Also I don't agree that hellbent has a lot of problems. Binary threshold mechanics are much easier to balance than ones that scale. Almost every broken mechanic scales - storm, affinity, and delve.
Hellbent being a threshold mechanic really limits what you can and can't do in terms of play patterns. In particular, Hellbent rewards you for emptying your hand quickly and then just keeping it that way. Also hard to balance; the difference between on and off is typically so large that you don't have very many design decisions that make sense: see Demon's Jester, Gobhobbler Rats, Hellhole Bouncer... basically most of the cards with Hellbent.
None of that makes sense, but I'm rather tired right now so that's the best I can muster.
Also, Channel + Emrakul is ostensibly a turn 2 kill. The issue is probably the enabler, not the creature? Thus Berserk being a problem (it's not going to be in my cube). But yea, I haven't really balanced the PT that well, but I also haven't tested it; it's all just theory atm.
Also, want to keep costs as low as possible but colour requirements high as I want these to function as madness enablers as well.
Furthermore, I assume most peoples' cubes are littered with really strong 1-for-1 removal (Swords, Paths, Dismembers, whatever) so I wanted to push the toughness a little to compensate for the inherent card disadvantage these imply. Basically don't want to remake Avatar of Discord in Rav/Kami standard.