Scandinavian and won opposite-bishops endgame

As usual, plugging a game I feel I played well into the engine is humbling. However, it's easier to understand human commentary.

Game: lichess.org/L3u1EU3g

I shamefully missed a mate in two. Despite that, I felt like I made some good strategic decisions, and I'm proud of the endgame I played - would appreciate analysis about whether simplifying to opposite-colored bishops, when mine was good and his was bad and I had an extra outside passed pawn, was a good idea.

Opening seems ok, I don't like 12.Nh4 & 13.Nxg6 healing your pawn structure. Then you are getting the initative (remember opposite colored bishop favor the attacker!) and you should have increased the pressure on f2 with 23...Qf6 & 24...Bc5 but after exchanging queens and missing 27...Rxd1+ the game should be in a draw range again.
33...Ra1 is one step further to a draw and I don't understand why your opponent gave up. His king can go to b3 and you should not get your pawns into motion with the white bishop on the a2-g8 diagonal, no matter it's 3 vs. 1 pawns.

Hi Mollus, thanks for your comments. All very helpful.

I played around a bit from the final position with the engine, and it does appear to be a win for black. The technique is pretty tricky, and i wouldn't have figured it out on my own. I can describe it if you want, or you can try to work it out for yourself.

I made quickly a study on the "classical" Scandinavian with some key ideas. It's not the end of the world for Black but I would try to bash it (and have done so!) that way!

Happy bashing!

Interesting study. However, most Scandinavians play 3...Qd6 or 3...Qd8 nowadays. This excludes the tactics associated with Bd2. I guess that even 3...Qe5+ or 3...Qd7 may be playable.

I don't know much about openings, the final position is more interesting for me. It is much deeper as I thought at my first sight but I am still not convinced how black should win this. Don't be fooled by the engine evaluation, anything better than -3 should still be a draw here. @the_breadler
My suggested line is:
49.Ke4 Bxc5 50.Kd3 Kg7 51.Ba2!!

[Not the engine move 51.Be6 which allows black to make an important move for free: 51.Be6 Kf6 52.Bg8 Bf8!! 53.Kc3 Ke5 54.Ba2 b5 and white is in trouble: the black king on d4 in combination with c5 is lethal.]

But what to do after 51.Ba2 - winning the pawn immediately?
51...Kh6 52.Kc3 Kxh5 53.Be6!! - it is essential to put pressure on these pawns! 53...Be3 (after 53...Kg5 54.Kb3!! -the white king has to secure a2- and NOT the direct 54.Bc8 because of 54...Bb4+ and black saves his pawns) 54.Kb4!! Bc1 55.Kc5 Kg5 56.Kb6 Kf6 57.Bg8 Be3+ 58.Kxb7 - draw

- activate the king? 51...Kf6 52.Kc4 Bf8 53.Kb3 Kg5 (53...Ke5 54.h6) 54.Ka4 Kxh5 55.Be6 (not 55.Ka5 Bc5 fail) b5+ 56.Kb3 b4 57.Bd7 c5 58.Be6 Kg5 59.Kc2 Kf4 60.Kd3 Ke5 61.Bc4 and now 61...Bd6 has a evaluation of -10.2 (depth 24) - after playing Bd6 it went down to -1.9

Also don't be fooled by my analysis, I'm not a 100% sure but would be happy if someone would correct possible errors!

Thanks.

Sarg0n, I hadn't seen the idea to skip Nf3 before. I guess basically that is gaining a tempo so that black can't play Bb4 later.

Mollus, I'm going to work through your analysis but I might need a couple days to get there... You are correct, I was playing black and winning against the engine move 51. Be6.

If white refrains from Nf3, then black should play Bg4 nevertheless as did Larsen in his classic win over Karpov

Who knows... Bg4 is not everyones taste and for example after f3 (probably best) there are even more critical lines, for example

1. e4 d5 2. exd5 Qxd5 3. Nc3 Qa5 4. d4 Nf6 5. Bd2 Bg4 6. f3 Bf5 7. Bc4 c6 (better Qb6!?) 8. g4 Bg6 9. f4 e6

has a score of >80% in my db.

Yes: who knows? 6...Bh5 is also possible but leads to transposition in case of 7 g4 Bg6 8 Bc4.
It seems good for white indeed.