How could I have played this opening better?

Question

This was a recent game on chess.com. The "Maximum" analysis tool they have says my first mistake isn't until move 9, but I felt like I misplayed the opening even earlier. Indeed black's white squared bishop came back to haunt me. So: am I correct and if so, what could I have done better?

I include the entire game, complete with several embarrassing blunders (I'm still very new), for the sake of completeness. It may be worth nothing this was a blitz game.

[fen ""]

1. e4 b6 {I don't see this response to e4 often, but I figure it allows me to control the center better.}
2. d4 Bb7
3. Nc3 e6
4. Nf3 Nf6
5. e5 Nd5 {I instantly regretted my pawn push here since, in particular, it opened up the long diagonal for my opponent's white squared bishop. Better was maybe 5. Be2 or Bg5?}
6. Nxd5 Bxd5
7. b3 Bb4+ {I was preparing for 8. c4, which as foiled.}
8. Bd2 Nc6
9. a3 Bxd2+
10. Qxd2 Bxf3
11. gxf3 O-O
12. O-O-O Qh4
13. h3 d5
14. exd6 cxd6
15. Rg1 d5
16. Rg4 Qf6
17. Be2 Rac8
18. Rdg1 g6
19. h4 e5
20. dxe5 Nxe5
21. Rg5 Nxf3
22. Bxf3 Qxf3
23. R1g3 Qh1+
24. Rg1 Qxh4
25. Rxd5 Rfd8
26. Rxd8+ Rxd8
27. Qe3 Rc8
28. Rd1 Qf6
29. Qe4 Qa1+
30. Kd2 Rd8+
31. Ke2

I resigned here.

Answer 1

Some thoughts, mainly about your opening (to be fair, this game jumped right from opening to endgame in short order):

• Your first four moves are very good, taking control of the center while developing.
• 5. e5. General rule of thumb: If you have pawns adjacent to each other and well-defended, you don't want to push one of them (unless you have a good reason! And achieve something concrete with it, say, if you had a bishop on g5). The reason is that by doing so you lose a lot of flexibility. As a pair on the same rank, the pawns control a line of squares of both colors, with no obvious opening to attack by your opponent while at the same time retaining the potential of two different pushes by yourself as soon as you see an opening in the opponent's structure in return. If you move one of them without a clear target, you weaken your own line of defense (as you noticed yourself, the bishop diagonal is now open) and you also commit yourself to one attacking angle, which means that your opponent now doesn't have to worry about the other as much anymore (you cannot easily play d5, so your advance has essentially come to a halt and is not much of a threat to him anymore). Bg5 is a much stronger move here, Bd3 is fine too (development + protection of e4).
• 6. Nxd5. Actually makes a lot of sense, despite moving a developed piece a second time, as it offers a way to "repair" your pawn structure in the center. A pawn on c4 makes it only half as bad, since then the e4 pawn has regained hope of being mobile again with the support of his neighbour.
• 7.b3. But then you fail to execute that plan. Why not play 7. c4 straight away? The pawn will be defended adequately by your bishop on f1 and scare back the enemy bishop with a gain of tempo. Why do you need more support of c4? The only way black can attack it once more is by ...b5, and he cannot do so because of Bxb5.
• 8. Bd2. Makes sense.
• 9. a3. Allows Black to play 9...Bxf3 10. Qxf3 Bxd2+ and you are forced to retake with your king and strand it in the middle of the board behind a less than optimal pawn structure. Lucky that he played it the other way around.
• Moves 10 & 11 are obvious.
• 12. O-O-O. Not the best choice, in my opinion. Your kingside might look just as weak as the queenside, but the doubled pawn actually offers some interesting possibilities, too. Many pieces have been exchanged already, so you don't actually have to worry about your king's safety after castling short: Bg2 fills the gap nicely (and can exploit the long diagonal where Black is actually weak now that his bishop is gone) and Black cannot bring in more pieces other than the queen anytime soon. Your f-pawn can support e5 from f4 to keep the center closed so that this stays this way and it may even attack on f5 in the future. You, on the other hand, might be able to use the half-open g-file to threaten his king. Black's counter plan will be to open files for his rooks. As this will happen on the queenside or center, castling long happens to have moved your king into future lines of fire. h1 would have been a much safer place to hide.
• 13. h3. Kinda unnecessary at this point in time, but it's alright, you want to make the rook more mobile.
• 14. exd6. Obviously, following my commentary on move 12, I don't like this move either. En passant looks fancy and all, but I think your strategy at this point should be to keep the center closed and use your (small, but hey) advantage on the kingside before he can break through on the queenside.
• Rest of the game, in short, as it doesn't really concern the opening anymore: After opening the center, his rooks can join the fight, and your inferior pawn structure quickly shines through. Castling long proves to be absolutely inefficient in terms of king safety because of said whole-y structure.

Answer 2

The game was totally correct but after the move 22..Qxf3 you could play 23.Rxd5 your f2 pawn is defended and opponents one pawn will down , and you could control the d file for that time which is very good . The opening was reasonable , but i think you need to prepare middlegame. For advice you can see the games of karpov and carlsen. sites : http://www.chessgames.com/perl/chesscollection?cid=1011420 ; http://www.chessgames.com/perl/chesscollection?cid=1034176 etc.