Every spoiler season is another opportunity to relearn what is actually good in Modern. There are so many new cards entering that, even when the broader metagame doesn't measurably change, specific matchups can be reshuffled thanks to new sideboard cards, role-players, and even lands. Also, despite years of trying, nobody has ever accurately predicted what cards will actually have an impact on every format. Testing often reveals unexpected additions, which is what happened to me this week as I tried a couple of March of the Machine cards in Modern, including a battle!

Testing in Gobakhan

As I mentioned last week, the battles were quite underwhelming. The only one I thought might actually see play as intended was Invasion of Gobakhan // Lightshield Array, mainly thanks to its cheap cost and low defense. The fact that the back synergizes with attacking battles was a nice bonus, but I wasn't expecting too much from the card. That said, it's a natural fit into Modern decks I like, so of course I tested it.

To begin, I grabbed a Mono-White Humans list I saw while gathering data for the metagame update and moved some numbers around to fit in Invasion. I'd also like to take this opportunity to thank the players working on this deck for cutting Chancellor of the Annex. A card that you'll never cast, can use only if it's in your opening hand, and whose primary purpose was to be pitched was never going to work out. Shining Shoal, on the other hand, was a great card to rediscover.

Confusing Discoveries

While I always try to keep an open mind, I was very skeptical of Invasion going in. Just because I said it was the most plausible didn't mean I thought it would actually be successful. However, a very strange thing happened. Invasion was good; playably good. Maybe even meta-shakingly good. That wasn't even factoring in Lightshield Array; the front side by itself was shockingly playable.

Cheap disruption is always playable, especially when it's a disruption effect that a color doesn't often get. White doesn't get direct hand disruption often, and Elite Spellbinder is just a little too slow for Modern. Wrenn and Six existing doesn't help, but regardless, Spellbinder always feels a half-turn too slow in the current metagame. Invasion being cheaper for the same ability means that the disruption is a bit more timely, and thus more powerful.

Lost the Beat

One aspect of Invasion's power has been players forgetting they can cast the exiled spell. Unlike Spell Queller- or Banisher Priest-type effects, opponents don't get the card back if Spellbinder or Invasion die. Thus, the spell is simply exiled, rather than put under them, and it's easy to forget that it's special as the game goes on. This doesn't really impact playability in a vacuum, but it is worth mentioning, especially since the card is new.

Instead, what I didn't appreciate (because it had never come up before) was how disruptive it was to remove a spell from a hand. That Thoughtseize is disruptive is well-known, but Invasion lets them cast the spell down the line. If it was straight-up Castigate there'd have been nothing to discover. However, even if the spell is cast down the line, often times it's lost a lot of its bite by then.

Sometimes, it's simply a case of an opening hand needing to play certain spells in a certain order to win. Take a link from the chain and the whole thing falls. Others, it's simply that a spell isn't good at the new mana cost. Expressive Iteration is quite mediocre, almost aggressively so, at four mana instead of two. It's like throwing a dance off by a few beats.

Compare with Thalia

The other unexpected result of my testing was how Invasion works alongside Thalia, Guardian of Thraben. Based on Pioneer experience with Spellbinder, I expected Thalia to be the main disrupter, with Invasion to simply be clean-up. As it turned out, the impact of Spellbinding something at two mana is very different than at three.

Thalia introduces inefficiency into a noncreature-based deck's gameplan. To use engines as an analogy, Thalia removes the grease from the gears. It's damaging, but might not be a killer. Thoughtseize is like throwing a wrench in the gears. It might shut everything down, or it might just get spit out of the machine. Spellbinder was like sticking a screwdriver in a critical gear after Thalia removed the grease.

Invasion is more like throwing a log of wood into the gears. It will rattle the engine and gum it up, but a robust one will eventually work it out. That's worse than Seize, but better than Spellbinder. There was a time that Kitesail Freebooter was a staple in Humans (A-Unholy Heat proved its final straw), and it filled a similar role. In the current metagame, Invasion being a noncreature spell and thus immune to Heat and Push was almost an advantage, making this effect much more lasting.

About the Array

It occurs to me that I've spent all this article on the surprisingly good frontside, and have been ignoring the back. In my initial evaluation, I thought that Lightshield Array was the big upside that would actually make the card playable. After testing, I'm cooler on Array.

Don't get me wrong, a few extra Thalia's Lieutenant triggers go a long way to winning a game, and may well put the game away against any fair deck. However, it wasn't especially common to have many creatures left to receive counters, and rarer to have the Array around. The sacrifice ability was frequently invaluable, and as testing progressed, opponents frequently tried to get me to pop it as quickly as possible. Trading Array for Akroma, Angel of Fury is a pretty good deal.

Bottom Line

I am very surprised at how much better Invasion of Gobakhan is and how medium Lightshield Array has turned out. That said, I'm not sure that Humans was the best test platform. The lack of synergy with the rest of the deck and being impacted by Thalia were major sticking points. I'm currently testing it in a Death and Taxes-style deck, where netting multiple triggers thanks to Flickerwisp is quite appealing. But the bottom line is that Invasion is much more playable in Modern than anticipated.

Diet Hogaak

Remember the early days of Hogaak, Arisen Necropolis? How the deck was a mill-combo deck first, and a graveyard beatdown deck second? I've been testing Kroxa and Kunoros with Altar of Dementia, and while it is nowhere near as good as Hogaak, it certainly has shades of that deck.

For some context, I see a lot of different decks gathering metagame data. A lot of decks are trying to cheat Atraxa, Grand Unifier into play, and the more glass cannon-y versions are gaining ground. While other sites list them as Reanimator decks, they're much closer to Legacy Tin Fins than Reanimator. It's about getting a big legendary threat out cheaply, gaining value, and then letting it die rather than keeping it around. This seemed like a natural fit for Altar as a value play, so I'd just need to add Kroxa and Thassa's Oracle

Combo Kroxa

This was actually a harder fit than I thought. There are two versions of turbo Atraxa, one featuring Ephemerate and the other, The Underworld Cookbook. The former is more robust, but the latter makes more sense for my purpose, so I modified it and started testing.

Between Profane Tutor and Atraxa, I didn't feel like I needed more than one of each combo piece. I can't cast Oracle or Kroxa, but the deck was intended as a proof of concept. I wasn't planning on playing drawn-out matches where I needed to. I just wanted to see whether the combo fit in.

I Forgot Something

The short answer was kinda. Altar fit in quite well, anyway. Not casting legends is what the Atraxa decks are about, so the mana wasn't an issue. Even without going for the combo, getting an Atraxa trigger, attacking, and then sacrificing Atraxa to Altar usually set up the next Goryo's Vengeance perfectly, making that Altar worth the card. Thanks to Profane's searches and Atraxa's draws, finding Altar was less of a problem than anticipated despite being a one-of, and this quickly set up a self-sustaining cycle of deadly value. Not having to exile Atraxa was nice value, too.

That said, adding the combo made the deck significantly worse. The combo wasn't bad, but it was only sometimes necessary. Vengeance for Atraxa and repeat was usually good enough. However, the biggest problem was Emrakul, the Aeons Torn. I forgot that when Emrakul is milled it shuffles the graveyard back into the deck. This made it impossible to actually combo off. Whoops, time to try again.

Kroxa 2.0

For the second try, I removed Emrakul, and added some more Altars and a Kroxa and another discard outlet. I wasn't interested in straying too far from accepted norms on the deck.

While this deck wasn't capable of the free-ish wins that Emrakul can deliver, it did work closer to my intentions. Alter filling the graveyard off Atraxa ensured that the combo could happen, and the deck got thin enough often enough that it was rare to not find a Kroxa if I wanted to try and combo. Each cycle of Vengeance and sacrifice saw 17 cards, after all.

The lingering question was whether this was better than the standard plan of just bashing with Atraxas or Emrakul. I'm not sure I have an answer. This deck is more vulnerable to graveyard hate like Abnormal Endurance but less vulnerable to City of Solitude or Ensnaring Bridge. The nuance leads me to believe that there is real potential here. It is more waiting for the right metagame than not being viable.

Bottom Line

Assuming they aren't already, Vengeance players should seriously consider retooling their deck around Altar. There's a lot of value to be had there, and even without the combo, the option to just mill out the opponent rather than exile the legends is attractive. The combo fits in without problem, but it felt like a Plan C and might not be necessary.

Expect the Unexpected

Testing only rarely confirms suspicions. Magic is so complex that there's always something new to discover, and some things are more powerful than they appear. My results from March of the Machine testing aren't what I expected. How are you all doing? If you've found something surprising, drop it in the comments!