A few years ago I wrote a series of articles (conveniently archived here for those of you who were pre-literate in 2000) called Building Broken Decks. My friend Zvi Mowshowitz really liked this series but for the fact that it wasn’t actually about building broken decks at all; in fact, in hindsight, it was more about identifying the principles of statistically solid, but fair, decks. Because of my associations at the time (we are judged by the company we keep, after all), I think I didn’t really understand the difference between a deck that was competitive and filled a nice niche in the metagame versus a really broken deck that somehow abused the mechanics of the game and pissed on the rules and any sense of fair play and propriety as it went about its process.
This preamble can be translated as “Ted has been bugging me since I signed on with Star City to revisit Building Broken Decks.”
Today I am going to talk about something very basic, which is – broken or not – how I pick a deck to play.
Short Answer: I try to figure out all the possible decks in the field and pick the deck that has the best chance of beating them.
Long Answer: see below
The Baseline: Advantage
Basically everything that I try to do in any article or newly designed deck is based on some perceived advantage. A simple example is the deck I used to qualify for Pro Tour: Los Angeles.
Creatures (17)
- 4 Sakura-Tribe Elder
- 1 Isao, Enlightened Bushi
- 4 Meloku the Clouded Mirror
- 4 Kodama of the North Tree
- 4 Keiga, the Tide Star
Lands (23)
Spells (20)
There were certainly other U/G Legends decks before Critical Mass which were successful to one degree or another, but ultimately flawed in their early game decisions for the field. In a head-to-head matchup, these decks might have a potentially busted late game via one or two copies of the narrow and clunky Overwhelming Intellect, but I thought that I would have the advantage by simply playing four Umezawa’s Jittes and four Sensei’s Divining Tops. I was willing to give up those few and far between games that Overwhelming Intellect demolished the opponent after an attrition war in exchange for the flood of games I would win with a more consistent early game of hitting drops or winning the Jitte war with greater frequency.
Explicitly speaking, I put a lot of stock in even small margins. In my experience, tournaments can be won and lost on tiny mental lapses and miniscule mistakes, so each and every percentage point of advantage that I can eke out from whatever sectors is welcome; these small margins are padding for all but my 40% errors. Together, they help to take advantage of my opponents’ countless mis-taps and ceaseless “judgment calls” in order to take close duels.
I have highlighted ways to get small advantages in the operations article; similar advantages, large and small can be had at the deck selection stage.
Playing Established Decks
Sometimes people ask me why I never play The Deck to Beat. The reason is that it is very difficult to get a competitive advantage when you pick that deck that is both known and targeted. The most glaring exception I can think of is Mirrodin Block Constructed, in the case of Vial Affinity.
The way Mirrodin Block decks laid out, the format at equilibrium should probably have been about 75% Vial Affinity, 25% other decks. At 75% Vial Affinity, Vial Affinity would have approached a 50% matchup against the field at large. The problem with the format was that the dedicated anti-Affinity decks were only passably better than 50% against the Vial Affinity decks and the non-Affinity/non-anti-Affinity decks in general had less percentage. So what happened was that with too many non-Affinity decks in the field (including anti-Affinity), they would all run up against each other, causing weird aberrations.
Case in point: the ideal Tooth and Nail deck does not contain maindeck Electrostatic Bolt. Look back at last summer’s Championship Season, when Tooth was arguably the dominant deck. How many maindeck Electrostatic Bolts were played? Despite being superb against select threats and highly mana efficient, Electrostatic Bolt is quite poor against even more. You might say “silly Flores, by Regionals last year, there were no Disciples of the Vault or Arcbound Ravagers to kill.” You would, if you said that, be missing the point.
For Reference:
Jeff Garza’s GP: NJ Tooth List
15 Forest
4 Mountain
4 Cloudpost
4 Eternal Witness
4 Solemn Simulacrum
1 Platinum Angel
1 Leonin Abunas
1 Darksteel Colussus
1 Duplicant
4 Tooth and Nail
4 Sylvan Scrying
4 Oxidize
4 Talisman of Impulse
4 Electrostatic Bolt
3 Reap and Sow
2 Rude Awakening
Sideboard
4 Viridian Shaman
3 Tel-Jilad Justice
3 Mindslaver
1 Reap and Sow
1 Triskelion
1 Mephidross Vampire
1 Duplicant
1 Sundering Titan
The Electrostatic Bolts in Block Tooth and Nail were designed to kill Slith Firewalkers, not Affinity’s banned list. That they were good against Affinity creatures was nice (even essentially necessary given the competitive landscape), but the cost of janking up the Tooth mana base in the pre-Sakura-Tribe Elder world for nothing more than four Electrostatic Bolts was nevertheless not inconsiderable. Abstractly speaking, a Tooth and Nail player would have preferred to have some man lands or even a lone basic Plains over those Mountains, run something more homogeneous – like Viridian Shaman or Tel-Jilad Justice – over the Bolts themselves. The problem was that not enough players were on board with correct thinking for this narrow format, so ripples were sent wildly from non-Affinity decks into the innocent bystanders of other non-Affinity fellows.
On top of that, there was the issue of the dissenting non-Vial Affinity decks. I don’t know why all the way to Champs 2004 there were Affinity decks littered with Ornithopters and Paradise Mantles and colored cards of all kinds, but any such players were breaking what I consider the prime rule of deck selection: Never play a strictly worse deck when you can play a strictly better deck. (It is actually embarrassing how many noted writers post decks that are just worse in some way or another than the “accepted” version of the same deck, often proudly stamping their names on the decks – what is worse is how often the said writers are clueless about how bad their math is.) Now everyone who was playing two summers ago knows how draw dependent the Affinity mirror could be, with the winner often being the guy who drew the most Disciples, but that doesn’t mean that there weren’t long run advantages to be had based on picking the best version of this disgusting deck; if you didn’t have Vials, you weren’t it.
In any case, Mirrodin Block Constructed as a format never hit equilibrium. With fewer than the estimated 75% of players playing tuned Vial Affinity decks, you had huge ripples of randomness that allowed for chaotic matchup interactions. For example, every time a Paradise Mantle Affinity deck slew an anti-Affinity deck with his dog-but-close 45% matchup, that made the tournament as a whole more likely to be won by any of the remaining Vial players… The Vial players would have a slight advantage over the crazy Mantle players due to greater consistency with essentially the same bomb tops, and a potentially bad matchup – say an overload Freshmaker with Oxidize, Electrostatic Bolt, Molder Slug, and even Creeping Mold – was no longer a threat; for his part, that Freshmaker player might have to duke it out with a Pristine Green player who had no reason to believe he could beat Disciple of the Vault… but also nothing to fear from the unhorsed G/R mage.
Now Mirrodin Block was a pretty special case. The Deck to Beat really was the best deck, and format never got to a level of maturity where inefficiencies could be exploited. Contrast that with Kamigawa Block Constructed one summer later. The Holy Knut will tell you that Gifts Ungiven was more dominant than Affinity, and that might be the case (though I think the statement is a myopic one); what is important is that unlike Mirrodin Block, Kamigawa Block was a land of strictly defined archetypes with very specific inefficiencies that could be exploited. On one end of the spectrum you had what amounted to hyper aggressive swarm decks (Black Hand and White Weenie) that played like mirror images of one another, such that late in the season, they even shared the same mana flashpoint for Spiritcraft breakers. On the other side of the format was Gifts Ungiven (and to a lesser extent Mono-Blue) for slow decks, with essentially nothing in between. This was a remarkably easy field to prepare for because the available cards to play were so powerful… and the battle lines were so specific. If you could figure out how to beat both White Weenie and Gifts Ungiven, essentially any decks falling in between became automatic because, accommodating as it might be, the card pool was not large enough to present many surprises. The solution to Mirrodin Block, even if it had hit the equilibrium point, would not have been as easy; in Kamigawa, Mono-Blue and Black Hand – decks at opposite ends of the spectrum – played at least some of the same breakers. Dominate the tempo of the faster deck with Shizo and Jittes, and it is implied that you can race the slower… But what about Mirrodin Block? What is the equal and opposite of an artifact?
That’s why I thought it was right to run the Vials. There was no way to get a real edge on a draw dependent format other than to hope the opponent would tap two artifacts for a Night’s Whisper.
Preying on Established Decks
For me, all deck decision making flows from the idea that I want to prey on format inefficiencies for an advantage, and that usually means not playing a highly expected deck, even if I am not playing a deck of my own design. For a long time, I played decks like U/G Threshold, and had what I thought were good reasons for doing so. I knew going into Sullivan, Nimble Mongoose, and Sullivan that my deck was good against The Rock, U/G Madness, and Psychatog, but that I would have to get lucky to beat combo. Not surprisingly, I missed Top 8 by losing a hopeless match against Enchantress after getting a little unlucky in the Madness “mirror.” If you want to win tournaments playing decks under this model, certain things have to go your way:
First of all, you have to be right about the pairings. I’m not talking about crossing your fingers and hoping you don’t face, say, Affinity in Mirrodin Block or Rebels in Masques Block. That’s just idiotic. If you do an accurate analysis prior to the tournament, you can usually figure out what the minority deck is and make that your bad matchup. In Sullivan, Nimble Mongoose, and Sullivan, I accurately figured out the pairings and lost to combo exactly once; the problem was that I dropped a round to Madness, which was a favorable matchup. This implies that…
Second, your percentages have to hold. The whole point of choosing a minority, unexpected, or otherwise “rogue” deck is that you have certain expectations about how your deck fares against a reasonably formed metagame. If you are right about your potentials and you hit the decks you expect, you should win the tournament… provided your estimations about who beats what are correct. This is of course a gigantic “provided” because players of oblique strategies very rarely have realistic initial expectations. For example, earlier this week I planned to go into tomorrow’s PTQ – which I expect to be covered in Ichorids – with Sitting Dead Red updated for Extended with Goblin Ringleaders and Matrons. I thought to myself “You know what’s good against a 3/1 haste? A Goblin Sharpshooter.” I was so sure of the advantages I had that I almost didn’t test. I tested Sitting Dead Red for Regionals 2004 and remembered the deck to be superb in a field that included Skullclamp helping out Disciple of the Vault. Surely a Jitte update on the Red Deck’s part and a Shrapnel Blast surprise out of Goblins of all decks would punish a format where everyone is on 17 turn 1, right?
Wrong. There’s A Reason Goblins is off everyone’s radar.
My teammate Tim McKenna pounded me with Ichorid. I won nearly half the matches, but you don’t play an under-powered surprise! deck to win 45%. If you’re not dominating the prey, then you have no business showing up with your clever sixty. Sure, I had the Ichorids under control most of the time, but the flying Psychatogs were a bit of a problem. In sideboarded games when I had Pithing Needle online, the Togs didn’t kill me, but the whole “draw three cards per turn” thing got pretty annoying for my team of 1/1 and 2/2 Goblins… especially as Tim had sided in Darkblast (a card I didn’t myself play last week). He would eventually win with something… Zombie tokens were many times sufficient.
In sum, sometimes you have to get that sass beaten out of you.
General Rules
I am almost constantly accused of playing “fair” decks. If you go over the original Building Broken Decks, you will see the articles littered with 1/1 creatures for G, 5/5 creatures for 2GG, and Disenchant variants aplenty. The sickest list I presented was a Nether-Go with main deck Submerge if you can believe it. Many of the decks in that series had above average percentages for their formats, but they were not “broken” as a general rule.
It is difficult to say which is the better deck overall, Boros Deck Wins or Heartbeat of Spring combo. Boros has won many more tournaments and boasts the more threatening early turns… but Heartbeat is generally considered to be the favorite head-to-head (not to mention that every deck from NO Stick to Slide is terrified of the “soft” combo pairing, regardless of what their masters tell you). It is not difficult to analyze the decks one against the other in terms of fair v. not so. One deck summons 1/1 creatures for one mana and 2/2 creatures for two. Surely it has a three point burn card for three mana and generally efficient drops… but it also has eight slots dedicated to Stone Rain. Heartbeat on the other hand draws three or more cards in a turn, pumps out mana in the mid-double digits, and can kill the opponent without dealing a single point of damage via one turn’s work. Consider also U/G Threshold v. Miracle Grow. The decks share much in common: their colors, their Werebears, half their overall strategies. Threshold can conditionally produce underpriced drops… Miracle Grow does the same but also operates on one mana and can stifle any strategy at will while simultaneously braining with lethal two-drops. Fair. Unfair.
There is no absolute difference between fair and unfair in terms of win percentage. Go back to Boros and Heartbeat. Heartbeat is unfair. It is broken. It is arguably the most powerful deck in the current Extended… by contrast Boros is considered the lowest powered deck in Extended. But again, that is not to say that you wouldn’t have a better time of it with Boros.
I like fair decks because I can understand them on a fundamental level much more easily than unfair decks. They work like decks worked when I learned to play Magic. If I play a Buehler Blue variant, I know I won’t have to mulligan very much; my next land drop is never in question. My deck will do the same thing, game in and game out, until I hit my win condition, at which point I will clock the opponent for a surprisingly swift 7 turns before cashing in the W. Unfair decks are not like this at all. Last weekend I played a deck with seventeen lands, eleven fewer than the Mono-Blue I just discussed (as a point of reference, in my basics articles I talk about avoiding the temptation to run fewer than 24 lands). The Ichorid deck has countless catalyst cards that do essentially nothing, or nothing efficiently… Putrid Imp, Zombie Infestation, Golgari Grave-Troll… One is a 1/1 for one mana that sometimes demands I pitch my grip to get a pathetic second point in play; one card gives me a bear for the cost of a bear… and three times the cards; the last is a nigh-uncastable five-drop in a deck where a second land drop many times doesn’t occur.
It is not difficult to identify the more steady deck between these two. One of them does the same thing time and again, and does it well. The other deck is given to higher highs… but also more miserably depressing lows and heartbreaking mulligan situations. Unless we are talking about High Tide – the broken combo deck that masqueraded as the best Draw-Go deck in the history of Extended – most broken decks play low on land, are crammed full of extra meaningless “stuff” and rely on cards that do nothing or less than nothing in and of themselves.
Regardless of fair v. unfair or any other normative labels we put on decks, the consensus worst way to lose in Magic is by being manascrewed. The best of the fair decks simply get manascrewed less than every other kind of deck. The reason I played Undead Gladiator in L.A.? It cycles. Therefore I will hit my next land drop. Me rikey rand drop (especially as the kill cards I played cost 7). When in doubt, I like to play the fewest number of colors possible (because that takes a variable out of the equation), and I generally play more land than is necessary, the opposite of Alan Comer’s less.
As much as possible I like to play cards like Magma Jet, Peek, Scrabbling Claws, and even the much-hated Sensei’s Divining Top. I like cards that do a little something in the short term but also get me my next damn land drop in the opening turns of the game. That is the reason I played Remand in Mono-Blue at States. No, I didn’t think Remand was the best counter in the world (even if it was hot for counter wars)… But I knew I could steal a turn and potentially set up my third drop for the Hinder for the next.
Don’t be an Idiot
The one thing that annoys me to a greater extent than any other experience I have ever had is the especially loud substitution of incorrect or inexplicable opinion over fact, logic, or common sense. Such affronts offend to different extents depending on context. The most obvious in modern American society is the idealistic young liberal, probably a college-aged and crunchy Caucasian, attacking the current president or a perceived Right Wing agenda via spurious arguments or fabricated data. Now obviously there are many places where we can criticize any administration – certainly the current one – on any number of fronts… which is what makes select attacks so distasteful.
The classic is of course the much popularized (and false) listing of Presidential IQs over the past 50 years, a metric that showed William Jefferson Clinton besting genius IQ by 30% and George W. Bush commanding precisely one-half Clinton’s purported score. Now Clinton was a Rhodes Scholar and genuinely charismatic person… I wouldn’t doubt him to be quite bright… but as a reasonable person I find it difficult to believe he eclipsed Albert Einstein’s estimated IQ by between 22 and 42 points. Despite the fact that, like anyone else who has ever heard him speak, I have noticed many communication hiccups from the President – and even studied the footage contrasting his first gubernatorial race, when he was a sharp and eloquent debater, with the current incarnation – as a reasonable person, I can’t believe that W’s IQ is closer in score to a bona fide mentally retarded person than the margin by which Clinton kicks “genius” down the stairs. In fact, to buy into numbers like these, I am pretty sure I would have to forget everything I know or have reasoned and submerge myself completely in rhetoric.
Completely sublimating good sense and even numbers in favor of “feelings” about decks is exactly what many Magic players do when choosing or making arguments in anticipation for the next tournament.
I am pretty sure that the number one problem plaguing Magic players as a whole is self-deception. “I am fine with that matchup” is poison to my ears. No one is fine with that matchup. You should be wetting your bed and calling home to mommy when that matchup sits down across the table from you. You should beg for a mulligan… and then not take one yourself. You should play so tightly in every other pairing you get the rest of the day, and run so quickly from table to table, that you can avoid that matchup at least until the Top 8, where hopefully the hated opponent will be manascrewed, or better, loses to someone who spends all his luck before he reaches you.
It’s Not That Bad.
Yes. Yes it is. It is that bad, you self-deluding moron!
The only way you can overcome the bad matchup is by 1) not having it at all, 2) dedicating yourself wholesale to its destruction, or 3) crossing your fingers and hoping the boogeyman doesn’t catch up to you. I played Friggorid last week, knowing that not only did I have a bad Affinity matchup, but that I would face it; 1/3 of my played matches were against Arcbound Ravager. The only solution I could come up with was to spend 13 of my 15 precious sideboard slots on defeating Ravager… I chose option 2. A realistic outlook about that matchup allowed me to take both pairings.
edt is famous for choosing option 3. “Pretend it doesn’t exist,” he used to say. Don’t deny that it’s bad… just don’t play against it. edt’s Michigan contemporary Aaron Breider used to play this R/W Iron Phoenix deck, the original Peace of Crap. Aaron would dump his Shard Phoenixes to Peace of Mind to get out of Trix range, and had more than enough elimination to stop most straightforward offensive strategies. He could not, for the life of him, defeat Survival of the Fittest. Ultimately, Aaron did the only thing he could: He didn’t deny that the matchup was bad, sort of pretend he had some plan to win; instead, he played in so many qualifiers, made so many Top 8s, that eventually all the Survival decks were in the other bracket and he could roll over soft and straightforward opponents until the Blue Envelope was his.
Rare is the deck that really and truly has no foil. Some decks, like Mono-Blue in the current Standard, have many enemies… but they just didn’t appear until the deck(s) were already entrenched at the top of the hop. Let’s ignore those decks for now; instead let’s focus on the opposite… You have one weekend to qualify for Honolulu, which I am pretty sure is the most desirable possible Pro Tour to be qualified for, given the current PTQ prize payout system, locale of the tournament, and actual Pro Tour format (Standard… which is essentially optimal for controlling your own destiny if you are willing to put in the time). You know you will have some foil… So what do you not want to do?
1) Ignore Ichorid. It’s not going to be mono-Black tomorrow, but it will be rare for someone to win a PTQ without either being or surviving the 3/1s.
2) Have no plan against Boros. It’s the most popular deck in the format in terms of number of decks making Top 8s… If you have a bad matchup, you don’t want this to be it for one simple reason: Most of the Boros decks that make Top 8 don’t win… but in order to win a Blue Envelope, you will often have to get past the Hound of Konda and friends.
3) Scoop to NO Stick. Where I live, some kind of U/W Control variant has been in three of the last five finals and four of the last five Top 8s. One old school U/W Control and two NO Sticks have taken the Blue envelope three of the last four PTQs, including NO Stick at the last PTQ at Neutral Ground (the season ender is at the Ground tomorrow) and a NO Stick mirror the week after.
There are lots of players in a tournament; even if you play 11 rounds, you will not play close to 200 of them. Therefore there is a huge possibility you can dodge all Ichorids. You might even be in the opposite bracket of the Boros decks in the Top 8… But lose to NO Stick? I know the sample size is small, but I wouldn’t want that as my bad matchup if I intended to actually qualify for the Pro Tour in the Northeast.
Sending a blind eye to your bad matchups is a fairly innocuous brand of self-deception. The downside is you just decrease your chance to qualify. However disagreeing with basic math is… I don’t even know the words.
There have been many discussions on the forums in recent months about which decks are good and which decks are not. One deck was said to be not as good as another deck of the same color because it had only three outstanding performances (statistical best win % at a large Standard four-slot PTQ, undefeated record at a premiere multi-format event, and Top 8 and second highest statistical win % at a major National Championships). The argument was that the less popular deck in question put up a total of one Top 8 versus countless National Championships level Top 8s. The reason this is an idiotic statement is that the number of players piloting the respective decks was differentiated by a measure of 1:20 or greater over the course of several tournaments… of course the generally accepted version of the deck would show up in more Top 8s.
Most recently there was an argument on the forums about Champs era Mono-Blue control from 2004 versus Affinity. Mono-Blue control could not have been a strong deck, one poster commented; it won only three Champs. Forget about the fact that Affinity was played conservatively at a 50:1 ratio to Mono-Blue control; no reason to bring that up. Even if Mono-Blue control advanced to Top 8 at ten times the rate that Affinity advanced, the hated artifacts would look much more successful based solely on the raw number of Top 8 appearances, such that it would be impossible for Mono-Blue to “look good,” regardless of how strong it was at winning matches.
Sometimes the best deck does the best at a tournament. Billy Moreno Madness Tog deck with Jitte, Gifts, and the now ubiquitous Loam engine was considered by many to be the best deck of Los Angeles and Billy finished second with it. At U.S. Nationals, it was pretty clear that BlueTooth was the best performing deck… It made up nearly half the Top 8, with 25% of its weaponeers making the break. What players sometimes ignore is that the facts are also clear when big decks played by many respected mages fail to make Top 8, or even leave the Swiss rounds with winning records. “It just had an off day,” they might shrug… When you or I get consecutive bad pairings, or fall to a manascrew, that we can blame on ill luck or a bad day; when 1/4 of a tournament, many of them Pro Tour Champions, all fail to do remotely well with a consensus powerhouse deck, that is grounds, methinks, for reevaluating the wisdom of a particular title or decision.
The last thing I want to say about this process is that sometimes, a deck shows up in testing that just beats everything. When that happens, it is a sure sign you should play that deck (Jon Finkel chose Napster after only one night of Standard testing for US Nationals 2000, and fairly early in the playtest process; he devoted all the rest of his practice time to draft, which helped him greatly towards his Day One perfect record). When I was testing for Regionals 2001, I had this deck Fires of Yavimaya and got “the Fix” (turn 1 accelerator, turn 2 Fires, turn 3 Blastoderm, turn 4 Saproling Burst) over and over. Poor Jon Becker. His Evil Eyes of Orms-by-Gore, which were meant to be very good against the said Fires, were rolled over by the Fix. “This deck is absurd,” I thought. “I always get ‘the Fix.'” My conclusion was to play a deck that beat the said Fires of Yavimaya.
I lost to it in Round One.
Summary on How to Pick a Deck to Play:
Never break the Prime Rule- Never play a strictly worse deck when you can play a strictly better deck.
1. Analyze the field; identify inefficiencies by which you can gain an advantage against that field.
2. Established decks are useful for multiple reasons; they give you punching bags to test against and a good idea of what your opponents will play. It is correct to play an established deck specifically when playing a non-established deck would yield a lower likelihood for victory (for most players, this is “most of the time”).
3. One empowering aspect of so-called “fair” decks is that their development is “normal,” as in parallel to the land drop and turn efficiency skills we all learned as burgeoning beginners. Playing “fair” decks with early game velocity and high land counts allows you to control how you lose. Due to potentially higher mana consistency, fair decks may have higher median values but not necessarily higher mean values when compared with “unfair” decks in the same format.
4. Self-deception is dangerous; close-minded approaches to decks one does not understand, especially in the face of statistics, is a sure sign of irrational deck choice decision making and potentially general incompetence.
5. When a deck can’t be beaten, that tends to imply that it can’t be beaten by you.
LOVE
MIKE