I illustrated earlier in the thread that the current model of running 7-10 counters and snapcaster mages aren't good for the deck.
It is
not a different deck to build this as a sligh deck, which it very much
should be. Notice the what, 1 month period this deck was a DTB? Because people weren't prepared for it at all and for a quick minute people were wrecking with it.
Except people then began trying to add more counter-magic to it, and cards that intend to go into the late game.
Put flatly, snapcaster mage in this deck is terrible. 3 mana burns or brainstorms aren't good, they're bad, and there's no actual reason to run him. He's done cool things for me, but ironically, it's because he was flashing back bolts.
Zilla probably has the best concept of the deck: all burn goes to their face, swing with your creatures until they're dead, only brainstorm/ponder for gas. And that's it. Daze is questionable at best and only really worth it around turn 2 or 3 when you already have sufficient lands down to keep playing things through it, but T1 daze'ing is awful for this deck.
This is the list I would / probably will be playing for UR Delver:
//Creatures: 12
4x
Goblin Guide4x
Delver of Secrets4x
Vexing Devil//Instants: 13
4x
Brainstorm3x
Daze4x
Lightning Bolt2x
Thunderous Wrath//Sorceries: 17
4x
Rift Bolt4x
Chain Lightning2x
Price of Progress2x
Fireblast3x
Ponder//Lands: 18
I'd be inclined to drop ponder for lava spikes in some number, but the card filtering will help set up turn 3 Thunderous Wrath bombs and I'm a pretty big fan of it in general.