Immediate aftermath analysis.

This sounds like it's a whole thing that had an introduction elsewhere, but as it's posted here I'm going to assume you want comments.

I don't know the Closed Sicilian at all but I have to comment on 8.b3 because I don't understand this. Assuming you want to play d4 (not 100% I would) if the play had instead gone 8.d4 c4 isn't the last thing White wants to play something like bxc4 bxc4 and open up the b file? That's where Black has the space and the pieces. It seems to me that b3 spends a tempo to do little more than blunt your LSB.

I don't understand any of your endgame comments I'm afraid. You talk about your position not being as good as you thought; you talk about 24...Nc4 being this this huge move; you talk about winning the Black DSB and the pinning f7 pawn and I don't understand any of it. I see a nice endgame where you're simply a pawn up and with good winning chances. I see no threats to ever win the Black DSB, I see no significance of the f7 pawn being pinned and I see 25.Bxc4 as leaving you with a very nice position.

Thanks for the post. Interesting game. Pretty tactical 

A few comments:

You mentioned after 1. e4 c5 2. Nc3 that you intended to dominate the d5 square. You have correctly identified THE most critical square in the Sicilian... d5. But in general terms the move Bc4 isn't really a great way to put pressure on it. (I say in general terms because there are a few specific lines such as the Sozin and the Velimirovic in which White's light-squared Bishop is indeed well-placed on c4 or b3). The Bishop is typically much too exposed on that square, and gets kicked around by Black... often ending up in an undignified position. The fact that you ended up with the Bishop buried alive on a2, blocked by its own Pawn at b3, should be more convincing than any argument I could offer.

If you want that Bishop to pressure the d5 square in the closed Sicilian, then after 1. e4 c5 2. Nc3 d6 (or Nc6, or whatever), play 3. g3 and 4. Bg2. The Bishop might LOOK less active on g2, but it will be far less exposed there so it can actually DO ITS JOB of pressuring d5, instead of running and hiding like a little girl.

I also didn't like the Queen trade. A large part of your advantage was in having more active pieces. Why trade your active Queen for his passive one? Either Qg5 or Qe5 looked more to the point.

16. b4 seems to win a piece.  B3 was a bad positional move as noted above, so make sure you understand that and ask questions if you don't, but you lost because of tactics.

Wouldn't think so. 16. b4 Ne6.

A tough loss but a good game. At 19...0-0 White looks better with a pawn up and Qside majority. But the B on a2 is kind of awkward. White's aim is to probably exchange as many pieces as possible and sooner or later get a passed pawn. Also get the King toward the center then qside but safely. 

I'm not sure about your plan afterward, maybe 21. Rad1 was the wrong rook. 21. Red1 and then the exchanges leaves your Rook on a1 protecting the B on a2 and a3 if B moves. Then c3 and Bd4 closes the file and maybe increases the chances of exchanges and a passed qside pawn. Though probably want to keep in mind opposite color bishops might likely draw even a pawn up. Kf1 then seems natural. 

Maybe that plan's not enough to win but whatever the right plan was I'd guess that 19...0-0 was a key moment to try to come up with one. 

Why are you having trouble with 2.c4? Most people think it is a patzer move but it is an interesting one. You will either get Maroczies or after 2...e5 (only way to stop you from playing 3.d4) a weird Botvinnik setup

15 Qxd5 is an automatic move, but clearly not the best. White is totally winning here.

E.g. 15 exd6 Qxd6 16 b4 Ne6 17 Bxd5 Rd8 18 Bxe6 fxe6 19 Bb2 pawn up with overwhelming position and Black can't castle...

There are some other strong 15th moves  too, like immediate 15 b4.