Well, the Pro Tour results are in, and boy were they not what anyone expected. The archetypes may have been known beforehand if you’d been keeping up with the Magic Online metagame, but I doubt that anyone would have predicted zero copies of Mardu in the Top 8 and an average-at-best record for the archetype. If not for B/W Zombies’ sideboard, we would have finally had a Top 8 without Gideon!
While maybe not the 6-different-deck platter people demand to see now, we had three distinct archetypes and a handful of variants among them. Meanwhile, if you look at some of the decks that finished 7-3 or better in Constructed, you find an array of options. Well… Unless you enjoy playing blue control. Turns out nobody has quite cracked that nut yet.
These records save Mardu a bit from the laugh riot that was the Top 8, but nonetheless, Mardu drastically underperformed as the most well-practiced and refined deck in the field. There were a number of pros who could transfer all their playtime with Mardu for the past couple of months to the new versions of Mardu. Of course, practice and experience doesn’t necessarily mean results. Pro Tour Aether Revolt champion Esper Berthoud jammed an interesting build of Mardu with throwback Inventor’s Apprentice, but only managed a pair of match wins before his day was over.
People will likely overstate the demise of Mardu Vehicles without taking into account that the real world won’t respond as severely. Just as Magic Online can deviate from paper, most tournament metagames will be manageable for a Mardu pilot. We can, however, glean some information based on the composition of the winning decks to determine what went wrong.
The first problem was how badly Mardu lined up against Marvel. That barely over 40% of the highest levels were stock Mardu lists makes it feel like the deck was getting shellacked by Marvel. The biggest takeaway from that is to recognize how much time Mardu gives the Marvel player. The curve-out draws from the past are mostly gone and reserved for the very best hands. Meanwhile, all the Marvel decks got more early game interaction, and turn-4 Ulamog still just ends the game.
This wouldn’t be so bad if you still had a winning plan against other aggro decks—but you don’t. Mono-Black Zombies had a 58.8% win rate, and that was with the midrange planeswalker sideboard package commonplace. Zombies had too many threats and too much resilience to punish them without chaining sweepers or the Zombies players stumbling—See the massive win percentage differential for B/W Zombies. When Zombies curves out with Dark Salvation and Liliana’s Mastery, there’s very little you can do.
A recurring theme of the event was how badly positioned Unlicensed Disintegration looked. Taking out one Zombie was hardly worthwhile, and none of the Aetherworks Marvel decks bat an eye at the card. All those sweet creatures you once wanted to hurl Murders at ceased to exist outside of a handful of B/G players. Fatal Push also had its own struggles, but at 1 mana it remains a reasonable play. It just didn’t feel worth it to tap out for a Disintegration anymore. Once you start questioning that spell, the whole archetype unravels, and you start asking why black is in the deck at all.
Stop and ponder for a moment: What does black give you if Disintegration simply didn’t exist?
Fatal Push: Good but not amazing right now, and could be adequately replaced by Magma Spray. In fact, the Mardu mirror is the best reason to keep it.
Cut // Ribbons: This card may be a cut above, but not by a huge margin, and these decks all generally win before you get a lot of value out of the Ribbons portion of the card. Declaration in Stone or Harnessed Lightning can fill the 2-drop removal slot.
Sorin, Grim Nemesis: This card is just bad now. Run Chandra, Flamecaller for extra sweepers against Zombies even if you keep the black.
Transgress the Mind: Most of the time you’d rather curve out against Marvel over casting a discard spell, and almost all the other matchups in which you’d want this card have disappeared or are already good matchups.
Scrapheap Scrounger’s ability: Ding ding ding, we have a winner.
This is where I’d start—a throwback to Kaladesh W/R Vehicles. Keeping a small amount of black in the deck doesn’t harm much and helps you utilize Radiant Flames against Zombies. Against Marvel, you keep an aggressive slant and can board in Kari Zev’s Expertise with a realistic shot of casting it post-Ulamog even if they nuke one red producing land.
Of course, the original Mardu base is still functional, and may just need to go back to blue for a fighting shot against Marvel. Andrea Mengucci made a case for that approach, and others will make similar changes. Obviously I’m not in agreement with the best approach to the Zombies matchup, and I think it’ll be interesting to see how the sweeper vs. non-sweeper board plan debates play out.
I think people are going to read this section and mistakenly believe I’m saying Mardu is dead. Mardu did badly at the Pro Tour because it was badly positioned and many of the builds were designed to succeed in a metagame that no longer existed. Sticking with the traditional Mardu midrange base is a mistake, as other decks simply overpower it or go too wide for it. Mardu can and will adapt and be a legitimate choice moving forward, so don’t lie to yourself when choosing a deck.
The other big known entity was Aetherworks Marvel. People expected it to take the place of 4c Saheeli as the new fun police. Based on the PT results, it sure lived up to that lofty title, spawning the most successful archetype at the PT, Flamecaller Marvel.
Both Team Musashi and Team Genesis have been among the best overall in the Constructed portion of Pro Tours this season. Both of them brought altered Temur Aetherworks decks that outperformed the stock lists, but I’m choosing to highlight the Genesis build because it’s a radical departure from what we’ve seen so far.
This is the step forward I was trying to make with my build last week, but rather than Glorybringer, the missing piece was the bigger Chandra. Flamecaller does a lot here, it provides the deck with a big play when it doesn’t see an Aetherworks Marvel and another sweeper. With the popularity of Zombies at the Pro Tour, many people are going to make this swap. By running the full playset you effectively double your relevant Marvel spins, since wiping Zombies’ board while keeping a planeswalker around is one of the best ways to beat them.
Now, I said that Chandra effectively doubles the good hits, but that’s not quite accurate. This build omits the 4th Ulamog from the 75, which is surprising. There are two ways to look at this cut: The first is that you’ve reduced the number of virtual mulligans you take when you draw Ulamog in your opener or pretty much any time before turns 8-10. The second is that you’ve made your Marvel spins that much weaker in a deck already plagued by variance. Ulamog also happens to be the best card in the mirror and cutting even one makes an impact, especially once post-board countermagic becomes the primary plan.
I’d cut Servant of the Conduit for the 4th Ulamog, a pair of Sweltering Suns and a Dissenter’s Deliverance or Baral’s Expertise. While this build has a much better reason to run Servant, my reservations remain. Unless you’re powering out Chandra a turn earlier, most of the time you don’t need the ramp and Servant is just another card that says make two energy. If you’re worried about losing some percentage to decks where you need to race, the Suns should mostly accomplish that. You can also swap Aether Meltdown for Magma Spray without a major loss if Mardu is displaced by Zombies.
While I’m not 100% sold on the total lack of card draw, it’s configured much more closely to that ideal with the extra threats the Flamecallers add to the deck. Ultimately, if you were going to take a deck that did well at the Pro Tour and start practicing and tweaking, you can’t really go wrong with this one.