawn gets quite tricky. I'd rather have black... 

I think 27...Re4 may have been better than either Rd8 or Rf7 but when you start needing to defend pawns with rooks like that it's not a good sign.  

31...Rd4 wasted a tempo and was not a good move -- You found the right idea but a move late with Rd1+ and then Rd2... but why would you think you have "lost the game at this point"? ... What was his winning advantage? No need for despair IMO.  

36...Rxb3 and the game is a long way from over. 

He could also ignore stupid advices, which sounds fair.

@BonyX3k, Na5 doesn't really improve and doesn't get you a pawn after 8.Bb5+ Bd7 9.Bxd7+ Qxd7 10.c4 really doesn't change anything so Na5 is just considered a normal move.

1...e5 is the one and only "politically correct" answer to 1.e4 for a newbie.

If, by chance, he manages to absorb its intricacies and subtleties properly, he can well keep it for a lifetime.

I think if anything e5 is a must for a begginer, when i was coached I was told as a begginer playing open games is as important as learning the swing in golf, you cant play fade, draw spin etc if you cant swing correctly in the first place. Saying a beginner should be playing a sicillian is nonesense, most beginners struggle with basic tactics never mind the wild/deeply theoretical sicillian systems

Despite all your "terrible" moves, as JG27 pyth points out, 36...Rxb3 and you have a good game with (assuming 37 RxRb3 BxRb3 38Nxb7, probably best for white) you have the bishop and a strong passer.  I say, for your level, you played well.

Thanks for all the comments! Noted all your advice and played out the suggested moves. I seem to have missed a lot.

I think,

1. I need to explore the e4 e5 opening a lot more, especially the pitfalls. Should use Game Explorer more.

2. Am too worried about doubled pawns, especially when I could've created a half-open a-file.

3. Could've had a good fight if not for the final blunder

4. There's a lot to improve.

Thanks again!

d5 did not lose the game, don't be ridiculous.

giusto.

6...d5 is a terrible move, which did not lose the game- as white replied quite poorly.

The old (and likely faulty) axiom "in chess the winner is whoever makes the penultimate mistake" probably applies here.

1. I need to explore the e4 e5 opening a lot more, especially the pitfalls. Should use Game Explorer more.

Not yet, openings can wait for later (or quite later). The fundamentals are the priority.

2. Am too worried about doubled pawns, especially when I could've created a half-open a-file.

The only problem I can see about a doubled pawn at your level, is that your pieces may have greater mobility because of the doubled pawn, and so it wmay be easier to blunder one of them.

3. Could've had a good fight if not for the final blunder

Just try to make the penultimate blunder yourself, and leave the final one to your opponent.

4. There's a lot to improve.

Sure enough, but with a lot of hard work, this is absolutely manageable.

As in tactics?

yep, but don't neglect positional understanding. my rating skyrocketed once i put effort into understanding positional basics (where do i place my pieces so that they're working towards a common goal, et al). the thing is, once i figured out positional stuff/planning, the tactical opportunities started falling into my lap.