Originally Posted by
JamieW89
Both options won against D&T, but after the match we agreed that Empty was perhaps a bit safer, unless you expect multiple sweepers. If they keep StP in they can beat it with 2nd land, mystic, plow and any follow-up creature. Maybe a fear of something like TiTi would let them keep some StP's, which would probably make the PiF line better. However, everyone knows I really like Empty (I usually run it main, used it against him in every game during our last match), so I went with PiF.
I should have given more information regarding the second example. The Miracles player had no mana open with just 3 lands in play. He also appeared to have no lands in hand. The top three of his library appeared to consist of 3's, 5's, and 6's as my spells (including LED,Rit,Cabal Rit, and PiF) either resolved or met spell-based interaction despite the lock on the table (I am not completely sure whether he ever flipped top or not.. We'll assume he never did - otherwise all the assumptions might fizzle :P).
As such the odds for the Tendrils kill are just those of dodging the 3 remaining Forces in 8 cards (5 unknown hand cards and top 3). The odds for the Empty line mainly consist of dodging 2 Brainstorms in 5 hand cards and dodging 3 Terminus in the top 4 (from which the top-3 seemed full of high cost spells). Snapcaster Mage would not have done it with just three lands and one turn.
Given these assumptions the chance of dodging a Force was about (32/35)*(31/34)*(30/33)*(29/32)*(28/31)*(7/10)*(6/9)*(5/8) which is 18.1%.
The chance of dodging 2 brainstorms was about (33/35)*(32/34)*(31/33)*(30/32)*(29/31) = 73.1%
The chance of dodging 3 Terminus in the top-3 was about (7/10)*(6/9)*(5/8)*(43/45) = 27.9%
The combined odds for the Goblins line are 20.4%
Assumptions:
1) None of 20 lands, 1 CB lock, and 4 used spells in his hand. (Actually 2 of the Terminus are unlikely to be in his hand, but didn't account for that)
2) Only 3/5/6 cmc cards in the top-3.
3) 3 Mentor, 3 FoW, 3 Terminus, 1 CJ as unplaced 3/5/6 cmc cards.
Given also that there are more hidden odds that beat the Goblins line (e.g. a Ponder kept in hand due to mana limitations or potential fow pitch could further increase the chance to find a Terminus, or playing SCM->BS to hit a land as 4th card if the top-3 does not have Terminus) the two lines are quite close.
I went with the Goblins, felt like it was wrong after, but if this math is any decent (not the greatest likely, but it's a complex situation and we need some dodgy assumptions) then it didn't really matter much.
I cannot find the Brainstorm spoiler, but in this spot I'd probably want to take our free kills if they exist (Infect isn't that easy, we're on the draw, our hand can work out badly if we just Probe first, Daze being annoying on our Brainstorm is still an issue turn-2, and they see a ton more cards potentially as well as having access to cards like Flusterstorm if they did have them).
If our first decision is to actually cast the Brainstorm on turn one, the key theme becomes the sequencing of spells versus Daze. We can lead with (1) Volc, LED, LED, Brainstorm, which minimizes the chance of a LED getting dazed, but begs the Brainstorm to get dazed, or with (2) Volc, Brainstorm, LED, LED when they will clearly daze the 2nd LED. It does give you more flexibility though as you could find things like Petal, discard+land etc. Option (3) is playing Volc, LED, LED, then putting Probe on the stack, and responding by Brainstorming, so that you can sacrifice your LEDs to pay for a potential daze. This seems quite weak, as it needs to hit something like Infernal + Dark Rit/Cabal Rit if they do have Daze. It also will not let you reevaluate your hand after the brainstorm if you'd want to.
I guess I'd just play the Volcanic and Brainstorm, intending to follow it up with two LEDs, and probing into AdN revealing the best card we have if they do nothing. But the cards we draw from the Brainstorm can change that (as will any Daze ofcourse).
What did/would you do?