Pure Khans of Tarkir Limited is coming to a close, and I haven't felt this confident in a Limited format in a long time. No, it isn't just because Five Color is just so good, though it really is.
A big part of this is that the format distills down to a couple core concepts. I'm best when I can focus my decisions around a more linear gameplan, and in Khans, those gameplans apply well on a ton of levels.
One of the biggest tenets is that Khans of Tarkir is a blocking format.
If you are trying to attack and your opponent can block, you are likely doing something wrong. If you have a problem, 90% of the time it can be solved by having the right blocker. If your deck isn't set up to kill them immediately, it better be set up to win a long blocking game.
We are coming right off the back of a pair of attacking formats. Being defensive in Theros was pretty much the worst, and it wasn't great in Magic 2015 either.
What factors divide formats into these categories?
Holding the Line
What are the requirements of a blocking format?
Being the player blocking has to be profitable.
Okay, beyond the obvious. What are some ways to accomplish this?
1) One blocker has to be able to hold off multiple attackers.
Why is this effective? You have generated virtual card advantage. One of your cards is negating multiple of theirs, even if they are still in play. Your opponent cannot attack without immediately becoming behind on board, allowing you to trade two life to change the gamestate from four power versus three power in their favor to three power to two in your favor.
Khans of Tarkir has some of this going on. There's a fairly large jump in creature sizing around five mana to X/5 where you start seeing a Sagu Archers making it so a morph and an Unyielding Krumar can't attack. Your Glacial Stalker can easily make it so their four-drop and their five-drop can't attack. Boards bog down and become difficult to maneuver on the ground. Because a lot of these creatures are higher power, it also makes it difficult for something "massive" to get through. By that, I don't mean a 12/12 Rakshasa Vizier. I'm more referring Woolly Loxodon or Armament Corps, which despite being a size tier above the X/5 barrier, often gets double blocked and trades down for a 5/5 or similar.
2) There have to be incentives for extending the game.
In Magic 2015 Draft, players were definitely incentivized to attack.
Why was this?
Part of it was that creatures capable of holding off multiple attackers really didn't exist. You had Rotfeaster Maggot as defensive champion and that was it. But blocking wasn't Zendikar levels of terrible. You could actually kill attackers in combat, and you can definitely have formats where making combat exchanges up the curve is profitable.
The key was that getting to the phase of the game where you had traded off everything just didn't provide any benefits.
All of the non-rare game-breaking expensive cards had convoke, so they discouraged trading. In theory, you could trade and live to the point where you had seven mana to cast Siege Wurm, but more likely the fact you had a dead card in an earlygame of trading just meant you fell behind and died.
What about making up that dead card? Good luck. You had Divination, Sign in Blood, and a super expensive and poorly sized Looter in Research Assistant at common. Even the good "control" decks were only a little better than the average deck at topdecking going late unless they were full of rares.
What about Khans?
Treasure Cruise. Bitter Revelation (better than Divination as the selection also locks it to hit two spells). Bloodfire Mentor as a better sized and cheaper Research Assistant. Abomination of Gudul. Even something as innocuous as the mana efficient Tormenting Voice is a big deal. If you want to play a long game, being ahead on cards is easy.
But that's not even all of it. The multicolored format means very powerful uncommons and rares tend to ship through a little later than they should. This is why Five Color tends to dominate as a lategame deck, as your card quality is high enough relative to your opponent that even at spell parity you are ahead in the long run.
And finally we have morph. In this set, morph is a huge incentive to go late. Old morphs had a lot of their value tied into the surprise factor of them, and often their face up version wasn't much bigger than the 2/2 Grey Ogre mode of the card. In Khans, with the "five mana to flip and beat a 2/2 in combat" rule that has been enforced, you have a lot of high cost morphs with a high morph cost to match. Morph in Khans functions very similar to kicker in older formats, where you get to pay a low cost for a small creature or a high cost for a big one on the same card, only there's an extra bonus of being able to net overpay for the large creature if you change your mind later about the small one (i.e. 3 colorless plus 2BUG to unmorph Abomination of Gudul is 2 more than 3BUG to cast it). There's no chance of dying before you can use your card, and there's no chance of playing a card that is only active in the earlygame.
Khans of Tarkir really has a confluence of all the things required for making lategames profitable if you want to play for them.
3) It is mathematically impossible to get your opponent in range before they make some key jump.
This last one is mostly a relic of Limited formats long gone, but Magic 2014 brought it back fairly recently so it's worth touching on. This was also the style of Magic 2010 for a bit of an older reference.
Sometimes the creatures in a set are so bad at attacking that being aggressive isn't mathematically possible. If you play a 1/3 on turn 2, a 2/2 on turn 3, and a 2/4 on turn 4, don't be surprised when they just play 2/4 into 3/5 and you can't deal damage.
I really hope this never happens again. It leads to miserable games of Magic. Sorry, you should have taken sixteen from their flier instead of casting your Doom Blade on it, because eventually you could find your Giant Spider to answer it where as you needed to Doom Blade their Air Elemental. Oh yeah, and no other creature dealt damage that game.
I like the future, where things that happen on turn 3 force real and involved decisions that actually matter.
That said, Khans comes close here. If you are a defensive deck, you play most games on 23-24 life off the common gainlands. The creatures are big enough that doing absolutely nothing gets you killed, but the amount you have to push back to make this free life untouchable before six-drops start flying is shockingly small.
On the Offensive
So, what turns this around for the person on the attack?
1) It is mathematically impossible to stay out of range before hitting a key jump.
This is half of the Zendikar issue.
People died on turn 3 in that format. Not often, but enough that I've heard the story more than once. Steppe Lynxes, Plated Geopede, Goblin Bushwhacker off a fetchland. Turn 4 or 5 was more normal, not in the sense that every game ended on it, but more that if your bad draw lined up with their good one you had only that long to recover.
Zendikar definitely had Territorial Baloths, Vastwood Gorgers, and expensive kicker cards like Heartstabber Mosquito on the high end, but living until you cast one of those cards was not a given without significant extra work on the front end.
There are also scenarios where the burn or Falters available make it so even normally reasonable life totals just aren't safe. This is how a lot of the actually good aggro decks in Khans operate, using Barrage of Boulders to kill opponents from absurd life totals like 21. The solution to this is very analogous to how you handle Burn in Constructed. If you will eventually die to something regardless of what you do, kill them before they draw it. So you have formats where it's smaller decks racing then hoping to have their breaker versus slightly larger decks racing back with slightly larger threats, mostly skipping the "stabilize and get infinitely far ahead in resources" step people traditionally associate with being the control. Again in Constructed terms, the blocking decks are Faeries and not U/W Control.
2) Blocking is most often chump blocking or poor tempo.
This is the other half of the Zendikar issue.
Play a land. Trigger landfall on Steppe Lynx. Attack. Would you like to block my 2/3 with a 2/2? I certainly wasn't blocking your 2/2 with my 0/1 Steppe Lynx on your turn. The best defensive creature in Zendikar block was Grazing Gladehart, and it was actively better to not block with.
I loved Zendikar block and all of the subtle decisions that mattered when games were so fast and brutal, but most people didn't. As a result, we haven't seen anything quite as in your face along these lines. There's the occasional Borderland Marauder, but it's not usually a major theme of a set to make creatures better when attacking.
That doesn't mean this doesn't still happen in a lot of different ways.
In Theros Draft, creatures that attacked were massive. Bestow, heroic, and monstrous played into a theme of "build a giant threat," and either your threat was bigger than theirs (winning) or their threat was bigger than everything you had (losing). Rarely was there a case where it was right for both players to not attack.
Once Journey Into Nyx came out, the format changed, but it still remained an attacking format. The reason was that the tricks were just so efficient on mana that blocking was tempo negative. You play a three-drop and it likely can trade for their three-drop, but they get to attack with their creature, play a one mana trick, and play another three-drop. Your block was effectively trading down, and you got swarmed and died.
If every time you block you lose board presence, it isn't a gameplan; it's a tactic to use in a race.
How to Be Offensive in a Defensive Format
1) Find the combo deck.
Raphael Levy notably found a Fortify-based strategy at Grand Prix Prague last year in Magic 2014 Draft and rode it to a 1st seed after the Swiss in that event.
Seth Manfield won Grand Prix Ottawa last month with a deck heavily reliant on Goblin tokens from Hordeling Outburst and Ponyback Brigade plus Rush of Battle or Trumpet Blast to deal lethal damage in one shot.
If your opponent isn't really pressuring you, assembling some one-two punch lethal combo is right where you want to be.
2) Play the Skies deck.
You are going to lose ground against people still trying to be normally aggressive, but often the Skies deck is partially taking advantage of the blockers and lategame available to it. The classic in my mind is Horned Turtle, but Monastery Flock is the Khans example.
In reality, the Skies deck plays the same game as the control decks. The assumption is just that if you go slightly under and around their big threats you will win the fight.
3) Metagame and play the unplayables.
If everyone knows aggro is bad, it's often possible to pick up some cards after the point they really should be in the pack. If people don't get the joke and fight each other for the cards it, won't end well, but if one person decides to be that guy, they can build something that might actually work.
A good example of this from years ago is the Mono-Red deck in Shadowmoor Draft. If multiple people were trying to chain Intimidator Initiate activations, the well of "playables" would run dry really fast, but if one person at the table was on the deck, it was hard to beat. I've executed similar plans in Magic 2011 Draft with Giant Growths and Garruk's Companion as the core of a green aggressive deck and in Shards of Alara with trash like Onyx Goblet.
Note that "playing combo" is really just a specific application of this. Ideally, the combo you build contains cards you can expect to see late. No one wants Canyon Lurkers early and Barrage of Boulders wheels all the time.
The subtext here is also that you can't always assume you will end up aggressive in a control format. It's still an archetype that has to be open. Obviously if you crack the case and no one else knows about it, you should be in a good position to move in, but even then the cards might just not be opened. If it's a blocking format, accept that sometimes you have to play like it is one.
Blocking in a World of Beatdown
1) Close fast.
I covered this when talking about how blocking decks developed to handle formats where no life total is safe.
It's still possible to block in an aggressive format, but often it's not possible to do so in a way that lets you bury them in cards or chip away their life total. If you kill them, they stop being able to attack you.
If you want to block in an aggressive format, make sure you use it as a way point to killing your opponent and not as a long term plan.
2) Play Skies or other semi-defensive tempo strategies.
Yes, this works for making blocking work as well as making attacking work. If you can interact with their ground game while they can't interact with your evasive aggression, you will likely win.
Just remember, blocking here is a tactical resource used to buy time to kill them with.
3) Metagame and play the unplayables.
Same advice as playing aggro into blockers, but it applies the other way.
Typically, if you are trying to make control work, the unplayables aren't commons people bypass but rares and uncommons that they view as fundamentally outside the bounds of the format. I have fond memories of killing people with Lorthos, the Tidemaker and Lavaball Trap. They weren't always there, but if someone opened them you could get them late and position yourself to take advantage of having such a powerful effect.
The Sweet Spot
Not every format is black and white like this. On occasion, attacking and blocking archetypes can coexist (Innistrad) or games can be extremely fluid (Magic 2013).
But more often than not, formats do pull towards one of these extremes. And when they do, there's a huge end in understanding which one it is and how to approach it.