I know that last time I said that I was going to do Splash Damage this time around (not that any of you remember, I'm sure), but I have been putting off writing Splash Damage for three years, so it comes naturally to me. Instead, I decided to talk about the elemental tenets of making your own deck over the course of a couple of articles. Splash Damage - and don't worry that you don't know this term, as it is used by very few players at this point - is an element of rogue deck design that comes at the end of the design cycle. It seems to me more organic at this point to start with the basic fundamentals of rogue deck theory first.
Let us begin by asking what we mean by rogue deck design. In my mind (and I know that this term pre-dates him)"rogue" was popularized by Adrian Sullivan around 1997 with the creation of his non-team, Cabal Rogue. Though today we know that Adrian has been at least partially responsible for some very innovative decks, he would be the first to tell you that structure around him is the aggregate force responsible for many of the decks that we associate with Adrian, and that any innovations that he has been able to put forward into the game are the result of working with other talented designers.
For our purposes,"rogue" = different. Rogue is good not because different is good, but because different wins. Rogue deck decisions are based on one thing, and one thing only: metagame predictions. There can be no rogue deck design outside the metagame, and the rewards for playing a rogue deck rarely occur when metagame predictions are not accurate.
People misuse the term"rogue" more than they do the term"fascist." They hear the word in context and they think they know what it means. Rogue doesn't mean bad. If you play with bad cards, please don't call them rogue. Bad cards are bad. Rogue cards are very good when the metagame predictions are right and the strategies hold, even if they may seem bad otherwise. A Gray Ogre is not a good card. No rogue deck designer would ever tell you to play with Gray Ogre even if you had a deck where Gray Ogre seemed like the perfect fit: you can get the superior Suq'Ata Lancer at the same cost and size; you can get the slightly worse - but still superior to Gray Ogre - Goblin Chariot at the same. A Stromgald Cabal might not look much better than a Gray Ogre against a G/R land destruction opponent, but man does it put the hurt on a White combo deck.
For our purposes, there is therefore only one model for rogue deck design, and that is the one espoused by Cabal Rogue. If you are going to appropriate Adrian's term, you must therefore speak his language on his terms. If you ever submitted a deck to Cabal Rogue, the first question you got was always"What exactly does this thing beat?"
Therefore, Rogue is distinguished by two things.
1. Rogue is different. Different means different from the decks that you might expect to show up, which implies that the opponent will not expect your archetype, twist on strategy, or some distinguishing element of your design execution that gives you...
2. Strategy Superiority (or at least deck advantage) over your projected opponents
These things are very important. When we finally hit Splash Damage, you will see how important the first one is. Until then, it should be obvious that statements like"It doesn't take a genius to guess the best of the Tier Two Rogues will be drawn to playing B/G this year" don't make any sense. If you can predict that some bad player is going to play some bad deck and you expect him, you bring a plan for him. He isn't rogue, he's bad. Either you discount him and beat him because he is bad, or you have a mechanism for strategy superiority. I can tell you that when I work on Standard decks, I am not"aiming" at Mirrodin Block TwelvePost decks, but have a plan in the back of my head for beating them, even if it doesn't involve dedicating cards to this matchup.
The second point is very obvious. There is no reason to play a rogue deck if it doesn't beat the decks that you are planning to beat. If you can't beat Goblin Bidding and Ravager Affinity, let me clue you in on something: play Goblin Bidding or Ravager Affinity. No one cares that it isn't fair that your stupid pet deck can't win because R&D didn't ban the card you needed them to ban. The other fellow forked over the $20 bucks or whatever Pete is getting for Arcbound Ravagers and he is laughing at you. Winning is a binary, and if you're not winning, there are usually a dozen or more variables under your control that you could have addressed in another way in order to affect your disappointing outcome.
It is therefore an important thing to distinguish Rogue deck design from what I call Natural Strategy. Natural Strategy occurs when you look at the available card pool and a synergistic deck appears based on that card pool. Natural Strategy decks are increasingly common as Wizards R&D pushes blocks of synergistic mechanics. Because Natural Strategy decks are implied by the card pool, it is very unusual for a Rogue deck to emerge from this class of decks (though it has happened). The most obvious and notorious Natural Strategy deck to have appeared in recent blocks is R/W Astral Slide. This is the version my friend Peppermint Von Corduroy III used to win the Pro Tour back when we still called him Joe Black:
2 Gempalm Incinerator
4 Lightning Rift
4 Astral Slide
4 Akroma's Blessing
4 Akroma's Vengeance
2 Daru Sanctifier
4 Exalted Angel
2 Jareth, Leonine Titan
3 Renewed Faith
4 Forgotten Cave
4 Secluded Steppe
All aspects of this deck archetype feed one another in an obvious manner. There is nothing either tricky or unexpected about its execution. It has expensive spells, so it has a lot of land. It has a lot of land, so it is not typically manascrewed. Many of its lands cycle, so it is mana flooded a disproportionately small amount of the time given its heavy land count. It takes redundant advantage of cycling to fuel Lightning Rift and Astral Slide. The trio of Exalted Angel, Astral Slide, and one-mana cycling lands ties its early game with a natural mana curve. Astral Slide can counter the symmetrical effects of its board sweepers to save its own fellas. Its board sweepers cycle, so they are not dead draws against non-beatdown decks. It should be no surprise, therefore, that in a small two-set Constructed Pro Tour, this deck was so successful; no other deck could claim its synergy, overall power, or the fact that it was an obvious place to build.
This is not to say that even Natural Strategy decks cannot be templated to tailor individual, or even rogue, elements. Osyp's individual card choices and his strategies for certain matchups are in places very different from Zvi's deck for the same Pro Tour, despite the fact that they share the same archetype. Especially in an untested format, Zvi would not point to many details and say they were rogue or unexpected; he would instead point to missing elements of another build and say that that deck was lacking.
The more common rogue strategy is therefore Anti-Strategy. Anti-Strategy is the opposite of Natural Strategy. Even if a deck is synergistic, its base relationship is directly predator-prey. It may have a plan to beat a general opponent, but its main goal is to eradicate a focused number of particular, specific, decks. If an Anti-Strategy deck is any good, it will beat its specific opponent consistently. Jamie Wakefield was a notorious player of this kind of deck, even if he didn't characterize himself this way explicitly. Bill Macey used to say that Jamie would pick a three-deck metagame and make some sort of horrible fatty deck (that he would claim was a"beatdown" deck) that crushed one of the opponents, beat one of the other opponents a respectable amount of the time when played competently, and never, ever, beat the third deck.
Anti-Strategy decks tend to be overloaded with Dead Weight elements that give them poor draws in some matchup or other. That's okay, though, they probably have some other, different, Dead Weight element that beats the other deck. Because Matt Vienneau says that I only ever write about one deck, I am going to use a different, and equally iconic, example of Anti-Strategy that altered the Magic - and especially Extended - landscape at Grand Prix: Kansas City in 1999.
1 Mox Diamond
3 Arcane Laboratory
4 Force of Will
2 Gaea's Blessings
2 Oath of Druids
1 Spike Feeder
1 Sylvan Library
1 Aura of Silence
4 Enlightened Tutor
4 Volcanic Island
4 Tropical Island
2 Mishra's Factory
1 Faerie Conclave
3 Flood Plain
1 Adarkar Wastes
This is Bob Maher's deck from the Top 8 of Grand Prix Kansas City (importantly distinguished from his PT Chicago and GP Seattle builds). What an unbelievably ugly deck. Only Adrian Sullivan could love playing Enlightened Tutor for the one Mox Diamond. Don't let Ped Bun make any decks without sending them to an image consultant first. You will see (below) that Robert's subsequent models were much more beautiful. Anyway, this deck might not be pretty to look at, but that's part of what makes it such an ingenious Anti-Strategy rogue deck. The dominant archetype of the format was High Tide, and this deck had all kinds of Dead Weight like Oath of Druids against it. Good thing that all Robert had to do was Enlightened Tutor for an Arcane Laboratory to win. It didn't matter what happened after, if he topped both Oaths in a row, then followed up with the lone Aura of Silence, then that Triskelion. Bob would probably still be able to beat the average High Tide player to death with a Mishra's Factory. The particularly cool thing about Ped's deck is that it has anti-plans for everybody.
3 Despotic Scepter
1 Feldon's Cane
4 Howling Mine
1 Ivory Tower
1 Zuran Orb
4 Arcane Denial
4 Force of Will
4 Lim-Dul's Vault
1 Land Tax
4 Adarkar Wastes
4 City of Brass
4 Underground River
1 Black Vise
1 Loadstone Bauble
2 Blue Elemental Blast
3 Mana Short
2 Wall of Air
1 Swords to Plowshares
The first really iconic Anti-Strategy deck was Turbo-Stasis. The above version is the one that R&D's Matt Place used to make the US National Team almost a decade ago.
During the Black Summer, most of the best players were running Necropotence variants. Their decks were mana hungry and typically had no way to remove enchantments. They had the Ultimate Nullifier in Nevinyrral's Disk, but it came into play tapped. The Necropotence deck was going to tap itself out to pump Knights of Stromgald or direct a Drain Life, so Stasis really hurt them. Howling Mine ceased being symmetrical, because the Necropotence player had no draw step. It was fairly obscene therefore how Turbo-Stasis took US Nationals 1996 by storm, claiming two of the four National Team slots, because the deck wasn't really that good overall.
A prepared opponent with lots of Disenchants and Divine Offerings would just draw two from your Howling Mine, kill it, and thank you for the extra card. As finalist George Baxter showed, and eventual World Champion Tom Champheng echoed a month later, you could just play a Serra Angel and kill the Stasis player with it.
An unprepared player, though, some guy who was aiming for Necropotence... he would have a rough match. If he tried the"draw two, Disenchant your Howling Mine" play, he would just get locked under Stasis, lamenting the fact that he only had three Disenchants between deck and sideboard. People who didn't understand what was what would get locked up even in games they were"supposed" to win.
But that's how many rogue decks work. They show up, no one knows how to play properly against them, and they win in the short term. The next week, when people have tested and figured out what is what, these lose their advantage.
4 Ankh of Mishra
4 Chimeric Idol
4 Tangle Wire
4 Parallax Tide
4 Rishadan Airship
4 Spiketail Hatchling
4 Troublesome Spirit
4 Wash Out
1 Rath's Edge
4 Rishadan Port
4 Svyelunite Temple
3 Mana Short
4 Rootwater Thief
Maher won with this pile because his opponents played against him as if he were playing a regular Aggro Waters deck instead of a wacky Ankh-Tide deck. They would see how he"refused" to play counters the whole game and assumed that he always had one back, never knowing that he just didn't have any. Maher's deck is a good example of a rogue deck that gains an advantage by affecting the opponent's ability to make the correct plays. If you actually spent any time testing against this deck, you would see that it really wasn't very good. It didn't have enough punch, and it couldn't stop the opponent from doing what the opponent wanted to do in any way but time control. The thing is, by the time anyone figured that out, Bob had already cashed his Top 16 check.
Kai has pointed out that the other possible route for a good rogue deck to take is to graduate to regular old net deck. When I beat him to win a GP Vegas Trial with The Rock, Mike Pustilnik had never heard of the deck. A few weeks later he used it to win GP Vegas (I won the Vegas PTQ the same day... impressive, I know). A week after that, a year after that, and another year later, and The Rock is one of the pillars of the Extended metagame.
A similarly concrete Extended example is Bob Maher and his Ped Bun Oath deck. A year after GP Kansas City, Bob took the same archetype to his home town of Chicago and not only made Top 8, but won the whole Pro Tour (beating Brian Davis's Free Spell Necro in the finals). Zvi Mowshowitz will argue until the sky falls that the Chicago Top 8 was the best one ever because of the diversity, that there were eight distinct, eight very good decks, but it was these two decks that became the iconic pillars of Extended. Necro was already up there, but Ped Bun Oath officially graduates to archetype. People have not only heard of it, they are not preparing for it. These are The. Decks. To. Beat. I mean who was really going to copy the Fire Whip deck?
1 Null Rod
1 Powder Keg
4 Force of Will
1 Trade Routes
2 Gaea's Blessing
2 Oath of Druids
1 Spike Feeder
1 Sylvan Library
1 Shard Phoenix
1 Aura of Silence
4 Enlightened Tutor
1 Ivory Mask
2 Swords to Plowshares
1 Faerie Conclave
3 Flood Plain
1 Reflecting Pool
3 Treetop Village
4 Tropical Island
2 Volcanic Island
1 Phyrexian Furnace
1 Powder Keg
2 Mana Short
1 Gaea's Blessing
2 Oath of Druids
1 Crater Hellion
1 Aura of Silence
1 Circle of Protection: Red
1 Light of Day
1 Sacred Ground
1 Swords to Plowshares
Bob showed that his deck was really the Deck to Beat, and followed up his Chicago win with a GP Seattle win not long after. Also at GP Seattle, we saw the perfect example of templated rogue deck graduation to Tier One. Both Jay Elarar and Alex Shvartsman made the Top 8 of Seattle with Pooh Burn, a very specific Red beatdown version designed by Seth Burn and onetime Deadguy Marc Aquino. Oh, was that not the deck you were expecting me to talk about?
This version was the one Alex Shvartsman played:
4 Cursed Scroll
3 Ball Lightning
3 Hammer of Bogardan
4 Jackal Pup
4 Mogg Fanatic
3 Price of Progress
3 Viashino Sandstalker
3 Ghitu Encampment
Seth's deck is one of the rare examples of how a Natural Strategy deck can flow into the rogue zone by templating to achieve Strategy Superiority. If you don't see what I mean, compare Pooh Burn with the typical, more predictable, deck Mark Gordon used to win Kansas City (same Top 8 as Bob with the triple Arcane Labs):
4 Price of Progress
4 Red Elemental Blast
Gordon's deck has a lot of natural synergy (I'm guessing there are four Fireblasts missing). It rushes and accelerates into a place where it can close the last couple of points with a burn flurry. It probably has more pure power than Seth's deck (keep in mind for a second that Seth isn't allowed to play with Lightning Bolt).
Now consider how Gordon's deck plays against our established Decks to Beat. Davis can't wait to play against this pile. Davis could care less that Gordon has Lightning Bolt. He can't wait to split a Contagion on two of those Goblins. He can't wait for the sheer gratuitous pleasure of aiming a Corrupt at a Jackal Pup. He's probably going to mop up game one, but can't wait to reach for his sideboard. He is going to go for two-for-one with his Powder Keg every time. If Gordon doesn't play into the two-for-one, Davis is just going to accumulate card advantage and make his land drops until he starts going to Mark's head with the Corrupts.
Compare that to how Seth's deck plays against Davis's deck. Seth is never going to be in a situation where he is going to lose two creatures to a Contagion. Never. He is going to slow roll his offense. He is going to make Davis pay a life and two cards to Contagion one creature. Seth has guys to play out, but he is in a lot better shape than Gordon against those main deck Spinning Darknesses. For one thing, he can pick and choose and use haste to try to dodge them. He can send his most dangerous creature - Ball Lightning - when Davis doesn't have three Black cards in his graveyard yet. Further, he has more redundant burn. Seth can blow burn on his own guy to defend against Spinning Darkness a lot more easily than Mark can. Mark needs his burn spells to close, but he might never get into closing position. Seth, on the other hand, can use haste and his long game burn like Hammer of Bogardan to win a direct head-to-head with Davis's Corrupts. The best thing about this matchup for Seth is that because he has so much burn and doesn't have to lean on his creatures (which might just go Spinning off), Davis has to be very careful with Necropotence. This is a liberty that Seth has that few other decks can claim.
Seth's deck is even better against Bob's. This is just a nightmare for Bob in every way, from the ground (lands) up. Bob is going to eat it to Price of Progress each and every time if Seth is careful. Bob's main creature sanction and deck name is worthless. Almost none of Seth's creatures stick around in play on Maher's upkeep (if you go back to Grand Prix Seattle, you will see that Bob took a game from Alex in their Swiss match only because Alex could literally find no way to remove his own Jackal Pup, allowing Bob to go off). Bob doesn't have enough permission to fight all of the Red deck's burn, either (especially Hammer of Bogardan). Because of this, Bob is going to have to pay for his creatures the old fashioned way if he is going to race. This is a terrible idea both because his creatures are vastly inferior to Seth's in the short term, and because tapping for Morphling is going to be an invitation to Price of Progress.
Seth's deck had what we call Strategy Superiority over the Decks to Beat. If Seth's deck operated normally, and either Necropotence or Oath of Druids operated normally, Seth's deck would win, because his baseline strategy could operate while simultaneously stymieing the baseline strategies of his opponents. He doesn't have to do anything special in order to make life difficult for Necropotence and Oath. When you see the reverse happening, for example, when Hacker with the Free Spell Necro beat Jay Elarar with Pooh Burn, it was because Hacker played a wildly different base strategy, and also fooled Elarar (Hacker played as a pure Masticore/creature elimination deck and raced Elarar, who did not realize he could kill Masticore with Incinerate due to the then-fresh 6th Edition Rules).
Strategy Superiority differs from what the AustiKnights used to call"deck advantage" in one basic way. A deck is said to have deck advantage over another deck when it tends to beat that deck. Deck advantage is pure numbers and doesn't talk about the mechanics behind those numbers, inevitability, who's the beatdown, or any other factors, such as having an idiot for a playtest partner. For instance, the notion of a three-deck metagame comes from this pyramid:
U/W has deck advantage over Erhnam
Erhnam has deck advantage over Necropotence
Necropotence has deck advantage over U/W
That is where we, historically, get the rock, paper, scissors model for Magic metagames that everyone uses today. The fact of the matter, though, is that these basic points of the metagame triangle existed for over two years using the same core card pools. There were a great deal of fluctuations in the decks, what was restricted and what kill cards were used in some cases. Ultimately, the model proved to be false, because unchecked Necropotence actually beat everything (even if the tables were turned around later), yet Jon Finkel prevailed with his U/W decks by doing things like playing main deck Circle of Protection: Black.
But let us assume that playing"normally" on both sides, a U/W deck was actually supposed to beat an Erhnam Djinn deck (presume a G/W Armageddon deck). The reason that U/W preyed on G/W was that it theoretically had Strategy Superiority. As long as the U/W player didn't make any mistakes, normal draw against normal draw, standard tactic against standard tactic, the U/W deck would win. It had lots of creature elimination - probably four Swords to Plowshares and four copies of Wrath of God - and could use permission against creatures if need be. The G/W deck had to make sure that it didn't get two-for-one'd too many times, or it would run out of creatures when compared to the U/W deck's amount of creature sanction. Now factor in that a U/W deck could play with Icy Manipulator or Kjeldoran Outpost to force the G/W deck into disadvantageous Wrath of God positions, and that, sometimes playing Millstone, it could also reduce the raw number of creatures the G/W deck could present as threats over the course of a game. The U/W deck would always have at least twice as many permission spells as the G/W deck would have Armageddons, as well as a higher likelihood of main deck library recursion. It had Strategy Superiority in all aspects of game development, which implied its deck advantage.
Here is a quick test to see if you could play U/W correctly against G/W in 1997.
All things considered equal, on turn 5, your opponent, who has not yet played a land, is attacking you with his Erhnam Djinn. Do you:
a) Take the hit, or
b) Aim a Swords to Plowshares at it?
(This is not a trick question)
If you said a, you are right! Many a G/W player would probably see taking the hit as a sign of weakness or vulnerability. Especially if he went first, he would probably now cast Armageddon. Preston Poulter, the icon of G/W play for many years since PT1, said that if Armageddon was in your hand, you should probably cast it. Most G/W players would follow Preston's rule, perhaps assuming you were waiting to draw a response card or waiting to play your fourth land for Wrath of God.
By casting Armageddon in this position the G/W player would put himself behind any likely loss of card advantage. You could just Swords to Plowshares his Erhnam Djinn and leave him with nothing. You would lose one card less than the G/W player at a minimum in the Armageddon exchange. If you used your Swords to Plowshares, he would just replace his Erhnam Djinn. Back then, U/W decks didn't have Fact or Fiction. They could not get into one-for-one fights over the long haul without having to use their permission against creatures, which could leave them potentially exposed to a disadvantageous Armageddon.
If the G/W player didn't have the Armageddon, he might play a second threat, playing into the U/W player's Wrath of God anyway. In either case, the strong U/W player would know that the G/W opponent would not have any burn, any way to really punish him for taking a four-point hit on the relatively early fifth turn once anti-creature measures were online. Yet the majority of U/W players probably used the Swords to Plowshares instead of focusing on the correct strategic elements that gave them superiority in the matchup. They might win anyway (perpetuating incorrect play over time), but when they lost, they would probably attribute the loss to something other than having an flawed fundamental plan and making an incorrect play during the critical early development of the game.
The interesting thing is that even though the U/W deck had Strategy Superiority, fluctuations in the card pool could allow the wily G/W mage to seize deck advantage without ever establishing it himself. In fact, it was in interacting with the U/W mage to every indication that the U/W mage had Strategy Superiority that the G/W player would win. In the summer of 1997, after the release of Weatherlight, a new technique was developed by Team AustiKnights, and was later adopted for the World Championships by members of the then-great East Coast Assassins and Team Deadguy.
The G/W player would Thaw into a position of eight mana but no Thawing Glaciers. He would act only if his U/W opponent had his own Thawing Glaciers in play. The G/W player would deliberately over-commit threats to the board. Ideally he would lay down his best creatures, perhaps two Maros, desperately trying to make up for his slow offense in the early game as the U/W player accumulated cards in hand and gathered defensive spells.
The U/W player would obviously use Wrath of God to kill the Maros (if he didn't, the G/W player would kill him anyway, having over-committed powerful and now unmolested threats to the board). Tapping down to"only" four or five mana, the U/W player would pass the turn, confident in his double counter ability.
However, this was exactly what the G/W player was expecting. In those critical nine turns, the G/W player used his Thawing Glaciers and Sylvan Library to sculpt a perfect hand. He would lead with an Abeyance, expecting it to be countered. He would then play his second Abeyance. It wouldn't really matter if this one got countered or not. If it didn't, the G/W player's victory would be assured. If it was countered, the U/W player would still have to have a Force of Will for the lynchpin third spell, an Armageddon. After snagging the U/W player's Thawing Glaciers, the G/W player would lay down his own and establish a huge advantage over the next several turns, ideally on the back of Sylvan Library. Almost always the Thawing Glaciers advantage after depleting so many of the U/W deck's defensive spells would be enough to win, even if the game turned ugly. If the Armageddon stole a Kjeldoran Outpost, all the better. If it didn't... well... that Outpost might be one of the only lands the U/W player had left in hand.
Notice how the G/W player wins the match by eschewing his baseline strategy for a rogue one. He makes the U/W player believe that he is playing a standard game in order to bait him into action. In actuality, the G/W player is focusing on the one resource where the U/W player's hungry deck does not have an overall strategic advantage: mana. The U/W deck is mana hungry and the G/W deck leaves the board with a Thawing Glaciers down. The way slow games developed back in 1997, the U/W player had probably already discarded a redundant Thawing Glaciers, because he had so many cards in hand and believed he would never miss a drop.
Notice also how the G/W deck can turn the tables on the U/W deck only if he builds a specific way, loading his deck with specific cards like Abeyance, multiple Sylvan Libraries, and Thawing Glaciers of his own in order to win a usually difficult matchup. Though he has a plan to win, the G/W player also relies on the expectations of the U/W player in order to execute on that plan. He has a very definite sequence of plays in mind, but no inherent Strategy Superiority; the G/W deck's entire plan is focused on resolving one spell under specific conditions. Rather than fight over an ephemeral creature-based board advantage, he wants to win a long haul mana war. It is important to note that any time you template an archetype to an extreme in order to gain a particular theater, you may be giving up something in others; the AustiKnights chose to imbalance the rest of their specific card choices against Black, playing with Whirling Dervishes and Scalebane's Elites as their supplemental creatures to make up for the fact that they had built a slow core.
You can also look at how a deck is specifically built into a"combination" deck. This G/W deck is far less efficient at executing on a traditional creature-Geddon plan, even if it has the tools for beating U/W if it sculpts the right draw. edt used to say that decks like High Tide and Trix had to be combo decks because they had a limited amount of space left after shaping themselves with all those powerful manipulation and disruption cards.
A more recent example of how these principles are implemented comes from templating AnGGRRRy Slug, the deck BDM and I designed for States this year. I am pretty sure that AnGGRRRy Slug was the best deck in the pre-Skullclamp world of States. I lost a mirror and to U/W to miss Top 8 (going 5-2 or 6-2 or something), though Aaron Muranaka won Utah States with essentially the same deck. BDM had told me before the tournament that U/W was the worst matchup, so I sideboarded heavily for it. Ironically, I was able to beat U/W in game one, but couldn't quite string together the threats in either of the other games, despite resolving a lot of Plow Unders.
Here is the States era AnGGRRRy Slug:
4 Birds of Paradise
4 Creeping Mold
2 Krosan Tusker
4 Molder Slug
4 Plow Under
4 Ravenous Baloth
4 Vine Trellis
2 Hammer of Bogardan
4 Stone Rain
2 Contested Cliffs
4 Shivan Oasis
4 Wooded Foothills
BDM continued to play the deck in Standard tournaments at Neutral Ground, where U/W decks were common. Even though we were supplementing our land destruction with Flashfires, U/W was a problem due to the low threat count. I actually theorized that AnGGRRRy Slug would have long game inevitability if it just played around Mana Leak, but U/W would sometimes get uppity and win with Exalted Angel or something.
We set out to find new sideboard cards for the U/W matchup.
BDM went with Xantid Swarm. He learned when to play Xantid Swarm early, and just made sure his spells resolved. I didn't like Xantid Swarm on principle. They didn't seem good. BDM assured me that he never lost to U/W with Xantid Swarm in his deck, and I believed him. It just seemed like Xantid Swarm was not just narrow, but, though it tipped the scales back in favor of AnGGRRRy Slug, it didn't establish Strategy Superiority. In my mind, I don't just like to win. I like to know why I am winning, and, even better than that, I like to win even when my opponent's deck is operating within the set interactions of the matchup. Xantid Swarm violates the interactions of the matchup, allowing AnGGRRRy Slug to resolve its land destruction.
My solution was to sideboard Dwarven Blastminer (if you read my article from a few weeks ago, you will see that I liked him enough to move him to the main deck). Dwarven Blastminer does a couple of things. First of all, he supplements the land destruction theme similar to the way Flashfires does. Secondly, he curves proactively under a U/W deck's fastest creature kill, comes online before it can be cast, and, in fact, helps to keep it from ever being cast when combined with the rest of the deck's suite of threats. Third, Dwarven Blastminer early establishes true Strategy Superiority, not just deck advantage. I am not contesting that BDM's sideboard creature could create a win if it showed up on the right turn. The difference is that an operational Xantid Swarm comes out of the sideboard to fight a particular class of cards that may or may not pair off with threats from my own deck, whereas an operational Dwarven Blastminer trumps the U/W deck's entire developmental process while at the same time enhancing the core competency of AnGGRRRy Slug. Besides, Blastminer has applications in other matchups, while Xantid Swarm does only one thing in one matchup. I think BDM was actually persuaded to move to Dwarven Blastminer, and has him maindeck now, as well (but if I showed you my all new, even more awful, AnGGRRRy Slug, you probably wouldn't even recognize it).
Provided your testing is good, you should be able to see which rogue decks you develop have deck advantage over those in the metagame. Sometimes answering questions is most efficiently done by playing cards against the specific cards that irk you, even when they seem narrow or not very good. Sometimes delving into the real nuts and bolts of the matchup in order to establish Strategy Superiority is the right way to go. I am generally comforted when able to establish Strategy Superiority because it means that even if the opponent's deck is firing at all cylinders, I am favored to win. If all I have is a matchup percentage, the games tend to be closer, more reliant on specifically tight play, resource allocation, and mistakes on the part of the opponent (for a player at my skill level, this isn't the best approach). The tricky thing about Strategy Superiority is that you have to actually understand the opposing strategy in order to mold a predatory relationship.
The best example I can think of this in recent years is the mono-Black deck I played at Regionals last year. We had two or three Visaras main and were crushing every deck in the metagame (I did not know about Zombies prior to Regionals and lost to it, after drawing with Wake, which we under-prepared for). U/G and G/R were said to side in Compost. U/G I didn't care about. All they could do with Compost was draw more guys and I just killed all their guys so it didn't matter. What mattered was if they had Upheaval or Stupefying Touch, which made for annoying Magic, but still favorable over three games, by the numbers. The problem was G/R. G/R would draw into more, like, Violent Eruptions. This was not acceptable. I could not tap Visara to kill a Firebolt.
Osyp suggested Mortivore to PJ for some reason. We never really did anything with Mortivore, but the concept of a huge regenerating creature that actually did something was intriguing to me, so I upgraded to Laquatus's Champion. This, on the other hand, was the stone cold nuts. I never lost to G/R in post-sideboarded games (though the percentages for pre-sideboarded games were still better). The goal was to attack their board with Innocent Blood, Chainer's Edict, and Smother per normal. If they didn't have Compost, game two would be a repeat of game one, and you would just mop them up with all the removal. If they did have Compost, you were essentially using your Innocent Bloods and Smothers as Unsummons and Aether Bursts. These Blue cards are generally considered very good at controlling the board; their philosophy is not to eliminate creatures, but to purchase mana and steal momentum from the opponent.
I realized that I could never win a card advantage war against a Compost with spot removal, so I focused on time. By buying time with my removal cards early (not to mention the Compost Time Walk I would probably get around turn 2 or 3, I would have a ton of life as I hit my late game drops. As early as turn 5, I would start with Corrupts to the nugget and Champions as soon as I had a free mana. It's not like the R/G player has any non-Phantom Centaur willing to bring it to the Champion. Yummy nummy Wild Mongrel.
The above is a great example of ignoring apparent card advantage interactions to establish true Strategy Superiority. It assumes not only that the R/G deck is working properly, but that it got the nuts. It has Compost if it wants. It draws twice as many cards my Precious. The sideboard plan works if the opponent has Compost. Great! You have Compost! I have fourteen life! So there! All I have to do is draw lands and spells. If the R/G player is"bad" though, the"good" sideboard strategy is bad. If he has Threaten instead of Compost? You lose your blocker, get your face smashed by an alpha strike, and take a Ball Lightning to boot. Potential yikes there (good thing that never happened to me).
My old friend Jamie drove me nuts me by trying to sideboard Repopulate against Living Death decks. To him, Repopulate countered the graveyard recursion element of Living Death at a low cost, and could be cycled when not needed. At the same time, it didn't do anything about his own graveyard, meaning that he should end up ahead, post-Death. The problem, of course, was that he wasn't addressing the real problem. He was trying to counter the Stroke of Genius or Donate, and ignoring the fact that he was getting wrecked all along the way. The opponent could easily two-for-one him on the back of Survival of the Fittest, hide behind Wall of Blossoms and Tradewind Rider, go Gray Ogre beatdown with his Ghitu Slingers and Bone Shredders... and never have to play the Living Death at all. He did not address the core strategy of this matchup, possibly because he didn't understand the actual variables and strategies involved, and was therefore long frustrated by his inability to beat Living Death.