I agree with Mulligan feature, but some seem to be saying it would allow for more strategic deck builds.
If anything, it would only keep the status quo and even things out. It would make bad deck builds (number of cards wise) able to compete with better deck builds consisting of the same cards. On the flip side, it would let good deck builders take into account the mulligan and include less of a certain card needed only in early game.
I have more to say, and from now on I'll use protect artifact in a rainbow deck as an example.
Against seism, you would be able half the number of PAs would have in a deck and still have the same chance at drawing it. This is a problem for me; the whole idea is to decide between possibly hurting yourself against other gods while giving you a better chance against seism. Not the best of both worlds.
The probabilities of drawing your protect artifact against seism, assuming you only have 1, is about 8/40 (assuming 40 card rainbow). Not exactly, since each card increases the chance slightly (1/40, 1/39, 1/38, etc), but basically it is 8/40. Which is equal to 1/5. So about one out of every five draws should include PA.
Now, with mulligan, you in effect have two chances at that 1/5 chance, (ie two draws per), equating to a 2/5 chance (not taking into account one less card). That is a DRASTIC difference. I think that gods should be more about the deck build and calculated risks, not about increasing your win percentage with mulligan.
Basically, I would like to see mulligan as an option for pvp and maybe other Ai, but not be available for gods.