Sullivan Library – A Case Study in Playing the Wrong Deck

On Saturday, September 6th, StarCityGames.com will be exhibiting at Your Move Games' 5K Magic Day in Boxborough, Massachussetts!
Friday, September 5th – The Block Constructed season is all but over, and it’s time to take stock of the lessons learned and the pitfalls avoided. Adrian looks at his deck of choice this season – Mono Black – and warns us about the dangers of unwarranted perseverance. He also brings us the updated version of the deck, for those players looking to rock out the Boxborough 5k this weekend.

I played the same deck this last two weeks in my attempt to qualify for Pro Tour: Berlin. Truth be told, it was probably the wrong deck. But I plugged away with it and played it anyway, even though I think that some part of me knew, deeply knew, that I should be playing something else. Despite that, I plowed on, and played it anyway. I devoted, perhaps, far too much effort on it. And, unsurprisingly, I didn’t qualify.

Here is a case study in playing the wrong deck.

A — Promise

I think it is hard to get lulled into going down the wrong path if you don’t see some particular lure or promise to it. If you’ve watched enough movies with con artists, you know that the con can get you the moment that you start thinking how you can get just what it is that they are promising you. You wouldn’t care if the con offered you a chance for something worthless. But if they offer you something valuable, you stand up and take note.

In my case, I had started working on a mono-Black midrange deck. It was inspired by a lot of things. One of the big ones was a conversation between me and Zac Hill on our drive back from Grand Prix: Indianapolis. It just seemed that there were so many tools that were available. You got to have the vicious Thoughtseize into Bitterblossom play. Further, once Eventide came out, you got to also add on four Stillmoon Cavalier. Finally, there was the lure of Demigod of Revenge. Holy wow! Demigod draws were one of those clearly unfair things… Those were some incentives.

I put together a crappy first-draft build and sat down against Kithkin. If nothing else, it seems like the most relevant place to start is with the aggressive deck in the format that can just “get you.” I sat down against a couple of random PTQ winning decklists (even if I wasn’t completely confident in their build). With a package of Black removal, and potent creatures including Stillmoon Cavalier to completely hold them down, it went surprisingly well. Further, once I included the Ashling the Extinguisher/Trip Noose combination in the deck, it went even better. The Trip Noose was going to be my nod to the real problem of Chameleon Colossus, but when you would get our Ashling/Noose, oh man!

Again, a small promise. It seemed like there was something there.

Here is the version of the wrong deck I played at the final PTQ in Detroit:

Sisters Grimm is named after Jamie Wakefield classic mono-Black midrange beatdown deck, Brothers Grimm, so named because of the two Legendary “brothers” Gallowbraid and Morinfen. Here, the “sisters” are Oona and Ashling (though, I’m sure Vorthos-leaning players will tell me that they aren’t related).

There’s a lot to say about this deck, but I’ll get into that in a while.

B — The Payoff

Of course, during this time, the world was up with Five-Color Control, and for good reason. I sat down with a slightly modified list against several different versions of Five-Color Control, and it seemed like I was just decimating it. I decided to actually actively test the matchup.

In my first formal session, I went 18-2 against the deck. That is eye-opening. Screamingly so.

When I visited Patrick Chapin in Detroit, he told me that he didn’t believe that could be the case. We played a short set of 8 games, where I won 5 to 3, and then boarded games, where he largely crushed me. We both walked out of our playtest session with different conclusions. He felt like I only barely beat him in the first game, whereas I felt like all of my wins were crushing wins, and his wins were more marginal. He felt like things got worse for me in game 2, and while I agreed, I felt like it was largely a question of poor boarding ideas.

In the time since I tested with Patrick, I’ve tested more, and I’ve come to the conclusion that it is definitely overwhelmingly to the advantage of Sisters Grimm. Unfortunately, though, that payoff is misleading.

At a PTQ, only a few players there are going to be skilled enough to actually pilot this deck that I want to be playing against. The rest are likely to fail with their Five-Color Control variants, leaving me with this powerful anti-deck that doesn’t really have a valid target. Further, more people will be likely to be playing strongly aggressive Kithkin and mono-Red beatdown decks. To push the deck in a direction where it can beat both Kithkin and Five-Color Control, unfortunately you find yourself struggling to fit in all of the cards you need. Then, all of this contorting aside, you’re open to other decks that might not be the best decks out there, that aren’t likely to win the PTQ, but are likely to be floating at the top of the field. All of the varieties of Chameleon Colossus decks, for example, are eminently beatable, but every step you take in trying to beat them comes at a cost.

The problem is something that Patrick brought up. While he wasn’t convinced about the power of the deck versus Five-Color Control, he did note that the deck didn’t have the intrinsic power of the big three archetypes. Kithkin, Faeries, and Five-Color all do a particular thing, and do it well. My deck essential threw a whole lot of stuff at the other player, and hoped one of them stuck.

Perhaps this is the reason it could do so well against Five-Color. If a Bitterblossom didn’t get them, perhaps a Stillmoon Cavalier would. If that didn’t get them, perhaps numerous Demigods would. If that didn’t get them, perhaps an Oona would. And as they were answering here and there, maybe a Corrupt would just finish up the job. Against Kithkin, you could reasonably expect to be able to kill and kill things, while defending yourself. But against the wrong opponent, some of the cards simply weren’t potent enough, or appropriate to their threats. What do you do about a Mistbind Clique if you don’t have the Unmake in your hand? You get Time Walked, that’s what.

The payoff of crushing the de facto “best” deck, according to a lot of the thinkers and players of the game, was having a deck that might not actually be suited to the metagame. It wasn’t just that you had some real problems to deal with, in terms of reasonably likely opposing decks. It’s that you might find yourself struggling against something random. A Standard Elf deck, like my Elf deck that Sam Black used to pilot his way onto the U.S. National team, can simply be a deck that is deeply aggressive while putting a little bit of disruption with Thoughtseize and removal to take you down. Sisters Grimm has a lot of avenues of attack or defense, but since it isn’t focused on one in particular, it is vulnerable to a diverse metagame (something that I think existed at the end of the format), as well as suffering just in general from a deck capable of a really focused attack, like Red or Faeries.

C — Options, Distractions, and Manpower

I had inklings of the problems that the deck might have, even if I did pooh-pooh them. So why didn’t I spend my effort on another deck?

The question might initially be “which deck,” and I felt like there were so many options. The PT Jank deck that I recently discussed looks kinda ugly, but it actually performed quite well. I knew that it wasn’t polished, but at the same time, I knew that it could beat all of the decks that a typical Kithkin deck could, but the existence of so much removal gave it real power against decks that traditionally could edge out Kithkin.

Fish was another deck I found interesting. I had already had a near-brush with qualification with a pre-Eventide build of Fish, and there were just so many new options that the mind boggled. Pre-Eventide, it felt like the big question was Fish versus Faeries — it seemed to close to being like Faeries versus Faeries, a coin flip. I felt like Fish had a 55/45 edge, but too much of the results were tied to who went first. Eventide, I was sure, could solve some of those problems. I believed, like Sisters Grimm did, that Fish would be sure to K.O. the big enemy of the moment, Five-Color. It might be promising.

Which Fish list should I even run? My own was clearly out of date. Jeff Fung’s Four-Color Merfolk seemed powerful, but ultimately wrong to me. I knew that I might be wrong on that count, but the strength of Fish, to my mind, was its ability to be a true aggro-control deck. Fung’s deck was pushing deep into mid-range beatdown territory, with the potential for aggro-control plays, making it more akin to a Faeries variant in some ways. I wasn’t sure, but it felt not quite right. Then, there was Mike Thompson’s second place list out of Seattle, which also felt misbuilt to me, but seemed a good deal closer to ideal, even though I was pretty sure it was at least nine cards off where I would want it to be. There were so many options!

But I only had so much time. In order to figure this out, I’d have to put in the level of manpower to which I simply didn’t have access. My own time was the big limiting factor. Without much time left on the clock before the first PTQ in Indianapolis, I let myself be lured in by the payoff of Sisters Grimm. It was so bad, in some ways, that I didn’t figure out two things I had severe questions about with the deck: was the Mutavault worth it (the answer, I now know, is a resounding no) and how can I cut down the deck to 60 cards and still have it perform the way I want (the answer, I think, is to massively overhaul the deck).

With more time, what could I have had? Maybe I could have had a more well-spun deck. Or maybe I could have come to the table with an actual, factual aggro-control deck — I could have had a deck that theoretically would have had a huge edge on any of the more controlling decks. Versus opposing aggressive decks, cards like Cryptic Command and access to Pollen Remedy from the board could give me a huge “Fog” edge, since I would have access to so many more ways to stop their damage, and if they chose to get into that fight with me, counterspells could stop their attempts.

I just didn’t have the time in. I knew that the potential was great, but I dodged it because maybe I couldn’t get as far as I had already gotten. And, besides the fact…

D — Wasted work! (And love…)

… what about all of that time I had spent?!

This is a terrible way to think about what deck to play. Just because you’ve put in a lot of effort, that doesn’t mean that you have a sufficient reason to continue on the same course. Essentially, you’re making the same foolish argument that has been made in another context: “You wouldn’t change horses midstream, would ya?”

Well, maybe you should, if the horse is drowning.

The thing to remember is that we should never fall in love with our decks. If, somehow, we can’t help it, we shouldn’t let that love blind us to the real problems that a deck has. Loving a deck might help propel us to do the work that we really out to be doing, but it might also make us “forgive” the deck for a failing, or overlook its faults.

Work isn’t wasted if it has given you conclusions. If you’ve put in the time and the effort, and it tells you that the deck you thought was The One is no longer the right call, you really do have to listen to it.

In my case, I knew somewhere, deep inside, that I was struggling to hard to contort the deck to deal with everything. I knew that in trying to accomplish this, I was costing myself against Faeries (because no one plays Faeries anymore, right? Har har…). But a part of me ignored all of this, partly out of love for the novelty of the deck, and partly out of love of being able to sidestep people’s expectations. There is real power in that, true, but you can’t pretend that that is enough. Going rogue, like love, is never enough for its own sake. It takes more than that.

E — Risks and Rewards

And what was the final mistake? It was examining the known risks and rewards and not weighing the correctly.

Gamblers often talk about EV, and they aren’t the only ones. EV is expected value. What can you expect the outcome of something to be? It isn’t the measure of a single instance. It is the measure of probable outcome, based on evidence. Betting a dollar on the outcome of a coin flip has an EV of zero. If you do it a hundred times, you’re liable to end up, on “average” right where you started.

If there is one thing that we don’t want to be at a PTQ, it is average. A fifty-fifty record is only acceptable in the rarest of circumstances, and then, usually only in mixed format events where you might expect to have a huge lead on your opponents going into that portion of the tournament, points-wise.

My error was not giving enough weight to the problems that would be afforded by Faeries, by Chameleon Colossus-wielding players, and by burn decks. By concerning myself primarily with Kithkin and Five-Color Control, I was giving myself a lopsided reading on my probable results. I knew, for example, that I was a heavy favorite in those two matchups. But I didn’t actually know the weights of many matchups. Doran, for example, felt bad. But how bad? Oh, I don’t know, I probably won’t play against it, right?

This makes an EV actually impossible to calculate. The more places you don’t really know how the matchup works, the more you are essentially just winging it. Merfolk? No idea. (I imagine it might be bad.)

Even if Fish might not have been a deck that I had done enough work with, at least there is a body of historical data that I can lean on with it. I know that historically aggro-control is good against control decks. Wherever I’m weak with data in calculating EV, I can lean on history, and make Bayesian-based predictions (i.e., reasonably informed probabilistic “guesses,” really).

Midrange Black? About the only short-cut I can make is pretty weak: Red beats Black. (And there, to be sure, I’m not sure how much certainty we could give that, if we were to try to use that to make informed predictions.)

If you don’t know your risks, it doesn’t matter how great your rewards are.

Still, though, there was a deck here. Yes, I was lead astray, but the reason I was lead astray was because there was at least something that glittered in that pan, even if it wasn’t that much.

A Good Deck Doesn’t Mean Good Enough

The thing that I really liked about the deck was that it could throw out so many potent forces. Even if the deck wasn’t quite good enough, there are some things about it that seem worth noting.

Thoughtseize: There is something really great about 4 Thoughtseize, if you can support it. So often, you can just make a game out of one that shouldn’t have been a contest. You can take away the only other Kithkin in a hand with a Stalwart, and easily win back the loss of two life. You can rip out a Bitterblossom, and leave a Faerie player with an ugly hand to try to win a game with. You can protect a threat so that it sticks when you cast it. The loss of two life is real, but the effect is potent. So many hands turn on a single card that the ability to strip that card away is a very valuable weapon.

Stillmoon Cavalier: As a protection from Black and White creature, Stillmoon Cavalier is able to naturally be an enemy to both Kithkin and Faeries. Even against Five-Color, Red, and Doran, there are a number of ways that it can be useful, often holding off a big creature all by itself. While it is easily killed by Firespout or other access to Red, it still remains a creature that can end the game by itself.

Demigod of Revenge and Oona, Queen of the Fae: Both of these creatures, in different ways, have the ability to end games. A resolved Oona, upon untapping, often strips away an opponent’s ability to win the game unless they have access to protection from Black. Demigod of Revenge, conversely, doesn’t care so much about resolving as it is able to cause unfair moments, when one or more come back into play from the graveyard. Even without multiples, as a kind of Tarox, it can just provide the kind of pressure that will set someone to backpedaling.

Ashling, the Extinguisher and Trip Noose: Ashling’s big weakness is that by itself, it can’t deal with anything that can face it in a fight. Doran, Chameleon Colossus, Stillmoon Cavalier, Horde of Notions, and anything else large can fight with an Ashling, and walk away the winner. Anything else, though, is stuck. You can’t race an Ashling unless you are so far ahead in the race that an Ashling can’t catch you up as it eats away your army. When you toss in a Trip Noose, it becomes all the harder to maintain a race, and problematic protection from Black creatures, albeit not killable by Ashling’s power, are still held down. Getting out both provides a super-powered version of the Abyss.

Bitterblossom and Leechridden Swamp: Bitterblossom, by itself, can sometimes provide a win. These days, though, most people are so well prepared for it that it is not enough. When you add on Leechridden Swamp, the damage becomes inexorable. As Patrick Chapin put it during our testing, “I can answer everything in your deck but that card.” And that would often be how it would go. I would power up a Leechridden Swamp with Bitterblossom or Demigod or what-have-you, and he just wouldn’t be able to kill me before he would die. This happens so often that you’ll find your opponent often realizing that they have to spend a scary amount of effort to kill everything that they can, lest a few stray points make the Leechridden Swamp kill them before they are ready to deal with it.

Unmake, Beseech the Queen, and the removal package: The combination of elimination can be varied here and there, depending on your metagame, but the Unmakes give you so much ability to take care of nearly anything. Add to that a healthy slew of two-mana removal, and a ton more access to yet more elimination from the sideboard, and you are well-armed to deal with the Kithkin lists. Corrupt can be a reasonable way to finish off a Five-Color player, recover from Kithkin, or have a shot against Red, but are often weak elsewhere. Beseech the Queen makes having access to even a single Corrupt a very different exercise than having none available whatsoever, and it also multiplies the useful effect of cards like Trip Noose without magnifying the clunky aspect of that card as well. Beseech, while often fetching elimination, is also useful in fetching that extra land or magnifying the impact of Demigods.

I sent the list to Evan Erwin. He tinkered with it and ended up with a list that was quite inspiring to me. Here is the version he took to an X-2 performance by the end of his PTQ:

4 Leechridden Swamp
4 Fetid Heath
2 Vivid Marsh
15 Swamp
1 Reflecting Pool
4 Demigod of Revenge
2 Oona, Queen of the Fae
4 Stillmoon Cavalier
3 Ashling, the Extinguisher
2 Beseech the Queen
2 Nameless Inversion
4 Unmake
2 Hallowed Burial
3 Bitterblossom
4 Thoughtseize
2 Runed Halo
2 Corrupt

This is far more ambitious than I had gone with the deck. A lot of this had to do with the unresolved questions I had about the needs of the deck. I was maximizing my Swamps as a means to power-up Corrupt, but the question I should have answered before I went there was “Is it worth it?” After a lot of testing, I think I’ve come to the conclusion that it isn’t.

Evan’s build, although more ambitious, still has some problems in the details, I think, though it is incredibly close. With access to Beseech, I still think that you want access to a single Trip Noose. Corrupt, even as a potential singleton, makes a lot less sense to me if you don’t have a maximal amount of Swamps. The land mix should be moved around to accommodate 27 land, a number that I constantly wanted to have as I played out the deck. The particular mix of land should also be shuffled up. Life gain, is still something that perhaps should be nodded at, though. While I’m dubious of a single Corrupt, it might be the way to go, or maybe Brian Kowal’s suggestion of a Divinity of Pride might be the way to go. I actually don’t mind giving up a little bit of early game to include a single Austere Command.

The synthesis of my thoughts with Evan’s deck gives us this:

I could imagine cutting the Noose or an Ashling for that Divinity if you went that way, though I think that this is pretty close to what you might want to play.

Briefly, this sideboard is based on some theory and a lot of actual testing. Shriekmaw, Soul Reap, and Recumbent Bliss are there to add onto your weapons against aggressive decks of all stripes. Austere Command can serve as a catch-all answer to anything, and Festercreep is a tutorable answer to Stillmoon Cavalier, or simply a mean to wipe out a load of Faeries. Mind Shatter is good against anything that wants to keep a hand, and Puppeteer Clique against anything that wants to keep creatures in the yard. Memory Plunder has tested magnificently as a means to be able to be a Cryptic Command deck on your own, or randomly be able to answer problematic cards with your opponent’s own weapons.

Without more testing, I still don’t know that this deck would have been good enough. My gut says that I would have gotten more mileage if I had cleaned up what I didn’t like about Thompson’s Fish list, but, that said, I love how Evan’s concepts give this deck so much resilience against the problems that I fell to in my own work on the deck. It looks like it still holds some of that promise and payoff that the initial list held, without simply being in deep trouble against a whole slew of cards. The White gives it the kind of pure power, I think, that Chapin critiqued it for not having. What is the cost to include these cards? A Trip Noose, some quick Black elimination, and Corrupt. Seems fine to me.

Overall, I’m really excited about how the deck played out, despite failing to perform like I’d like it to. It was a good reminder to me about where going rogue can go wrong. You can’t really afford to take shortcuts or take things for granted when you go rogue. You have to put in the work, or else you’re just gambling with dice that might well be loaded against you.

See you next week!

Adrian Sullivan