Shifting Metagame: SCG Charlotte Analysis

You may not have heard, but there’s a Modern Pro Tour coming up. This is leading to an influx of interest in our beloved format and increased tournament support. The past weekend had an SCG Modern Open in Charlotte; next weekend there’s another in Cincinnati, SCG Regionals are in November, and then the SCG Invitational and GP Oklahoma City in December. That’s moving in the direction of a reasonable data set. Therefore, it’s time to scrutinize the metagame again.

We’ve been over this before, but Wizards’ decision to reduce and regulate what MTGO results get published hampered our ability to analyze the metagame. Selectively picking five different decks rather than the random ten they previously released ruined MTGO as a data source. Curated samples reveal the biases of the curator, while random samples more accurately reflects the population. If every data point has the same chance of being chosen, then the final data set will proportionately represent the population. This no longer being the case for MTGO, we have to use paper events and preferably big ones to get a decent sample size. These don’t happen often so it has become harder to accurately model the metagame. Just like Wizards intended.

However, when there are a number of big events close together, as there will be between the SCG Opens and Classics, we will get enough data points from random samples that we can in fact do some analysis. So over the next few months I’ll be putting these big event results together and attempting to project the Modern Pro Tour metagame. It’s hard to say at this point how well it will reflect the “real” metagame, but that’s what research and experimentation are for, aren’t they? My plan for this week to start putting together the data set and see how things look. As more big events take place I will add them in and start to evaluate the trend. Hopefully patterns of success will emerge and from those I can make conclusions about the state of the Modern metagame.

A Colorless World

I’m starting with the broadest data available, the Day 2 Metagame Breakdown. Lacking Day 1 data, this is the total population for our next steps. There are a lot of decks, but I want to highlight the upper quarter of the data.

Deck Total
Eldrazi Tron15
Affinity13
Jeskai Control10
Burn9
UR Storm7
Grixis Death's Shadow6
Jund6
Abzan5
Elves4
GR Ponza4

That is a lot of colorless decks. Affinity and Etron represent 35% the table and 22% of the total Day 2 field. Etron being popular is not exactly surprising; it’s been doing well for months. Mana acceleration coupled with big, disruptive threats is a fine strategy, especially when you have a good matchup against the format’s boogeyman. Affinity had fallen off over the summer, so it’s made a pretty surprising resurgence here.

The Affinity surge may help explain the return of Jeskai Control. Largely replaced by UW during Death’s Shadow‘s heyday a few months back, Jeskai boasted a better matchup against both Affinity and Counters Company and has been steadily climbing in popularity. I do wonder how many of the decks listed as Jeskai Control are actually tempo decks, but that changes nothing. Bolt-Snap-Bolt decks are making a return.

The really interesting part, at least to me, is at the bottom of the table. Grixis Death’s Shadow, after spending months as Modern’s “best deck,” only put six copies into Day 2. This is the same as Jund, the deck it theoretically replaced, with midrange Abzan just behind them. This suggests that the metagame warp I noted months ago is weakening. Eldrazi Tron and GDS are powerful, but far from unbeatable.

The problem with both is that they have lots of air. Etron has lots of lands, land search, and mana rocks. GDS plays the most efficient threats, removal, and disruption, but it doesn’t play a lot of them. It’s held together by cantrips. Both decks have a tendency to do nothing if their initial hand isn’t good enough to win. BGx is a pile of the best cards that are relevant the entire game. They even have lots of creature-lands so their lands remain live draws. As counter-strategies emerged and the novelty wore off, I’m not surprised that players went back to the old reliable standbys. From that overall population, what actually rose to the top?

Charlotte Open Top 16

Storm won Charlotte, though given that the finals were against Affinity this isn’t too surprising. Fast combo is traditionally advantaged over creature decks, though Affinity was hampered by some subpar draws. What really surprised me was the semifinal game against Grixis Death’s Shadow. The matchup is very favorable for GDS and yet it just couldn’t close the game. The fundamental flaw of the deck I mentioned above was on full display, as Andrew Jessup had plenty of early disruption but not the early threats to close the game before Paul Mueller recovered.

However, all that is less important than the overall view of the metagame that the tournament gave us. I’m just using the Top 16. The Top 32 is nice and all even if the prizes are a bit weak, but when trying to find the best decks in the tournament we should pick the best results. It would also make the table prohibitively huge.

DeckTotal
Storm2
Affinity2
Grixis Death's Shadow2
GW Company1
GR Ponza1
UG Merfolk1
Breach Titan1
Infect1
Abzan1
Jeskai Control1
GB Tron1
BW Eldrazi1
8-Rack1

Not a single Eldrazi Tron. Out of fifteen decks, nobody made Top 16. One did squeak into Top 32, but only just at 31st. That is a remarkable failure to convert. Jeskai also had disappointing results. Of the previous list toppers, only Affinity remains, both representatives in the Top 8 no less. Exactly what all this means is hard to determine. We don’t have enough data yet to really make any determinations, and if we just go by conversion rates, then UG Merfolk is the best deck with 100% Day 2/Top 8 conversion. We’re just looking for a place to start—if these trends continue that’s something to investigate.

GDS placing two copies into Top 16, even just barely, does still speak to its power. The observed drop-off in popularity and failure to win events may have taken it down a peg, but it’s still a powerful and terrifyingly efficient deck. Don’t sit on Storm either. Combo is back. And far faster than it used to be. Ad Nauseam was the only real combo deck for years because of its resistance to disruption relative to Storm. It appears that isn’t good enough anymore and speed is ruling the combo world. Be aware and be ready.

Charlotte Classic Top 16

The Open results are nice, but 16 data points just won’t get us anywhere. The Classic may not have been as harsh a crucible, but it’s still a large tournament and yields far more valuable data than four-round Modern Leagues or a PPTQ. Therefore I’m going to bring in the Classic Top 16 as well.

DeckTotal
Jeskai Control3
Affinity2
Storm2
Temur Aggro1
GR Scapeshift1
Mardu1
Saheeli Evolution1
Jund1
Knight Company1
UW Control1
Abzan1
Grixis Death's Shadow1

That’s very interesting. Affinity and Storm placed two copies again. As above, it doesn’t mean anything yet, but it is interesting. This marks them as the decks to watch. Placing well in two events speaks well of their power. However, this is the Classic accompanying a Modern Open. The player base of such events tend to be players from the Open who didn’t make Day 2. As a result, you tend to see similar decks. However, we do assume that the best decks win more, so that could also be an indication of their place in the metagame. We need to wait until next week to see.

The other interesting thing, from a metagame analysis standpoint, is Jeskai’s outstanding result. Representing 19% of the Top 16 and winning the event is nothing to dismiss. Or it looks that way. This is one of those cases where decks that aren’t really the same are being lumped together. Two decks, including the winner, are actually Jeskai Tempo decks with Geist of Saint Traft and Spell Queller. They assume the control role most of the time, but they’re not real control decks. The other is a true Nahiri, the Harbinger control deck in the classic model. It’s a fine enough distinction that I won’t split hairs, especially since it’s not universally accepted as a distinction.

There are a lot of fringe and rogue decks in this Top 16, but you shouldn’t read too much into them. It was a relatively small event compared to the Open, and this gives weird brews a better chance to spike the event. As I’ve covered before, the smaller the dataset the more likely it is that outliers will affect the data.

Charlotte Metagame

That’s nice and all, but how is any of this helpful? By itself it isn’t. It’s when we put the data together that the overall metagame picture will start to emerge. We’re not to the point where we can really make valid arguments about the metagame, but we can see the relative strengths of the decks at SCG Charlotte. With the data from Cincinnati next week and more in coming weeks, we will build a far more accurate picture. Eventually we will have enough to actually draw conclusions about the Modern metagame.

DeckTotal
Storm4
Affinity4
Jeskai Control4
Grixis Death's Shadow3
Abzan2
UG Merfolk1
Breach Titan1
Infect1
Jund1
Saheeli Evolution1
Mardu1
BW Eldrazi1
8-Rack1
Temur Aggro1
GR Scapeshift1
UW Control1
Knight Company1
GW Company1
GR Ponza1
GB Tron1

You could probably see this table coming. Storm, Affinity, and Jeskai decks had the best weekend, accounting for 37.5% of the total Top 16 finishes. If you wanted to do well, these decks gave you the best odds in Charlotte. If they continue to do well this will indicate high-tier status. Following them are GDS, Abzan, and then all the singletons. These individual finishes aren’t statistically meaningful and if they remain alone after the Washington DC Classic I will be trimming them from the table. They’re just meaningful to the rest of the data. What you should take away is the affirmation of the old saw, “Play your deck.” The deck you know and love above all others is still the best choice for you at any event. It may not have staying power of the tier decks, but that doesn’t matter if you win.

The other thing to conclude from this table is that Etron was a very poor choice in Charlotte. Whether this was because of weaknesses in the deck or a very hostile metagame is impossible to say. If this continues, a reexamination of the deck and its place will be necessary. Who cares if it grinds results well on MTGO if it falls apart on the bigish stage? Of course, having said that, it’s pretty much guaranteed to come roaring back next week. Of course, it could all be a double-bluff, but at that point I’m just chasing my tail and need to abruptly transition to the next section.

What Does It Mean?

Nothing in terms of the actual metagame. But as a result there is quite a bit to investigate. I keep coming back to the failure of Etron in Charlotte, but given the results, perhaps that isn’t surprising. Etron feeds on midrange decks and Chalice-soft decks. There isn’t much midrange present here, and most decks have adapted to Chalice enough that it isn’t game-ending. Even GDS can push through a Chalice, even if two can be problematic.

This may help explain why Storm did well. The only spaghetti-monster that Storm really cares about is Thought-Knot Seer, and that’s not as intimidating as you might think. Chalice is also not that effective on its own since the cantrips aren’t too important. Chalice for one only hurts if you’re struggling to go off, while Chalice on two is devastating. But that doesn’t happen often before turn four, giving Storm plenty of time to just win. The Eldrazi’s lack of interactive spells really hurts. I’d guess that Storm fed off slower combo decks and Etron on its way to the top.

Affinity and Jeskai being the other two big decks is interesting because Jeskai has such a good matchup against Affinity. The deck is mostly removal and unless Affinity can Cranial Plating an Etched Champion it won’t fight through all of it. This isn’t even getting into the sideboard cards. I’m guessing the Affinity players dodged a lot on their way up the brackets. These decks also probably explain Burn’s drop. Lightning Helix and Vault Skirge are bad news for red decks. This isn’t even considering Affinity’s speed advantage or the counters from Jeskai. Neither is exactly cold to Etron or Storm either.

Based on these results, I think the top decks were the right choice at the right time given what they faced. Next week I will see if they had any staying power, and that’s when the real analysis will get started. See you then!

David began playing Magic during Odyssey block, quit playing Magic when Caw Blade ruled the world, and returned to Modern shortly before Deathrite was banned. He’s made an appearance at the Pro Tour, made money at GP Denver, and is constantly grinding and brewing in Modern.

11 thoughts on “Shifting Metagame: SCG Charlotte Analysis

  1. Good article, the best in a while on this quite in all honesty, very glad to see more metagame analysis back on Nexus, and definitely looking forward to the next few weeks where the dataset will grow and give us our first clear picture of the actual metagame in a long time.

    Speaking of metagame, shouldn’t the “current meta” sidebar be removed from the front page? If it’s impossible to update now that’s one thing, but leaving it up after it’s 6 months out of date feels a bit useless.

    Lastly, to address (partially) your question about E-tron and whether the meta in Charlotte was too hostile, consider that Affinity was the second most represented deck and it CRUSHES E-tron. The match-up for Affinity is about as positive as against Mono-U Merfolk, so that definitely hurt E-Tron’s conversion rate.

      1. I didn’t want to phrase it that way, but yeah, it does. If it has to stay until it gets updated, maybe we could have a little fun with it? For example, you could change the title to “Current metagame (from 6 months ago)” or “Current metagame. Just kidding”. Or have fun with the archetypes, throw in some spicy ones and unheard of combinations:

        Pirate Stompy
        Death’s Shadow Tron
        Affinity Control
        UB Valakut
        Ground Seal Combo

        On a more serious note, if you intend to update it, I’m curious to know, how? After this site stopped updating the metagame other sites took over for a while, but ever since WotC cut down on published results, it’s been radio silence everywhere. How do you intend to cobble together enough meaningful data to update the list, other than sporadically in periods with lots of events, such as the weeks to come? Will the updates ever become regular again? When can we expect an update?

        1. First, let me say I have deliberately kept mum about this as I don’t want to make promises I can’t keep. I’m sort of a one-man team when it comes to the metagame updates, and the work load is… sort of breathtaking.

          We actually only relied on MTGO data for a section of our analysis. The rest of it came from a bunch of different sites (MTG Top 8, MTGdecks, etc.), which reported results from paper events. Part of the reason our metagame analysis always looked different (and usually more nuanced) than other sites’ is because we gathered ALL the decks that were available publicly, and assigned different weights to different types of events.

          So, while the change in MTGO reporting definitely hurts, it doesn’t make that impossible. I will probably assign slightly less weight to the MTGO data points, and play around with it to see what makes sense intuitively. The biggest obstacle at this point is much more mundane—I have to program some code that will automate inputting all that tournament data, as I can’t sustainably do it every month/week by hand. Unfortunately I don’t have a clear time frame on that right now.

  2. I recognise that it’s a common human trait to cherry-pick aspects of their environment a weave a personal narrative of blame and doom, but I’m surprised to see it in effect so clearly at the moment surrounding Modern.

    Your article is refreshing, because it doesn’t pander to this at all (and why should it!). Simple unbiased analysis, as much as it’s possible to do so.

    Yet outside of this protected bubble we see a clamouring vocal minority crying foul and waxing lyrical about the death of modern and how ‘broken’ eldrazi and titanshift are. These two decks (as opposed to, say, affinity which is consistently one of the best things to do in the format) seem to draw more ire than I can previously remember in modern. Cries for bans and of the death of modern seem rife, and I’m genuinely confused by the lack of any sort of evidence to support these viewpoints.

    However, I can and do recognise that in our imperfect world of magic, many players will bring a casual or suboptimal pile to FNM or their local playgroups (or larger events) and those ‘big mana’ decks often seem the most brutal and merciless when you’re not expecting to see them and are piloting (for example) BW vampire tribal with Felidar Sovereign as a win condition. Also, it’s a provable phenomenon (for many reasons including cost) that people are resistant to changing their decks, sideboards to respond to a metagame shift. Perhaps these are underlying causes of the rhetoric I’m seeing plastered about on the Internet?

    Apologies for the tangential comment! I just felt like somehow it meshed in with the early metagame insight that we’re getting at this point from big events starting to flow in.

    Bring on the ProTour 🙂

    1. Also if I may gush slightly; this article is the kind of article I come here for. To the point, data-driven and a great basis for open discussion on the format, with some interesting points thrown our way as well. It’s helpful, insightful and once a few decklists are thrown in, we’re talking about a properly decent bit of writing with substance and analysis.

      Wizards have made it harder for you guys, and I sympathise. Is there room for you to apply this sort of methodology to something a bit more regional? There are numerous 100+ player events happening all the time, and while they aren’t “all of modern” they are at least representative of the format.

      Thanks!

      1. The problem with that kind of data mining is ignorance. I’m sure that you’re right about these events, but I don’t know where they are or who’s running them. I frequently don’t hear about events in my own area until right before. To get that data we have to know about the event in advance and then find the lists, which most don’t publish and they may not be willing to give us even if asked. Could it be done, certainly. Is it feasible? I’m not convinced.

        We’re looking at various solutions and want to bring back the metagame updates, but we keep having this problem of finding enough usable data in a timely manner. Jason’s working on it.

        1. It’s possible perhaps to partner up with a store that does semi-regular large sanctioned modern events, and maybe even go as far as offering a little branding to the tournament.

          For instance, Birmingham in the UK is the site of Manaleak, which consistently holds the largest regular modern tournaments here. Less frequently we have the MagicMadhouse large tournaments as well, one of which is modern.

          There will be equivalent places in the US, as well as Japan, France etc etc etc.

          Who knows, it’s possible however unlikely that you guys could broker a relationship where they provide the list of decks from a tournament and then maybe a top-8 decklist trawl.

          Maybe? Haha 🙂

      2. Glad to hear this piece captured your interest. Right now my priority is getting the metagame database fully automated so that I can do updates without hours of data entry. We hope to be bringing you more “big picture” stuff on Modern soon.

  3. Good article! I’m very excited to finally see some form of analysis article once again.

    Jason, I know that you’re working on the metagame updates. I hope it’s a priority; I feel like the majority of the people who used this site in the past primarily used it for metagame data.

    1. It is a priority. I’m just trying to create something sustainable that won’t require constant upkeep, which is something of a project. The actual writing of the article each month was the easy part.

Leave a Reply