Most of my time before GP Charlotte was spent trying to decide between playing Grixis or Twin. Both decks had a lot of cards that I like to play—Vendilion Clique, Snapcaster, Cryptic Command—but the difference was that one had the “oops I win” component and the other did not. In return for that potential, you had to play cards that aren’t even very good in Limited, such as Deceiver Exarch and Splinter Twin.

Historically I’ve always disliked Twin, which is weird because looking at its cards you’d think it should be the opposite. I think part of that has been the somewhat flawed Modern testing on TeamCFB, at least where this deck is concerned. Twin’s bad matchups—black/green in all its forms—are always the most popular decks at Modern PTs, so they are always the decks we test against first. When Twin naturally underperforms against these, we sort of dismiss it, and never get to the point where we play it against the decks it beats—which are not very widely represented but, on the aggregate, are a big portion of the metagame.

The way I saw it, playing Grixis would make my matchup better against Jund/Junk and against the mirror. Playing Twin would make my matchup better against Burn, Amulet, Tron, Affinity, and random combo decks. The main difference, however, was that playing Twin would completely swing matchups. Tron would go from being a very bad matchup to a very good one, whereas Grixis would turn a 40% into a 60%. Besides that, the Twin deck could realistically morph into a bad Grixis if that was necessary, but the Grixis deck would have a really hard time morphing into Twin (though now that I type this I think that maybe this is not the worst idea).

In the end, I settled for Twin because it seemed like it would improve my bad matchups more. And also because I liked what people did to their decks after board to try to beat my combo. This is the list I played:

Grixis Twin

The big difference between this list and most other Twin decks is that it’s a dedicated UR Twin deck game 1, but morphs into Grixis Twin post-board. This is not usual. Most of the time, dedicated lists are UR and remain that way. If someone plays Grixis Twin, then they usually have some removal and Tasigurs main. The reason I decided to play the list I played was that I thought being dedicated to the combo game 1 was better, and you didn’t need another kill condition, but at the same time I didn’t like the sideboard plan for UR when you needed to “morph.” In most blue matches, including the mirror, having a cheap creature is very important so that you can apply pressure and then force them to react, and there isn’t much of that if you’re straight UR. When I asked Sam Pardee what his plan for the mirror was and he replied with “board in Roast to kill their Tasigur,” it was clear which side of the exchange I wanted to be on.

The tournament went relatively well—I went 11-4 for 1 Pro-point. My wins were against 2 Burn decks, 2 Affinity decks, Infect, Storm, Jund, and Zoo. My losses were to Tron, Amulet, Junk, and Jund. Junk and Jund are bad matchups and I lost in close games, while Tron and Amulet are good matchups and I think I got a little unlucky to lose to them (Amulet killed me on turn two on the play, for example, when my hand was 3 lands, Remand, Blood Moon, and the combo).

I think I played well enough—I did not make any great plays or any horrible mistakes (which is actually unusual as far as I am concerned. I usually swing between the genius-horrible column rather than staying mostly in the middle). I think a lot of the nuances I missed were due to being inexperienced with the deck, which is a big thing in Modern—you can see that the people who play Modern with the same deck usually do well with a greater degree of consistency than those who keep switching decks, like me. The moments I felt I lacked the most were those in which I think I should have exploited the fear of the combo, but I didn’t. There were situations in which I could have played an untapped land turn three but decided to play a tapped land instead, forgetting that passing the turn with three untapped mana would put the fear of God into them even though I didn’t have any piece of the combo in my hand or even in my deck. When you play Twin, this is a fear you have to exploit, and I’m sure I’ll do better in this regard if I play the deck again.

Some of the more unusual card choices:

1 Thought Scour: Well, I had one slot, it works with Snapcaster Mage, and it works with Tasigurs in the sideboard. A lot of people play 1 Peek, but I think if you have Tasigurs in your 75 then you should play Scour.

2 Spell Snare, 2 Dispel: I like both of these better than Spell Pierce. Spell Pierce is better in some situations, of course, but Dispel is the best to protect the combo and Spell Snare is the best to stall the game. Once you morph into the control plan, then having Spell Snare in your deck is very good, as that card is very powerful in this format—it counters Goyf, Bob, Spellskite, Snapcaster, Ravager, Plating, Summer Bloom, Roast (!), Terminate. And with Collected Company gaining ground in Modern, Dispel is very rarely a dead card.

2 Kolaghan’s Command : A couple of weeks ago, I asked on Twitter why everybody was playing Kolaghan’s Command. That was before I cast Snapcaster, targeted Command, blocked with Snapcaster, and then flashbacked it bringing it back. I’m a believer now.

I think Kolaghan’s Command does a lot for you. It’s a grindy card that is good in your grindy matchups post-board (Jund/Junk/mirror) and it kills Spellskite/Shackles, two of your big problems. More importantly, it doubles as hate against Affinity and it’s an answer to Torpor Orb, which you might need because you don’t have Ancient Grudge anymore.

1 Anger of the Gods, 1 Pyroclasm: I’m sure one is better than the other, but I don’t know which one. Anger is better versus the Collected Company decks and Junk with Voice of Resurgence. Pyroclasm is better versus Affinity, Burn, and Infect. I think that if I had to do it again, I’d choose two Pyroclasms.

2 Dismember: I’m a big fan of Dismember, though it’s probably worse than it has ever been right now because it doesn’t kill Primeval Titan or Tarmogoyf in the late game. The upside, of course, is that it only costs 1 (so it’s easy to use in the mirror), it doesn’t require black mana, it kills Spellskite (unlike Ultimate Price) and it doesn’t cost BR like Terminate, which is very awkward in a deck that is base-blue and wants to fetch for Island on turn one a lot of the time. In the end, they all have pluses and minuses, and it’s up to you to decide which removal you want, though I do believe you want an instant-speed card that kills big things.

Sideboarding

Sideboarding with this deck is VERY tricky. I honestly think I would be doing you a disservice if I gave you a complete sideboard guide, because it changes so much depending on what your opponent has and what she plays. I don’t even know how I’d sideboard for any match before I see what kind of player I’m playing against—so if I were to give you straight numbers, I would have to make them up. A general guideline is the best I can do.

  • Against the black/green decks, it’s better to take out a big portion of the combo. Against Junk, I usually leave in either one or two Splinter Twins, based on how they play (if they play afraid of it game two, then I can take the second one out for game three). Against Jund, I leave in one or zero (they have access to Rending Volley, so the combo is generally worse). Against Junk I board in one Blood Moon, against Jund I board in zero. If in game two they aggressively search for basics, then I take out all Blood Moons. If they only search for duals, I could bring in two. I also bring in Keranos, Tasigurs, and Commands (but no Thoughtseizes). If they are playing Junk and have Souls and Voice, I bring in Anger of the Gods. I usually take out some number of Twins and two/three Deceiver Exarchs (Pestermite is better because it threatens Liliana more), as well as Dispels most of the time.
  • Against decks that race you, I keep all of the combo in. Tron, Amulet, Storm, Affinity, Burn, Infect. Against all of those I keep the full 10 cards. Against Burn I also bring in a couple of Tasigurs. Against the other combo decks, Thoughtseize is great.
  • Against blue decks, I take out part of the combo. I don’t love Keranos versus other Twin decks, but Tasigur is very good, as is Thoughtseize. I don’t like taking out all of the combo because it keeps them honest—if they try to tap out for Keranos and a counter war, for example, you can just kill them.
  • Remand is a good candidate to come out on the draw, but it’s usually good against almost everyone on the play. It’s never great versus Affinity or versus Burn, but I still keep some on the play versus Burn.
  • Spellskite is good against almost everyone—Burn, Twin, decks with removal, Amulet. I think I would like two in the future.

General pointers:

  • If you are unsure on whether you want a card with Serum Visions or not, you can keep it and play a fetchland or cast a Thought Scour. This will give you more time to decide if you want to draw the card or not.
  • If you’re going to play Serum Visions, make sure you fetch first—you don’t want to have to shuffle away your scry cards.
  • Always sideboard a little, even if you aren’t changing anything in your deck (take out a card and bring in the same card). It’s important to keep them guessing—if they see combo game 2, you don’t want them to be sure that you still have the combo game 3.
  • You choose tap or untap with Pestermite when the ability resolves. This is useful if they have Vedalken Shackles and only two mana—you can then target it with Pestermite and wait for them. If they don’t do anything, you tap Shackles. If they tap it to steal Pestermite, you untap it when the trigger resolves and get it back. This doesn’t work with Deceiver Exarch or if they have four mana (because then they just use Shackles again) but it’s an interaction that almost nobody knows.
  • If you target something with Snapcaster, you don’t need to play it immediately. Targeting a Bolt or a Dismember midcombat might end up dissuading them from playing another creature. This is especially good with Kolaghan’s Command, so that you can return the Snapcaster again after it dies.
  • Don’t forget that you can Remand your own spells if they get countered.
  • After you have Tasigur, it becomes much easier to kill Tarmogoyf. You can Bolt it and then play Tasigur to shrink it to 2/3 and kill it. You can also do this game 1 with Snapcaster Mage.
  • Most people will kill your Exarchs at the end of the turn if you play them and tap a land. Almost every opponent of mine did that—they never waited for me to untap if I had access to more mana because they were afraid I’d topdeck a way to protect my guy. You can use this to draw removal spells out of their hand.
  • Playing Splinter Twin on Snapcaster Mage is very legit. You can even do it on Vendilion Clique to get more triggers, but it will die.

Moving forward, I’d probably like to try the Griselbrand/Goryo’s Vengeance deck that Top 8’d the GP. I’ve always had a soft spot for that deck and I tried to make it work for every Modern PT, but I kept running into problems that involved my inability to connect with Griselbrand against things like Affinity and Twin when I was running Fury of the Horde. I think the Shoal plan fixes all of those issues, and even lets you kill instant-speed, so it’s a big upgrade to the deck. If that doesn’t prove good enough, though (or if people react to it and play a bunch of hate), then I’ll probably just go back to playing Twin.