I just lost to a player 400 points above me. The thing that gets me is that I found this nice "tactics trainer" type move to win a pawn plus the exchange in the opening. I was so excited to pull this stunt on a higher player! But somehow I managed to give all my gains back, and then walked into oncoming traffic in the back end of the game. What could I have done better in the first part so as not to give it all back to him?

Nice initial tactic! I haven't analyzed in depth, but Qe2 does look to be the first mistake. Why not Qh5+ provoking another pawn weakness followed Qd1 ... or just Qg3? I remember seeing something (an article maybe?) about common errors when one takes a big lead. One of the major points was that players in the lead tend to overrreact by laying very cautiously or by thinking, I'm up, now I can relax and the game will win itself.

14.Qf3 was a panic move.  I've learned from what little otb tournament experience I have as a 1520 ish player that you always need to be as objective as possible if you're up or down unless you want to crumble under pressure or blow a won position.  I don't see much wrong with Qxa7 as the center is closed and white is two moves from castling with Bd2 0-0.  You can even afford to pitch back a pawn or two if you feel under fire in the future, but the center is closed and your queen is not getting trapped.

I agree with RyanM, 14. Qxa7 is better than 14.Qf3. A pawn is a pawn. But what about more fireworks: have you considered 14. Rxa7?!

Black jumps with joy, plays 14. -Bc6. And then 15. Rxe7+ with these possibilities: 15. - Qxe7 16. Qxb7+ or

15. - Kxe7 16. Bg5+ Ke8 17. Qxc6+ Nxc6 18. Bxd8 (not quite good enough, but the threat of Bg5+ remains). Better 16. Qa7+, with possibly - 16. -Nd7 17. Bg5+ (now you get his Queen, the knight is pinned!

lol, I got the same idea, but in a serious game I would probably be wise by Qxa7.

14.Qf3 isn't great, true, but 15.Qe2 ? is really painful to watch. 15.Qg3 is way better : the queen stay active, attacks g7, doesn't block the bishop, is on a safe square and also defends f2 - everything is better here.

16.f3 ? : I wonder against what you are defending. 16.Be3 develops. The best is maybe 16.Qg4, but after you moved the queen to e2, you can't find it OTB for psychological reasons.

22.Qe3 is advantageously replaced by g4 (or g3 followed by g4) (it's too late to develop the Bf1).

23.Kd1 ? lets the enemy soldiers invade the 8th rank, while Kb2 ! followed by Ra3/a4 and Rha1 later started the counterattack. In these lines, you need not to push g3 and then g4 because the black dark bishop will have to transfer to the attack of a3 anyways (no need to give him a tempo to do this).

27.Rxa4 killed an enemy pawn (the zwischenzug Ra8+ kills any discovery tricks with the knight).

After 34.Ke1 (?) there seems to be no hope left. 34.Qa8+ followed by Q or Rxa4 (depending on Black's answer) was a good hope to draw, based on the poor placement of Black's rook.

Please be relevant, helpful & nice!

Irontiger - these comments are very sensible, I think. Metastable, there is a hopeful message here: After 15. Qe2 you felt hemmed in, the game drifting away from you. But even as late as move 34, there was hope for a draw! So don't be intimidated, and if it looks like a sacrifice, it sometimes really is just a pawn up for grabs.

Irontiger and all, thanks for the comments. I recall seeing some of the ideas you mentioned during the game, but I guess I didn't look deep enough or think hard enough in the lines enough to realize their true value. I like how you put 15.Qe2: "painful to watch" - that's pretty accurate :-)