Attention to the End – Building Towards The End Game

As we’re all aware, Mike has had a modicum of success of late. His State Championship win, piloting the innovative “This Girl” deck, goes a fair way to banishing the “Bad Player Flores” nickname once and for all. By his own admission, his success is largely due to his fine analysis of the intricacies of the endgame – sculpting the proceedings in order to steer the game to an inevitable victory. Today’s theoretical excercise takes us through the thought process behind such structured play…

This theme is something that I plan to write about in greater detail in the future, but for now I decided I would do an introductory piece on the idea, and current applications of, the end game in Magic. The end game is closely related to the concept of strategy (please remember that most so-called Magic: The Gathering "strategy" articles actually have little or nothing to do with strategy), at least in the sense that many branches of tightly strategic play move towards a set end game, but gets even less specific attention. The end game is an element of my strategic game – and I would assume the games of most of my readers – that needs a fair bit of work (less than mulligans, less than unlocking the Limited board, more than figuring out what land to play on turn 2); I’d guess that you don’t think about it discreetly at all. Oddly, end game awareness is probably the one area where the average Vintage player greatly exceeds the average competitive player (of other formats) in capability, even if he does not speak about the game specifically in the terms I am going to discuss today; there is actually a very simple reason for this: Vintage players only have one or two scripted possible end games (for example "Tendrils you" or "Darksteel Colossus, Time Walk" in recent months) and therefore lack the capability (possibility?) of losing their paths the way that Standard players can as they progress from the middle turns (what few there are) to the end game. Any competent Vintage player using one of these end game plans essentially has no choice but to pay attention to it as he proceeds.

My attention was drawn to specifically the end game in recent months due to the disparate sideboarding strategies for Combo Deck that Josh Ravitz and I had in the one week we both played it (Josh at Grand Prix: Madison with Ant and Alex, me in Connecticut with Steve and Paul).

Because that is the most logical time to discuss sideboard strategies, Josh and I went over ours the week after we played Combo Deck. I was complaining about how I had beaten Heezy Street in a Game 1 but lost the match (Paul and Steve pulled it out, and we finished second due to my poor play against Brassman in the finals); Josh had lost to Mark Herberholz himself in the Grand Prix, and finished Top 8 but out of the elimination portion. I was surprised to hear that Josh left in Angel of Despair. As a seven-mana card, and one considerably less durable long game than Ink-Eyes, Servant of Oni, I had sided mine out despite the fact that one won me the first game.

No, they’re quite good against Heezy Street, I was informed. I wanted answers to Blood Moon and to lower my curve in anticipation for creature beats, bringing in Naturalize and Putrefy (I could have sworn I had a fourth Faith’s Fetters in the sideboard but this is apparently a pernicious fiction). I incorrectly assumed that I could just ride the minor mana acceleration of my Signets and Karoos, and win on incremental positional advantage (Loxodon Hierarch) and value trading (Faith’s Fetters). If I just held on early, my more powerful cards would win out eventually (wrong).

I ignored the monolithic, Keiga-like presence of the admittedly expensive Angel of Despair. Angel of Despair takes out Heezy Street’s best card, then trades with a Rumbling Slum. Or it nukes a Rumbling Slum and scares away a Giant Solifuge. It is tempo, card advantage, and a finisher all in one. While not an end game card in quite the same way as some of the other options we will discuss today, my deficiency in this matchup sideboarded put me on the path to thinking about the various phases of the game, the when and why of how to address them.

Why Do You NEED An End Game Plan?

The short answer is "you don’t." Many competitive decks such as Budget Boros and Zoo variants cannot by nature play discreet end games. Their value as deck choices comes largely from the homogenous natures of their opening sevens. They do not play end game cards save one or two Demonfires (and even then only in rare circumstances), so their versions of the end is to throw as many early game cards as possible at the opponent in the hopes of overwhelming him. We all know this to be a reasonable game plan in at least some formats.

However, the end game is a rich and important element of other deck styles with actual mana curves and strategic progressions, so thinking about it specifically, and understanding what you might want to do when you hit the end game will only help you win more if you are prepared. Recently my friend Paul Jordan quite glowingly wrote this about me:

He is constantly plotting turns in advance, essentially writing the script to the figurative murder of twenty separate life points owned by his opponent. I don’t want to steal too much thunder in the details, as I’m sure Mike will have a tournament report of his own, but he really is actually playing well.

One of the main reasons I am playing better Constructed Magic than Paul (who is a vastly better Limited and tempo-oriented player than I am) is that I figure out how exactly to get to my end game so that once I am there, I can finish out the duel, QX. A simple way to think about it is "how do I plan to actually win this game?" Once I know that, I can engage in all manner of tomfoolery to keep the opponent off the right plan while I back office my way to the way I want to win. A big problem with my no Angel of Despair anti-Heezy Street strategy was that, while my cards were certainly good, I didn’t really know where I was going with them. Many players have the idea that you can make a dozen profitable trades, and by the end, you have so much added value that you just win on domino effect, automatically. That actually happens most of the time, which feeds the aimless "you can’t beat Trish" camp, but a lot of the time, profitable trade after profitable trade on non-essential resources is just going to put you into a position of false security; all war is deception, remember.

It is one thing when you are playing Limited and literally every move you make is a two-for-one Zap, Exclude, and Flametongue Kavu. At some point if you are pushing your cards further and further into enemy territory and he has literally no board presence, you probably have tempo enough. Conversely we (and here I definitely mean "I") tend to lump "profitable trades" together with whatever positive elements besides overwhelming card advantage and tempo. Like Faith’s Fetters is basically always a one-for-one (even when they Disenchant it after a few), but you get something out of it. A bunch of intangible virtual cards and a little unrelated life gain is nice, but it doesn’t mean that you are knocking down the doors to the castle, a sacrificial virgin on each arm.

I once asked World Champion Julien Nuijten who had the better Gifts Ungiven sideboard strategy, the Dutchies or the Japanese. Julien chuckled and – admittedly biased – said "Dutchies, obv." I don’t know which one won more, but from the perspective of attention to the end game, Dutch was definitely correct.

The Japanese plan, generally credited to World Champion Katsuhiro Mori, involves siding in Godo, Bandit Warlord and equipment to create a positional advantage. The Godo plan is interesting in that the Japanese specifically play to the beatdown role, but do so while creating tempo and card advantage.

The Dutchies also sided in additional threats. However, their game plan is to go to four copies of Kokusho, the Evening Star – and a redundant Time of Need – in order to strike and then double up on Kokusho, lethally. The Japanese plan is a very good mid-game plan that puts the Gifts deck on the offensive and backs the opponent into a corner; with the Dragon’s Fang, it makes every Sakura-Tribe Elder into a potential breaker. The Dutch plan, though, when executed correctly, immediately landslides the game.

Teddy Card Game wrote a fantastic match report contrasting the two plans when StarCityGames.com columnists Rogier Maaten and Kenji Tsumura met in the Top 8 of Grand Prix Salt Lake City. Japanese sideboard got Rogier this time.

In today’s Standard, there are about two-and-change legitimate end game plans. Almost all of the successful Standard decks utilize one of these plans, or runs a specific foil to stop the opponent’s ability to hit that end game plan. The most obvious is the fast (or at least persistent) Akroma-as-trump. In a very "Vintage" way, the Akroma strategy can accelerate to the end game with unreasonable speed, lapping the opponent (who may have not yet crossed the minimum game line), tearing into the final phase of the game as quickly as turn 4. Like Upheaval plus Psychatog, the Akroma end game (though trump in the abstract) is creating its own and actual mini-metagame. There was a time when players were using Innocent Blood to counter Desolation Angels, Balanced Terravores, and Upheaval-fed Atogs; there was even a time when Seth Burn declared Psychatog "dead" as a strategy due to the printing of Chainer’s Edict in Torment. Psychatog players shifted their strategies to accommodate the low mana environments they themselves were creating, waiting a turn to set up Circular Logic, moving to Fact or Fiction or Gush-based Wonder attacks (eschewing Upheaval entirely in some builds, especially after ‘Tog’s term in Standard was up), or incorporating alternate – and Edict-proof – kills, like one maindeck (or multiple sideboarded) Zombie Infestation(s). In the same way, the present format is trying to find the right angle to attack the theoretically best Akroma. Condemn is the obvious answer: at one mana, it fits the puzzle at any stage of the game, whether turn 4 or turn 24; Condemn, like Skred a few short months ago, is also essentially uncounterable by conventional means given sufficient mana. Akroma players are aware of the presence of Condemn. How do they deal with this answer? What counter-counter-measures will they employ to ensure the safety of the Angel of Wrath before entering the Red Zone? Recently, Nick Eisel and Brian David-Marshall have begun to advocate Temporal Isolation as the preferred anti-Akroma measure. Temporal Isolation is only one more mana than Condemn, and solves one of the basic Akroma counter-counter-measures that works against Condemn, which is just playing or Resurrecting a second Akroma. Temporal Isolation is in one sense a great solution that also helps to stunt Dragonstorm’s Standard finishing flurry, but in another sense, is flawed in that more than half the Akroma end game decks play a mode count of three copies of Mortify main, will in general be loathe to side down to fewer than two copies even when the opponent presents zero main deck enchantments, and will in most cases have few targets to point them at save creatures (for which these decks generally have many other answers).

The most powerful end game play in Standard is the unchecked and Hellbent Demonfire. I would guess that most players do not understand why this is an end game strategy and not just some card you play out of your hand, point-and-click, the way we spoke about other cards and admittedly legitimate general strategies, above, viz. Godo. I can tell you from personal experience that the Demonfire end game is one of the most skill-intensive, strategic, and deliberately deceptive available in Standard (and previously Ravnica Block). I have learned more about playing my middle turns well by running Demonfire in the various Rakdos hybrids and later KarstenBot than probably any other skill set in Magic, ever. You will find as you learn to play for the Demonfire that even when you need what may have seemed in younger days a Herculean amount of mana to win with a single spell, when you play for the Demonfire, that mana comes. Your opponents will often throw up their hands – angry – when you win, assuming themselves "outside of burn range." Pro Tour Champions may never realize that you had Demonfire in your deck until it springs, and lesser men will assume you just got lucky. All of these things mix and mingle into the mystique and splendor of the Demonfire kill, and sublimely contribute, I’d assume, to why otherwise reasonable thinkers like Brian Kowal put Demonfire at Standard #1 (where, to be fair, I too had it until I studied the numbers) over Wrath of God.

Demonfire is special. There is no other way to describe it. Chad Ellis once said it was, sure, the best X-spell ever printed, but nevertheless overrated. He was, I think, very wrong. Demonfire is still underrated, and probably underplayed. There is something legitimately magical about Demonfire that transcends the mere cardboard, or even how it is positioned in the metagame. The way it encourages you to play against sophisticated opponents or in the face of ostensible trump strategies due to (and via) Hellbent is… The closest I can come up with is back in 1994 when I first saw Magical Hack, how a door opened for me and I became aware for the first time that there was some secret knowledge locked in the interplay of different, specific, cards that could be combined with methodologies much more intricate and strange and wonderful than throwing Kird Ape into a deck with basic Forest. Demonfire is the world’s best Blaze. It is the end, yes, but it is also the path. Hellbent was attached to Demonfire, we presume, as a Rakdos stamp… I wonder if R&D ever envisioned that decks full of Compulsive Research – or even Blue decks with Dark Confidants – would sculpt to their end games to no cards in hand, sloughing their lands and spells strategically and tactically, incrementally, over six turns, or three, or just one dizzying flourish of quick drops and Signets to the often unpredictable (or at least unseen) fiery conclusion of a duel.

Demonfire is special. It encourages a certain style of play out of non-Rakdos decks that is not intuitive, and requires looking many turns in advance. I think that there is probably no other card that requires so skilled an attention to setup, requiring long thought beyond the immediate tactics and operations of a duel. There will be many games, like that desperate turn when Mills fumbled and Buehler became the Chicago champ, where Demonfire will be the random X that steals the game (good in its own right), and many others when a powerhouse mana deck like UrzaTron will use it as a blunt force instrument to finish a “helpless” vanilla opponent with double Remand backup… But when this card is played to perfection, against a hand full of permission spells, staring at a Circle of Protection: Red or Story Circle… it’s so Magic, it’s magic.

But we cannot discount that Demonfire, beyond all its skill in its best moments, is one of the most blatantly obvious "get lucky" cards in Standard. That’s a big reason why I tend to play four. I really enjoy the inexorable Demonfire, a juggernaut unless the opponent has some kind of tricksy Signet-lifting plan, or perhaps a Willbender, but I am also fine with tapping eight or ten lands with three other cards in my hand, just hoping that it will resolve. Point-and-click. Point-and-click. When you are applying pressure on the opponent, even a Blue mage will often have mana tapped on his own turn before he realizes you are pushing for the end.

Beyond Akroma and Demonfire, there are a couple of other potential "end game" cards and strategies that deserve mention. I don’t think they are specific and discreet in the same way, or at least not as ubiquitous, but in many cases you play to them or fight to hit them in the same way you would one of the two most important ones.

Bogardan Hellkite
Bogardan Hellkite fills the same role as Akroma, Angel of Wrath in some decks, and is part of a very specific combination kill in one specific deck. In the former case, it sure can be scary on turn 4, but the Hellkite still has its share of holes. Most mid-range decks can answer it at the cost of five life, three mana, and a single card, and humble plays like Temporal Isolation rob it of a tremendous amount of punch (sometimes all it’s punch). In Dragonstorm, Bogardan Hellkite is "just" part of a combo kill, which I guess can be characterized as an end game strategy (especially if we are willing to give the nod to Vintage players, as we did at the beginning of this article), but is not expressly the same thing because while a true end game is defined by playing through the minimum game through the middle or setup turns into the finale, combo decks in the abstract tend to compress the middle, accelerating directly to their flash, bang, and panache as quickly after passing the minimum game line as possible.

Glare of Subdual
This card may be the best anti-end game card in Standard, certainly better than Circle of Protection: Red or spotty one-for-ones. While it can function as the last hammer blow, I don’t think that Glare is a truly legitimate end game card simply because it is so broad and abstract in application. While Demonfire can sometimes kill a Birds of Paradise on the second turn, that’s not really why it’s there, and given the resource stump you’ve just inflicted, you can trust that the next Demonfire will clinch it in a few turns. By contrast, Glare is a defensive mid-game card at least half the time, a tapper of Howling Mines, a very randomly good – and random – card that you play to get some sort of incremental advantage at some point in the game. For sure it can be a great end game card in the sense that you tap all the opponent’s blockers and then crash for the kill over one or two attacks, but the fact that you probably never attack with the same configuration of creatures using this technique over the course of an entire tournament relegates it to very good and oftentimes highly strategic utility card (at least the majority of the time). That said, Glare is a format definer and one of the main reasons U/W Akroma decks either have to play Black (Mortify) or Sacred Mesa (alternate kill when Akroma is kold to a Saproling token).

This is an awesome finisher, and comes closer to Akroma and Demonfire to signalling "the end" than almost any other alternative. The only reservation I have is that there are numerous tactically correct ‘Vore games where you will just punch out a 1/1 Magnivore because it is mana convenient, lining up the land destruction, card draw, and Wildfire death knell over the course of numerous turns before the ‘Vore actually becomes a singularly frightening proposition.

Firemane Angel
For sure this card is "the boots," as Heezy might say, making games impossible to win for numerous deck styles. It definitely suffices as an "uncounterable" ten mana threat in the extra-long game, especially as the cap to a well-exhausted attrition strategy, and may be the second best creature in Standard. However the fact that Firemane Angel is so good comes largely from its versatility, such as dumping doubles on turn 3 via Compulsive Research against Zoo… where it probably wins more games than as the relentless ten drop we yearn for against Draining Whelk control. As soon as you are willing to mark Honden of Cleansing Fire down as your Angel of Despair, then sign me up. I’m all for a Firemane Angel. That said, like Demonfire, Firemane Angel is definitely the kind of long game threat that you can "play for" or "play to" strategically (as is Magnivore). I just wouldn’t characterize them as specifically important to deal with as Big Two Akroma and Demonfire, which is where we draw the distinction.

Why Differentiate?

The end game cards of the present format are important as discrete ideas for a couple of reasons: Unlike the best decks of some other formats, these cards are only trump at a certain stage in the game, and essentially only one-at-a-time (contrast with Skullclamp in Ravager Affinity 2004, or Umezawa’s Jitte or Damping Matrix in White Weenie).

Few players ran "anti-Tooth and Nail" cards. Our goals were to stunt the UrzaTron before it hit nine, or to blast Tooth and Nail out with multiple Beacons and Pulses if the worst happened. We can attack Demonfire the same way (have a plan to keep it off Hellbent if we have counters, or stunt the mana or end the game before the mana is sufficient if we are beatdown); Akroma, for all her awesome offense, actually has many exploitable angles.

What we do not want to do is treat these cards like any other threats. Terry Soh showed the world what he thought about "Cranial Extraction for Tooth and Nail" on the way to his sweep at the Invitational. Demonfire probably wins exactly as many games with Circle of Protection: Red in play as not.


Post Script: I have been reading this over a couple of times and while I think I hit most of the points I wanted to, I’m not sure that I got across what I wanted to (and keep in mind I have been working on this conceptually since like April). I actually kind of feel like Grant Morrison in his long leather coat and bald pate screaming that people should not just sit there and listen to his bullspit (he didn’t say "bullspit") but instead invoke his actual actualization techniques. Shouting, like pulling hair and biting, is borne from a barrier or other difficulty to communicate, and I’m not sure if I got what I wanted to get out across. If I didn’t, I’m sorry, and I’ll try to flesh this out more at another time. It just seems to me that the game is played in different ways at certain stages, and that attention to the end game, specifically, as a time to prepare for, as a time where you want to keep your play and specifically strategy tight even when it seems like you are trumping should be a way to improve your game. That’s it. Thanks for reading.