So here's the deck I've been finding myself using most often for FG farming:
6rn 6rn 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6u2 6u2 6u2 6u2 6u2 6u2 6u7 6u7 6u7 6u7 6u7 6u7 6ve 6ve 6ve 7t8 7te 7te 7te 7te 7te 7te
Pretty straightforward extension of the classic unupped liquid antimatter, but the addition of chaos power is really transformative since it gives you bigger vampires that stay alive longer; a lot of gods that are nearly impossible with unupped liquid antimatter are very doable when you can power up their creatures for antimattering (Octane becomes downright easy.)
The overall win rate is still pretty mediocre (~35-40% maybe) but games don't take too long and you can simply skip 40% or so of the gods that come up and win against the rest 70% of the time with a decent EM rate, so you earn money at a decent clip. I still wind up using RoL-Hope and dune scorp decks now and again for variety's sake, and honestly they're probably better, but I seem to have the most fun with this one, and it makes a decent complement that's good against some of the gods that give other options difficulty (particularly for RoL-Hope, since it's vulnerable to CC decks and antimatter bypasses most of the AI's creature control entirely.)
I'm pretty happy with where this is at. It's very lean on pillars, but this actually seems to be ideal with the way the deck plays; you're usually saving cards and not looking to play them as fast as you can since you need to build up combos of 2-4+ cards and wait for good opportunities to play them. I find I usually have juuuuuuust enough quanta to do what I need to (another area where upgrades transform this deck; this pillar distribution would be suicidal if you didn't have an upgraded deck.) There's a lot of versatility though as both darkness and death can bring a lot to the table here--steals, mutations, etc. I've tried various tweaks without much success; it seems like it may be possible to scrape by with -1 antimatter or liquid shadow (you rarely need all 6, thanks to the extra oomph from the chaos powers) but in practice this doesn't seem to work too well. Has anyone else done any experimentation along these lines before?