We love deck lists.
We love them enough that Constructed PTs are much more popular from a coverage point of view than Limited PTs, such that we get to have mixed-format PTs now so that we can incorporate drafting while still retaining those deck lists we adore. Of course we might, in a moment of introspection, wonder why we love deck lists.
From my perspective, I look at deck lists for a number of reasons. I want to understand the ongoing metagame. I want to know what decks are succeeding in that metagame. In my case, that’s because I need to beat those decks to win, but for many, it’s also an indication that they should be playing those decks. Finally, I just want to see cool ideas, and I’m hoping that someone out there has come up with something cool, or at least the seeds of something cool that I can work into a developed, successful game plan.
Where are the deck lists?
We’re awash in options in terms of finding decks, but the path we choose can and should differ depending on where we are in the Magic product release cycle. Leading up to and soon after a set’s release, you’re looking for ideas, and that will naturally lead you to columns like this one, as well as the forums. There our goal is to just scrape information about how cards might be used, to leverage the imagination of others well ahead of the formation of an actual metagame. Once we’re several weeks into a product’s release, as we are now, it’s time to turn to my two favorite resources.
First, we have the Magic site’s Decks of the Week feature. This aggregates decks from both Magic Online events and real-world PTQs. I adore Decks of the Week during Constructed PTQ seasons, as it gives us a window into the ongoing, shifting PTQ metagame. One thing to watch for in Decks of the Week is that it does not report on tournament size, which can lead to “How did that win?” moments, before you realize you’re looking at the results from a 15-person PTQ in Anchorage, Alaska.
My other primary deck list source is DeckCheck.net. DeckCheck tends to overlap significantly with Decks of the Week, but has the advantages of casting a somewhat wider net for results, encoding the deck lists in forms that are useful for subsequent testing, and being syndicated. This last feature is great, since it means I’m just subscribed in Google Reader and don’t have to actively track down new data. DeckCheck listings also tend to be good about saying how large a tournament was, so that we’re less likely to be misled by small tournaments. It’s important to remember that Nationals-level tournaments in many countries are smaller than routine PTQs in many parts of the U.S., for example.
The most recent Decks of the Week feature was particularly abundant, listing the results from twenty-four Magic Online Standard tournaments.
What do they tell us?
As I mentioned above, we look to published deck lists to give us an idea what we might face in a tournament, as well as what’s succeeding right now. However, except in those rare instances where we have complete information about all the deck lists at an event, we’re naturally trying to draw conclusions based on a limited, successful subset of decks played. In one sense this prepares us for the right task, since we want to beat the winners if we’re going to win. In another sense, it can be dreadfully misleading.
In bioinformatics, we’re often faced with situations where we have information, but don’t quite know what it means. For example, we may want to try to find genes within a DNA sequence. We have examples of known genes, and we can use this information along with a tool called a Hidden Markov Model to try and find new genes. In this, as in other computational tools, we look at the results (known genes) and try to extract a broader understanding of the world from them (in our example, rules for “Is this a gene?”).
In analyzing successful deck lists and attempting to understand the metagame and what makes a successful deck, we’re carrying out a similar exercise. It’s important to understand this because it influences what conclusions we draw and the validity of those conclusions. I’ll look at this now in the context of the recent giant list of Standard Decks of the Week.
The overwhelming take-away from last week’s lists is “Jund.” Most of the Standard events listed feature a majority of Jund decks. What does this tell us?
First, there’s the possibility that Jund is the “best deck.” We need to be cautious about this because we don’t really know how many people are actually playing Jund. At the most recent Pro Tour, half the decks in the top eight were Zoo. Of course, there’s the issue of a mixed format to confuse us here, but even leaving that aside, some 22% of the players at the event brought Zoo decks. Clearly, Zoo is somewhat enriched in the top eight, but it’s not the same as if 5% of the field were Zoo with the same top eight results. If a majority of players are bringing Jund to these events, it’s not necessarily meaningful if we see a plethora of winning Jund lists.
Second, there’s the possibility that “good players play Jund.” If you have some time to burn and want to really wade through those Decks of the Week entries from Magic Online, you’ll notice the same names turning up over and over again. Who’s playing a deck is just as legitimate an axis of differentiation in these types of analyses as which decks are being played. To put this another way, we may be learning that you can expect to play against Jund in the late rounds, but that it still isn’t necessarily the bulk of the field. I don’t think this is the case, but we have to be cautious about it nonetheless. This issue brings us the idea, one you may have seen before, of making decks that can win in the top eight of a tournament, but won’t make it there – and vice versa.
My overall impression from the current swathe of deck lists is that Jund is not necessarily the best deck, but it is abundant enough in later rounds of tournaments that it remains the quality benchmark for doing well. In other words, Jund is the “you must be this tall to ride this ride” sign for the current Standard metagame.
As we continue to piece through deck lists, we also need to be cautious about the conclusions we draw from cards in those lists. Here we endanger ourselves with the problem of being “results oriented,” assuming that a card in a winning deck list is the best choice. To return to biology for a moment, new students often assume that evolution makes the best choices, when, in fact, it really just makes “good enough” choices. The same is true for deck lists. A winning list was good enough to win that tournament under a specific set of circumstances.
Consider that at Pro Tour: Austin, Tsuyoshi Ikeda had two Riftsweepers in his sideboard and did not know why they were there. He’d accepted the deck list without asking for an explanation, and while they happened to prove useful in the Next-Level Blue matchup, he wasn’t really sure why the designer included them. Were they the best choice for those two slots? Tsuyoshi certainly couldn’t tell you, and he took second place in a Pro Tour.
Similarly, recall GP Los Angeles 2009, where Mat Marr ran four Rune Snags in his Faeries deck because he couldn’t find any Mana Leaks. Various Magic forums spawned long discussions about the merits of Snag over Leak in this spot, but the simple fact was that Mat couldn’t find any Leaks, so he went with plan B.
Returning to the current metagame, I was initially fascinated to see that a number of Jund decks ran three or four copies of Terramorphic Expanse. That was certainly interesting, especially in light of my discussion of the value of Terramorphics in Nayamorphic last week. However, after plowing through more lists, I noticed something important.
Some lists ran Terramorphic Expanse. Some lists ran Verdant Catacombs. Almost no lists ran a mix of the two.
My tentative conclusion, then, was that we were looking not at conscious, effectiveness-based decisions. Rather, that looks like the footprint of a card availability issue, like Mat Marr’s Rune Snags. Catacombs is an expensive card right now, and the difference in effectiveness between Expanse and Catacombs is probably not so extreme as to keep one from making a 4-0 or 3-1 record in a Magic Online event – at least, not in these still-early days of the Alara-Zendikar Standard environment. Perhaps bolstering this hypothesis is the fact that as we move to those events later on in the month, we start to see Catacombs creeping into more builds in small amounts (for example, a mix of three Expanses and one Catacombs).
It’s very useful, and important, for us to remember that as much as we attempt to be the best, a winning deck need only be “good enough.” About a month or so into a new set, we should suspect that a lack of Zendikar cards in a deck may be less an issue of conscious choice, and more one of card availability.
Or, to reframe that, there’s ample ground to optimize these decks by porting in more cards from Zendikar.
Aside from issues of whether a deck is the best or not, the other useful consequence of list-gazing of known deck archetypes, such as Jund, is in generating an overall understanding of everything that archetype can be. All Jund decks are not the same. Beyond choices in the mana base, I’ve noticed a few major points of differentation between Jund lists.
A default Jund list might look like this:
Default Jund (not recommended)
However, lists in the past month have also included Resounding Thunder, Slave of Bolas, Burst Lightning, Chandra Nalaar, Liliana Vess, and Sign in Blood. Rather than attempt to generate one, Karsten-esque “average Jund build” for your testing gauntlet, I recommend pawing through this recent list of decks and trying to gather a list from each of these variations. Does Jund with Thunder play out differently than Jund without? I’m not sure yet, but you can be sure I’ll include both in my gauntlet for the coming week.
I have no novel deck options of my own this week – I’m considering control again, and expect to return to thinking about it next week. In the meantime, however, I’d like to highlight a couple cool lists from the most recent Decks of the Week. All competitive intelligence purposes aside, I really do enjoy just finding cool ideas, so here are a handful:
Cradle Aggro (by Yemeth)
Yemeth has done reasonably well with this G/W fattie-based aggro deck that splashes red for Uril in the main deck and Pyroclasm and Manabarbs in the side. I’ve somewhat facetiously named it after Cradle of Vitality, a card that has appeared in every listed performance of this deck, and which I’ll want to test in action so I can figure out why it’s even there. This deck attacks Jund with a combination of high toughness, Troll Shroud, and life gain, and looks like it would be a great deal of fun to play.
Ascension Walker (by SamuelBeckett)
This build caught my eye because it deploys an idea I’d been considering for control – Kiss of the Amesha. A little while back, Kiss felt like a poor man’s Primal Command. In the current Standard, it seems like a decent late-game option for control decks, giving you the same massive one-up that Primal Command managed combined with slightly more, if less focused, card advantage. Is it the best choice? I think that’s worth testing to find out.
Grixis Control (by firstshot)
Yes, four copies of Cruel Ultimatum. That’s almost all I have to say about this design, other than to point out that it has no sweepers in the main deck. This may well be the right choice in an environment where people are running fat creatures in an attempt to beat Jund. Why waste space (and screw up your colors) by trying to pack in Day of Judgment when you can just one-for-one them with Terminate, Wretched Banquet, and Cruel Ultimatum?
There are a number of other cool new deck concepts in the sea of data from this most recent Decks of the Week. In suggesting you go take a look, I’ll remind you one more time that while you’re taking in the cool stuff, you should try not to just glaze over as you’re paging past Jund list after Jund list. Take a close look at what they’re doing. Are they shifting over time? Can you find some kind of optimal Jund list buried in all that information? At the very least, we want to come away with a sort of comprehensive image of the likely constitution of any random Jund deck we run into, and the easiest way to generate that image is to actually slow down and look at a bunch of individual Jund lists.
Taking down Suffolk
Last week, Josh Tovell of the U.K. contacted me to ask for some words of advice on deploying Nayamorphic in defense of his title as Suffolk County Champion (the equivalent of a States championship in the U.S.). With an eye toward a metagame that he thought was likely to include rather more Jund and “Edward Cullen aggro” and less in the way of draw-go or tapout control, he ended up opting to leave the Ascensions behind and run [card]Celestial Purge[/card]s instead.
On Sunday, Josh wrote me to let me know that he’d taken down the Suffolk Championship again, making him the first person to win that title twice. He wrote a bit about his experience with the deck, and I wanted to end today’s article by excerpting some of that, and by pointing out that Josh’s willingness to tweak the deck (in this case, the sideboard) to match his expected local metagame is exactly the kind of thing we want to do if we’re going to be as successful as we’d like to be. Do take note of his suggestion about adding another fetch – it’s something that’s worth considering.
Here are some of Josh’s thoughts from his victory:
The deck ran like a dream all day, I can safely say it is one of the best decks I’ve played for a while. The “Reach” package is very well named; it puts the game within reach of winning. Elspeth won me about fifty percent of games, being able to angelic blessing a Thoctar or Knight is absolutely crushing. I was asked all day whether “Is that type 2 Rubin Zoo?” I dispelled these thoughts all day and explained you had conceived it prior to Austin.
I don’t know how much you’ve toyed around with the fetch land quantities but I think cutting a Forest for a Misty Rainforest would work well, however this is just a thought I had after playing the deck.
I played Soldiers in the quarters; the matchup was harder than I anticipated. The early game was mine, however late game spells like Conqueror’s Pledge and Captain of the Watch left me a little distressed, however I managed to burn him out thanks to the early game damage.
My semis and final matchup’s were a dream, Vampires and Vampires. My sideboard Celestial Purges were working overtime. The semis were a breeze; Vines of Vastwood saved my arse on more than one occasion (that is amazing sideboard tech Alex).
I swept the first game in the finals but was staring down defeat in the second. My board was Elspeth, a bunch of lands, Knight of the Reliquary (currently 5/5), and Bloodbraid Elf. My opponent had calculated the win, attempted to Tendrils my Bloodbraid Elf and alpha strike with his whole team. Luckily I had the Vines to negate his Tendrils; however he swung with his team. I chumped with Bloodbraid and dropped to 1, he was at 13. I was holding the second Vines and drew Terramorphic Expanse. I promptly played and sacrificed it, bumping my Knight to 6/6. Elspeth blessed him and a kicked Vines sent him into the red zone for 13 damage; exactly lethal.
Congratulations to Josh on his successful title defense.