Well, it's been quite a week hasn't it? The banning of Splinter Twin really touched a nerve with the whole community. However, that is now firmly behind us and continuing to bemoan (or extoll) that decision adds nothing new or useful. Instead, let us look to the future and answer the burning question: What now?

Looking to the immediate future, next weekend there is a StarCityGames Open with a Modern Classic in Columbus. The weekend after there will be a Modern Pro Tour in Atlanta that some of you will be going to, and I wish I would stop getting second at PPTQs so I could join you. Therefore I (and most of the rest of you) will be attending SCG Regionals the Pro Tour weekend instead. To that end I, and all of you I expect, have been extensively testing Twinless Modern. I'm going to share my testing process today, as well as where I am in my deck selection process. Hopefully this will help you to refine or jumpstart your own, or at least avoid some of the pitfalls I've run into. Let's start at the beginning.

[wp_ad_camp_1]

It's a Good Place to Start

When a new set "normally" comes into Modern the main considerations you need to make are:

1. What was the metagame before the release?
2. Did the release add any new cards/decks?
3. Do those new cards affect my deck?
4. Does this affect any of my matchups?

This will obviously depend greatly on the set and your deck. Khans of Tarkir obviously added many new cards and even some decks (Jeskai Ascendancy), drastically changing the Modern landscape (beyond Treasure Cruise I mean). In the Khans case, most players should have been answering "yes" to all those questions. A set like Born of the Gods on the other hand? Not so much (Brimaz, King of Oreskos and Courser of Kruphix see a little play).

This questioning doesn't work when a banning is thrown into the equation. It suddenly becomes a far more complicated affair, with the additions of:

1. Does the loss of a good/bad matchup make {deck} (un)viable?
2. Is {brew} now playable?
3. What becomes of {banned deck} players?
4. How does my sideboard need/able to change?
5. What will the metagame look like?

Etc., etc., etc. While these are important questions to consider, addressing them all at once is not just overwhelming but unhelpful. You need to simplify. Scientists seek to test as few variables as possible for a reason, and in analyzing a new metagame, especially one where a format staple is gone, we should follow suit. My education was in economics and in that field we theorize and test variables ceteris paribus or all other things the same. While this isn't the most "complete" way of thinking about a new metagame, it is certainly more digestible, so I always start with a few ceteris paribus assumptions. Here are those assumptions after the Twin banning.

1. Decks with a bad Twin matchup get better
This shouldn't surprise anyone. Everyone, myself included, has lamented the fact that Affinity's worst matchup is now gone, a matchup that was bad for Tron, Scapeshift, and (depending on who you ask) Infect. Now that the turn four, two-card combo is no more, any deck that struggled to race it suddenly has more time to set up and/or can become more linear by eliminating the interaction they had to run to keep the matchup manageable. Exactly how this will pay out is unclear, but the baseline assumption still holds.

1b. Corollary: The converse is not necessarily true.
Just because a good matchup is suddenly gone does not necessarily mean your deck is not viable. Twin was 12.8% of the metagame on average over the last year so if the only reason to play your deck was because it beat Twin then it probably wasn't all that good to start with. Your deck must still have other good matchups and they remain good without Twin in the picture ceteris paribus.

2.  Everyone knows about Assumption 1
This also shouldn't surprise anyone since it's obvious and has been what everyone has talked about. If you aren't aware of the above Assumption 1 then you're doing a very good job hiding in your cave (can I join you until the election is over?). Across numerous sites and message boards, including our comments section, players and authors have been saying that Affinity is the big winner from the ban so it stands to reason that every enfranchised player is aware of this whether they agree with it or not. Thus, at least some will act on this belief. There will be attempts to "Next Level" the tournament, either through unusual deck selection or metagamed sideboard choices.

3.  Bx Eldrazi will continue to gain popularity
This was true before the ban. There's no reason to think it will stop now. Eldrazi has been everywhere online and in articles and there is no reason to think it is going away.

4. Popular decks will stay popular
Modern decks are expensive so it stands to reason that most players, having invested in a deck prior to January 18, will try to stick with their deck from before unless forced to change. Following this, a Burn player is more likely to remain a Burn player than change decks because a) they're invested both financially and time-wise, b) they already know how to play the deck, c) the Sunk Cost Fallacy means people tend to stick by their investment regardless of the value of doing so, and d) the deck may actually still be good.

You could certainly make more assumptions, but I'm sticking to these since, again, too many variables makes for unfocused study. Also, I have no way of knowing what the former Twin players will be doing. I'd imagine they'll stay URx, but whether or not that will be Delver or Grixis-style control is impossible to know, so I assume nothing about them. Looking at these three-and-a-half points, Assumptions 1 and 3 seem like good places to actually start our investigation. 4 is useful for purposes of testing since it allows us to use known decks from the previous gauntlet without worrying about wild changes being present, but it's not really something we can analyze. Basically it's me calling ceteris paribus on the rest of the metagame.

The problem with Assumption 2 is that it is very murky and relies on your ability to predict your opponents biases and deck selection. Go too far into this line of thought you fall down the rabbit hole of I know you know I know and that just confuses everybody. Unless you're going to a very small tournament, either an Invitational or your LGS, where you know exactly who will be there and what they tend to play, don't try to metagame deck choice in a new format. In an open tournament there is far too great a risk of Next Leveling yourself and picking a deck with no chance of winning against the actual metagame.

It's like picking Scissors in a field full of Rock, which Mike Flores, Brian Kibler and Richard Feldman wrote about a long time ago. For those too lazy to read those important pieces of theory, picking a targeted anti-deck pays off when you have The Read on a given tournament and fails unless you have a more general Plan B or a transformational sideboard. With this being Modern and sideboard space at a premium, that transition is quite hard. Stick to the variables you can observe and control and disregard everyone else's deck choice in your decision making. It will save you from inbred testing and a severe headache.

What do our Assumptions Mean?

What 1 and 3 suggest to me is that linear aggro and big mana strategies will be more popular than you would "normally" expect, at least in the short-term. Scapeshift is kind of an outlier since it is a combo deck that uses lots of land, similar to how Eldrazi is a midrange/ramp deck that uses a lot of land. That said, all these decks are trying to execute their gameplan while ignoring the opponent as much as possible. This suggests if you want an edge in deck selection you should either seek to force interaction and cripple the opposing game plan or have a faster goldfish. I consider these assumptions my testing parameters and use them to determine which deck(s) I spend my time on. Assumption 4 means that once I establish my deck's viability against the testing parameters, I need to run the full gauntlet and find out if my results are actually valid.

Where I Started

Those of you who read last week's article will remember I was working on both Esper Mentor and Merfolk. However, they aren't the only decks I keep in stock. I also keep UW Titan and Wx Hatebears mostly built in case the meta becomes favorable. With Twin gone and a poor matchup for the latter two decks eliminated, I gave all four decks a stress test under my established parameters to see how things went. I picked representative Affinity, Tron, and Infect lists from the Charlotte Open (they're all almost the same at this point), Thien Nguyen's Scapeshift list, and built an processor-based and Heartless Summoning version of BW Eldrazi. I modeled it on numerous lists I saw online using Thought-Knot Seer and Matter Reshaper. Then I played as many games against them as I and my testing partners could stand.

UW Titan

My Titan lists harken back to the pre-Deathrite Shaman era of Modern when Bloodbraid Jund dominated and the best way to grind out a win against it was recurring Kitchen Finks (not the best way to beat it, just to grind it out). Instead of the dedicated recursion engine Trevor showcased a while ago, my lists are more about maximizing individual value than the whole being greater than the parts.

When I've played the deck previously it had a very good matchup against Grixis, Burn, and BGx, and could hold its own against Tron and Bloom thanks to Ghost Quarter. The countermagic kept Scapeshift reasonable. The problem was it was a bit light on creature removal and so was vulnerable to Affinity and Infect pre-board. This was still true now. To make matters worse, Eldrazi is far more resilient to Ghost Quarter than Tron or Bloom since Eldrazi Temple and Eye of Ugin are more readily replaced by basics than Tron pieces are. UW Titan also wasn't fast enough to race Oblivion Sower and Ulamog, the Ceaseless Hunger. While it was still a solid choice against the rest of the metagame, I didn't think it could be made good enough to overcome the test parameters and so I tabled Titan. If BGx comes back in a big way then I'll run it, but not now.

Wx Hatebears/Death and Taxes

The thing about Hatebears/Death and Taxes is that you never ask if the deck is good but rather is Leonin Arbiter plus Ghost Quarter and Thalia, Guardian of Thraben good enough. In my experience, the more degenerate the format is, the better these cards become: making degenerate decks play 'fair' usually causes them to fall apart (just look at Legacy DnT vs. Storm). During the Treasure Cruise era they frequently were, although you also lost to Forked Bolt quite a bit.

In the previous metagame, Thalia and Arbiter were rarely enough. Affinity was too fast and creature-heavy for Thalia to be good and too many other decks played too many lands for mana denial to be effective. You'd sometimes get lucky but more often you were just delaying how long it took for your creatures to be hopelessly outclassed.

This proved to be the case again, and while Strip Mine was surprisingly good against Eldrazi, if any of their threats ever resolved it was game over. Also, while the disruption package was decent against Tron, Pyroclasm was much better against us and swung many games in Tron's favor. It's harder for this deck to play around sweepers than Merfolk since your creatures are also your disruption, do not exponentially scale, have low toughness, and you need a more of them in play to win. I tested several different color/card combinations and configurations and the results were the same in all lists: if you couldn't set up the soft lock, took too long to win, or if they wriggled out of it on their own, you lost. Craig Wescoe and Eric Froehlich might disagree, but I don't think Hatebears is a good choice for the new Modern.

Merfolk

The list I showed you last week was the result of changes made based on the above assumptions. So far they've been right on the money and the deck remains unchanged since that past Monday.

Prior to the banning, the deck had poor matchups against Affinity and Infect. It also always lost to Tron when it played Ugin, the Spirit Dragon, but normally won otherwise. Scapeshift was a manageable race and Eldrazi was a cakewalk. The Bx Processors didn't interact enough and was not sufficiently fast to win before you swamped them with fish. With the updates, the Affinity and Infect matchups improved slightly, though you were still unfavorable despite Echoing Truth and Harbinger being very good in both matchups. Scapeshift was worse without the counters, though still reasonable, and Eldrazi was still good even if Thought-Knot Seer proved better than expected. This satisfied me enough that I'm moving ahead with testing this deck for Regionals.

Esper Mentor

Like I said last week, I was very disappointed this deck lost a good matchup in Twin and was worried about both Affinity and Eldrazi. I also noted Affinity was probably fixable with Drown in Sorrow. The price, as it turned out, was that I had to cut the Monastery Mentors to make Drown work.

The Drowns were surprisingly good against both Infect and Affinity, greatly improving the matchup pre-board. The loss of Mentor was not terrible since it was never great in my bad matchups and died too readily against the midrange decks. He could potentially produce a huge clock against control decks but isn't where you want to be against big mana. Tron, Scapeshift, and Eldrazi were still problems, but that was mostly the result of Inquisition being very bad against them. As this seemed fixable, I decided to move ahead with testing Esper as well.

Where I am Now

There isn't much more to do with the Merfolk list. I've tested Sea Gate Wreckage as a one-of and it's good against attrition decks but pretty useless elsewhere so I'd leave it at home for now. The sideboard is still in flux and I need to wait until we get a better idea of where the metagame is heading before I commit to anything. Harm's Way is looking attractive for both Burn and aggro matchups, but I'd rather not board in purely reactive spells. I may end up wanting Sunlance if aggro is big. I'm not as keen on 'Folk as I normally am since it did very well in Atlanta this past weekend and will be on people's minds going forward. It also suggests I should update my sequencing guide for the Merfolk mirror. However, my pet deck continues to prove the Merfolk core is still very strong and resilient despite being a well-known quantity. Harbinger of the Tides finally has the aggro and (judging by Todd Anderson's deck) tempo matchups it needs to really shine. Spreading Seas is proving its worth slowing down Tron and Eldrazi. If Affinity and Infect weren't looming in the background, it would be an easy pick for me.

Esper Tokens, on the other hand, is showing considerable promise. It is very strong against Affinity and Infect thanks to discard and Lingering Souls. Swapping out the Inquisitions for Thoughtseize improved the big mana matchups and are still reasonable against aggro (Burn, less so), though I've been swapping them for the less painful option in sideboard games. I want both since extra discard is good against Grixis, Scapeshift, Grishoalbrand, and Jund so it's more a choice of which is the four-of in the maindeck. Esper Tokens is still somewhat weak to them, and that problem may not be fixable, but now you can prevent their big turn more readily and are more likely to race with Finks and Spirits than before.

I like the scry from Drown, but if Abzan Company is big in your area then Flaying Tendrils is better. This deck is the Abzan to Grixis' Jund and grinds very hard with Lingering Souls and Sorin. The manlands are also very important in a lot of matchups. I added another Finks over Clique due the former's value in offsetting the additional life you're losing from Thoughtseize. I've still got two weeks to work on it but I'm curious to hear your thoughts as well.

This still doesn't answer the question of what deck I will play for Regionals and the reason is that I don't know yet. Merfolk will be better against a metagame full of big mana while Esper is the pick if aggro decks rule the day. Testing against the rest of the field is underway and, frankly, both decks are about equal for me. Tokens has better aggro matchups while Merfolk's speed is better against combo. Both are reasonable against midrange. In this situation, there is no true "right" decision. Each deck is weak against an opposing end of the metagame spectrum and are good to even against everything in the middle. The choice comes down to personal preference and what you perceive the metagame is doing.

Unfortunately, we just don't have enough information to make a guess (and no matter what anyone says it will be a guess) at what the February 6 metagame will look like. The SCG Atlanta results suggest an aggro metagame, but one datapoint is not a trendline.  If this continues through Columbus at the end of the week, that will be significant and will make my decision easier. If not, then things will get interesting. I'd advise you to keep watching SCG for IQ and Classic results and make your metagame call no later than next Tuesday so that come Friday you will be ready for whichever event you're attending.