I'm continuing to tinker around with the format of the articles as I get more to grips with the pros and cons of PGN/HTML editing. The next "big" update will be on the Albin Counter-Gambit, which I've been using frequently with Black over the past couple of years, with reasonable results and some exciting games.
However, the latest update to my site has been on the Englund Gambit (1.d4 e5). This is Black's only way to get a gambit on the board by force after 1.d4, but the downside is that it is not as reliable as the Albin or Budapest Gambits because White has not weakened the b4 and d4-squares with c2-c4.
I played the Englund Gambit with 1.d4 e5 2.dxe5 Nc6 3.Nf3 Qe7 quite a lot in casual games between 2004 and 2008, but have largely abandoned it since, partly because of the strength of 8.Nd5 in the main line and partly because I began to find the Albin and related lines (1.d4 d5 2.Nf3 Nc6!?) more attractive as well as being sounder. The line is still playable at club level in my opinion, since there is only one particularly troublesome line for Black and in practice, White is unlikely to both get that far and know what he/she is doing, but it does arguably amount to "hope chess" in the sense that you're relying on your opponent not being booked up on the refutation. I tried it out in one serious match game, which I won, though ironically it was me who ended up the pawn ahead and facing an attack:
Having looked through the different lines of the Englund Gambit, perhaps the best practical tries for Black after 1.d4 e5 are the Zilbermints Gambit (2.dxe5 Nc6 3.Nf3 Nge7) and Soller Gambit (3...f6), which are the lines that Stefan Bücker examined in Kaissiber 5 in some detail. They concede a greater-than-usual advantage to White with best play but I don't know of an outright refutation. That said, my occasional experiments with the 3...Nge7 variation in casual games have typically been unconvincing, and I typically felt that I was playing an inferior version of the Albin Counter-Gambit lines with ...Nge7. 3...f6 should not give Black enough for the pawn, since Black is playing a Blackmar-Diemer Gambit a tempo down, but a scan through the databases suggests that in rapid games, the BDG can be dangerous even after losing a tempo. In slow games I have rather more doubts about 3...f6, however.
If White does not play 2.dxe5 and 3.Nf3 then White gets no more than a normal edge, and sometimes less (e.g. after 2.d5?! White is the one struggling to equalise). One area where I differ from established theory is that I believe that 1.d4 e5 2.dxe5 Nc6 3.f4?! should be met by 3...d6!, rather than 3...f6?! 4.Nf3! fxe5 5.fxe5 which is probably as good for White as 3.Nf3.
I have to say, though, that if I was to take White against someone like Lev Zilbermints (I've seen quite a few of his short, sharp kingside attacks in the Soller Gambit in particular) then I would be strongly tempted by 1.d4 e5 2.e4!?, where it's often White who gets to be the gambiteer, making it a good psychological tactic.