Only 34 games in the Db with 2.Bc4 with a win ratio for White of only +29% -53% =18%. Surprisingly (or maybe un- considering how dubious it is), the first try was as late as 1988 with a White win: J Young vs R Huntington, 1988

4.Qh5 is the Schaeffer Gambit with roughly the same percentages. 5...Bg7 is the only game: other ones have 5...Nf6.

It appears Simon was honour bound to play the Hillbilly and walked into a prepared line.
6...a5 was a TN played after a couple of minutes. Simon had to slip in an odd move to to get Gawain out of his prep. Having made this OTB decision odd moves will normally require more odd moves.

Gawain too dallied in the 'odd moves' stakes. 8 pawn moves in the first 10 moves is not standard in a Black Caro Kann. Eccentric moves often have to meet eccentric replies.

The game is noted up by Richard Palliser in CHESS December 2011. who writes; <"10 g4? And this is truly disgusting."> A bit harsh, but coming from a fellow GM rather than a C.G. punter this reads OK and Richard never pussy-foots about in his notes.

Being a non-GM and always keen to offer a reason for so called poor moves. I'd have gone down the path. "Two choices, either get crushed like a bug in a longer no hope game or a dogs dinner game where it either works or the end is mercifully quick." Simon went 'dogs dinner.'

Richard's reports starts;

1 e4 c6

"No Dragon today! The Caro strikes me as a good choice against the highly–aggressive and ever–dangerous Williams, and Gawain was actually hoping for White’s next move... 2 Bc4!? The so–called ‘Hillbilly Attack’ on which Simon has previously written an article.

It’s not without some punch, but whether it was really such a good choice against 2600+ opposition I’m somewhat less sure. "