The first way that Bag of Holding works better is that it can trigger itself. Scry is an even narrower theme to build around than discard, so I really should have put a Treasure Map-type activated ability on here. Second, the Bag puts less of the mana payment up front and more into the card-advantage activation, which improves the ability's balance and helps avoid running out of exiled cards. Together, these differences mean that Bag of Holding is a less swingy card than Vault of the Dross-- less useless when it's bad and less oppressive when it's good. If I had a fresh crack at this design, I'd probably cost it at 2 mana with 1, T to scry 1 and 2, T to get back an exiled card.
With a single cascade card with CMC 3 this would let you mill out and win with Laboratory Maniac style cards AND tutor for that card. Maybe limit it to the first ~3 cards put on the bottom of the deck each turn?
ReplyDeleteHmm, good point, I'd forgotten how open-ended some of those "bottom of library" effects were. A 2-card combo that requires ~7 mana on 1 turn (3 for the cascade spell, 1 to get back LabMan, 3 to cast it) doesn't seem too scary compared to Living End, but it's worth keeping in mind (and might actually be a good reason to change the costs around as I described).
DeleteAnother option could be to only let you exile cards that are put there in a non-random order. Might lead to a strange wording though, especially to make it clear that a single card can be put there in a random order.
Delete