Annotated chess game

Brings to mind the old epigram by William Ewart Napier: "In the laboratory the gambits all test unfavorably; but all gambits are sound over the board."

Yes, and I was once taken off guard by that and, inadvertently allowed black to capture on e4. I was very upset with myself afterwards because I was lost, as things kept going downhill from there. In other words, I had no way back.

4. Bd3 is also played here. In this position however, I have a fondness for a gambit-y sideline in 4. Ne2!?, daring black to take on e4. This was an old weapon of Alekhine and Bronstein who had some spirited games with it.

I would've answered with 4.exd5 here. The only time that I'd play 4.e5 would be if black had played ...Nf6 earlier.

Interesting! I have a particular fondness for 5...Ba5 in this position, and though I've had some success with it (and interesting games), against top opposition I've crashed and burned. I've been forced to abandon one of my favourite lines.  

7. Qg4 seems good here, activating the queen and prompting a weakening ...g6. If black takes on b2 your QB is on the long a1-h8 diagonal and solidifying defense on e5. Nf3, Bd3 and O-O seem to naturally be in the works. With the 2 bishops and ownership of the half-open a-file compensation for the pawn seems adequate.

Possibly the reasoning for delaying the recapture is that, for the pawn (if Black takes on b2), White gets an easy, well developed game. At that, Black has to be wary, lest this disaster overtake him: 7.Nf3 cxb2 8.Bxb2 Ne7 9.Bd3 0-0(?? - 'castling into the attack') 10.Bxh7ch Kxh7(?!) 11.Ng5ch Kg6 - retreat simply leads to checkmate - 12.Qg4 f5 13.exf6e.p. e5 14.Ne6ch Kxe6 15.Bxe5ch!! ... (15.Nxd8? loses) 15...Kxe5 16.Qe2ch ... and picks up Black's queen. Although Black will have three minor pieces for it, Black's overall unfavourable position should give White the win. The dear old 'Greek Gift' sacrifice. Always fresh, to my mind - not so much the motif itself, as where it may crop up.

I agree. It's an interesting setup.

8.Nf3 was played by New Zealand GM Murray Chandler against Australia's Ian Rogers in 1986. White won.

And, I would've thought that this scored better. The reason that I'd refrain from 8.f4 that it advances the weakest pawn on the board. And, I would assume that the K would be close to exposure.

Going pawn-grabbing at the expense of development and/or initiative in the opening should have a Surgeon's General Warning label attached to it. I've never been comfortable being the "grabber."

I have been, but I learned that the reason must be very good. Once or twice, pawn grabbing at the expense of development costed me the game.