Objective
This is intended as a summary guide laying out fundamental principles of good deck building. This is not intended to be comprehensive nor is it intended to introduce cutting edge concepts, just summarize established wisdom. A player reading this ought to be able to learn the established best practices of deck building.
Who is this for?
This is for players with some magic experience who are looking to take their deck building skills to a higher level.
This will probably be a good read for true novices also but there is an assumption of familiarity with core game concepts (like Tempo and Card Advantage and Mana Curve) that a true beginner might not have. This might be a good read for experienced players also, but it will not teach you anything you don't already know. You may enjoy reading it anyway because sometimes seeing something written in somebody else's words makes it easier to think about.
The real target audience for this guide are players who are familiar with the the game but have not mastered deck building. My assumption is that a player who wants to build a better deck wants to build a deck that wins more and is more competitively viable. If that is not your goal then this guide is not for you.
Who am I?
I'm Metamorph. No I will not tell you my real name or DCI number. Thats for my real life friends only. I'm an experienced magic player with almost 17 years under my belt. When I started playing the core set was Revised Edition and the two expansions you could buy in stores were The Dark and Fallen Empires. The first box of cards I ever bought was Ice Age, and I bought it because I coveted the awesome new artwork on Icy Manipulator in that set.
I first started playing competitively when Mirage/Tempest was the Standard environment and I've played competitively off and on ever since. I've had some mixed success but I'm certainly not a pro. I've Top 8e'd PTQS twice, a $5K open event once, and back when I was a kid the Junior Super Series multiple times. I've top 16ed more than 20 PTQ/$5K level events though so I'm quite familiar with what competitive play is like and how to produce a winning record, though I'm not quite good enough to play at the highest level.
I play lots of different types of Magic and I play on a budget most of the time so I absolutely do understand casual play and budget issues. But I won't be talking about that here. Or anywhere really. I stick to pure strategy. Thats what this guide is.
A Note on Format
This is intended for Standard Constructed Magic. Format provides context and context is the framework of meaning. Most of what I will say here is not at all true of formats besides Standard. If you are interested in learning how to build decks for Legacy, Vintage, Commander or any other format then this post will be of only minimal usefulness. Finally, the orientation of this post is for Standard in its modern form (meaning effectively Standard since about Invasion block onwards). Standard from the bad old days was very different and should not be used as a context for this discussion.
Summary
1. Good Decks Play With Good Cards
2. Good Decks Have Good Plans
3. Good Decks Have Good Mana Bases
4. Good Decks Respect Their Opponents
5. Good Decks Have 75 Cards
6. Sometimes Even Good Decks Are Bad Choices
7. Sometimes Your 'Good Deck' Isn't
Good Decks Play With Good Cards
The concept is simple but the idea itself has many wrinkles to it. The power level of a deck is directly related to the power level of the individual cards that the deck is composed of. Synergy enhances this further, but at the most fundamental level the standalone power of each card in the deck is extremely important. While it is possible to build decks where the Synergy of card interactions allows for otherwise weak cards to become powerful this is an exceptional case and must be justified independently. The most common case is that independently powerful cards are the best choice for building a deck.
This raises the crucial question: how can I tell when a card is independently powerful?
Thats a deep question in its own right and there is no easy answer when it comes to card evaluation. However, there are some very useful and well established guideline principles we can use to get it right most of the time. Not always though. Sometimes things aren't as they seem for one reason or another, but if we make our usual assumptions and consider the card in a typical Standard environment.
Efficiency
Efficiency is a rough measure of how much bang you get for your buck. How much effect does a card provide given the resources you spend on it. Every card requires resource expenditure, there are no exceptions. Even if a card costs zero mana and zero alternative costs it still costs the card itself. This is something that is frequently overlooked by inexperienced deck builders. Always remember that at minimum a card costs the card itself and there are many cards whose effects simply don't justify that cost.
How do we measure efficiency? There's not standard method. Its very contextual. A card is efficient if it produces more then other similar cards in the metagame. Sometimes we get "strictly better" scenarios where it is obvious that Runeclaw Bear has a better body for its mana cost then Goblin Piker. But how do we know that neither one of them is efficient enough for constructed? No easy answer. Generally its because the 2/2 body itself is not particularly worth spending a whole card on. This is a very good example of why considering the card cost is so crucial. We are readily willing to play other 2/2 for 2 mana cards if they provide an additional effect. Grand Abolisher or Leonin Arbiter or Kalastria Highborn or Goblin Wardriver are all readily playable for the same converted mana cost. The additional effect on these creatures is arguably the more valuable part. Would you play that effect if it was just on an enchantment with no body? Probably not. So really its the value we're looking at. We want to have it all. We want the additional effect AND the body.
I realize its not exactly useful to tell people they shouldn't play Runeclaw Bears. Lets extrapolate from the example and try to form a general principle: when evaluating a card for efficiency consider how much value the card provides. If a card provides just a body, could you instead play something with an effect AND a body? If a card provides just an effect, could you instead play something with an effect AND a body? This value maximizing notion of efficiency is a good guideline. We prefer to play with cards that are almost like built in 2-for-1s. If you can get a body and an effect at the same time the card is almost doing as much work as 2 cards.
The other notion of efficiency thats important to consider is raw mana efficiency. This is not a hard concept to measure. Lightning Bolt is the king of mana efficiency for burn spells. Nothing is better. Creatures like Leatherback Baloth are the kings of mana efficiency for creature body size. Mana efficiency comparisons are straightforward. Just look at the numbers. There's not much more to say except to emphasize that often it is incorrect to favor mana efficiency over card efficiency. In standard it is frequently the case that card efficiency is the more useful measure. Simple Example: Squadron Hawks is a 2 mana 1/1 (very mana inefficient) but is a 4 for 1 in terms of cards. Its one of the best creatures in the format.
Consistency
The other measure of card power that is useful to consider is consistency. Consistency refers to how likely a card is to be able to perform as desired under normal play circumstances. A card like Phyrexian Obliterator has very high mana efficiency but its restrictive mana cost makes it extremely inconsistently castable except in almost mono-black decks. Mana consistency is always related to what your mana base is capable of, so keep this in mind when including cards. You must be able to cast them consistently or they will be useless.
Mana cost consistency is not the only form. Some cards have functional dependencies that raise consistency issues. Any aura or equipment or creature enhancing effect (like Honor of the Pure) is functionally dependent on there being one or more creatures in play on your side of the table. This is a pretty soft consistency requirement (you usually do have creatures) but it still must be considered. A common mistake is to underestimate even soft requirements. If you are playing an equipment, could you have played another creature instead? That would add more power to the board without being functionally dependent on another creature. Does the equipment provide a powerful enough effect to justify using it instead of another creature? It probably has to be about as powerful as a Sword of Feast and Famine to be worth it. Thats a high bar to clear. Most creature enhancing effects are not worth it for this reason. I don't want to get caught up on this one example. There are many strategies that utilize synergy to make creature enhancement truly worth it (Tempered Steel decks, for example) but you should be aware that these synergy based strategies suffer from consistency issues because of the nature of the strategy. Its sometimes worth it anyway though.
There are other forms of functional dependency too. Any card that relies on you having draw another card first has a functional dependency. Be very wary when playing cards like this. Only the softest functional dependencies are consistent enough to be worthwhile. Very hard dependencies require ENORMOUS support to achieve consistency. If you're going to be running a hard dependency be prepared to build a dedicated combo deck, otherwise its probably not worth it.
To clarify the terminology: a soft dependency would be between categories. Equipment + Creature, Ramp Cards + Bombs, etc. A hard dependency would be between specific cards such as Joraga Treespeaker + Myr Superion.
Niche Cases and Role Players
There are times when it is not possible to only use the most efficient and consistent cards in the format. Maybe your colors simply can't do something efficiently. Maybe the format card pool simply doesn't have any consistent options. There are too many exceptions to list so my purpose with this brief section is just to make it clear that there are good exceptions to the general guidelines. Sometimes you just need a certain job done and have to work with the tools you have, even though by historical standards they are inefficient or inconsistent. We do what we must. However, if you find that your deck is including a large number of inefficient or inconsistent role players (cards that serve a highly specific purpose) then you might want to seriously question whether or not its worth it. If your strategy requires too many inefficient/inconsistent cards its probably not good enough. It might have worked in past environments but is not well supported in the current card pool. It might just not be a good idea, period.
Good Decks Have Good Plans
Magic is a strategy game. Strategy is about planning as well as about execution. If your plan is bad then you are at a huge disadvantage. Make sure you have a good plan before you show up to the fight. The classic way of defining your plan is by strategic archetype. Most players have heard of these before, we throw the terms around all the time without having a clear idea about what they really mean and the permutations of them are: Aggro, Control, Combo.
Mission Statement
A Mission statement is one or two sentences that describe in the most concise terms how your deck plans to win a game of magic. Every deck should have its own mission statement that is based on its archetype. Here are the general purpose archetypal mission statements, and a few examples of deck specific versions of them.
General Aggro Archetype: "play too many threats too quickly for my opponent to deal with them all before his life total has been reduced to zero."
specific example: "Deal as much damage as possible with my 1 and 2 mana red creatures. Finish the game with a second wave of red creatures or direct damage spells."
General Control Archetype: "prevent my opponent from executing his game plan. Then, when my opponent cannot defeat me anymore play a threat that my opponent cannot answer and win with it."
specific example: "utilize countermagic and creature removal to deal with early game threats. use card drawing spells to make sure I have more answers than my opponent has threats. then, when my opponent is exhausted play a Grave Titan and kill him with it in 2 turns."
General Combo Archetype: "assemble a combination of two specific cards as quickly as possible. the two cards, in combination, will produce an effect so overwhelmingly powerful that my opponent will lose to it."
specific example: "utilize cheap blue cantrips and card drawing effects to get Deceiver Exarch and Splinter Twin into my hand. Then, cast these two spells in combination to produce infinite Exarch tokens with haste and attack for the win."
Mission statements are very important. If you cannot produce a succinct mission statement for your deck then you probably don't have a good enough plan. How can you expect to win a game of Magic if you don't even know how you're going to win it?
There's more to archetypes and strategic plans then just what I've covered here. It could be a whole article series discussing the various hybrid archetypes (aggro-control, mid-range, ramp, aggro-combo, combo-control, multi-combo, etc.) but for the sake of preventing this already sprawling article from getting out of control i'll leave that discussion for another day.
Have a Plan B
No plan survives first contact with the enemy fully intact. Sometimes things just don't work out according to plan. Maybe your opponent had all the right answers for your attackers. Maybe your deck's own internal consistency issues screwed you and you just didn't draw what you needed in time. A good deck should be able to win even when things go a little wrong. If things go alot wrong even the best deck will probably lose, but having a fallback plan to cover some hiccups in your main plan is a very good idea.
What constitutes a good Plan B? Totally specific to each deck. No general way to characterize this. What makes a good plan B? Its not vulnerable to the same things Plan A is. If your Plan A is to attack with creatures then your Plan B should not also be vulnerable to creature removal. If your Plan A is to assemble a combo then your Plan B should not be vulnerable to the same things that disrupt your Plan A combo. Plan B doesn't need to be complicated or independently viable. Simple things will suffice. Maybe you'll just use Burn spells to finish the job. Totally valid plan B. Maybe you'll just get 20 points in with Manlands. totally valid plan B. maybe you've got a second, weaker combo in your deck to supplement the first. Totally valid plan B. Doesn't really matter what it is or even that its particularly good. What does matter is that you have a backup plan that still works even when your primary plan has failed.
Good Decks Have Good Mana Bases
This section can be short and sweet. The topic itself is deep and interesting but fortunately can be summarized in a few simple words.
On Mana Curve and Land Count
Mana screw will kill you every time. So will color screw. You can't play spells that your mana base cannot support. The single most common error I see from any deckbuilder is playing insufficient land. It is absolutely imperative that you play enough land for your deck, based on what the deck's mana curve is.
In general we can think of there being three typical mana curves: low, medium, and high. In practice there's a huge amount of variation but at minimum we can categorize any mana curve into these ranges.
Example of a low mana curve:
CMC 1: 8-12
CMC 2: 8-12
CMC 3: 4-6
CMC 4: 2-4
CMC 5+: probably none, maybe 1-3
Example of a medium mana curve:
CMC 1: 0-4
CMC 2: 8-12
CMC 3: 6-8
CMC 4: 4-6
CMC 5 or 6: 4-6
CMC 7+: probably none, maybe 1 or 2
Example of a high mana curve:
CMC 1: 0-4
CMC 2: 6-10
CMC 3: 4-6
CMC 4: 4-6
CMC 5: 3-5
CMC 6 or 7: 3-5
CMC 8+: probably none
most aggro decks have low mana curves, most control decks have high mana curves, most ramp decks have high mana curves. alot of decks end up with medium mana curves just because its fairly consistent. some decks are called midrange for no reason other then they have a medium mana curve.
here's a very basic rule of thumb for assigning land count to your deck. 24 lands is the default number for medium mana curve decks. if your deck has a low mana curve you can justify playing fewer then 24 lands. 22 works for most aggro decks. a very slim one (with lots of 1 drops) can get away with 20 or 21. if your deck plays fewer then 20 lands it has to have a curve that is lower than low. lots of free spells, maybe a mana cheating mechanic. if your deck has a high mana curve you definitely need more then 24 lands. 26 is the most typical number for control decks. ramp decks often play as many as 27.
On Colored Mana Balance
additionally we must ensure that colored mana costs are covered. it is usually correct to play as many quality dual lands as exist in the format that can fit into your mana base. some of the best decks of all time were good in no small part due to their mana base being composed almost entirely of dual lands. UB Faeries (a previous standard boogeyman, one of the best standard decks of all time) played 16-20 dual lands in its mana base. the advantages of this are profound. many have tried to come up with a mathematical formula for computing color ratio but there's usually some flaw with whatever formula you might come up with. here's a potentially useful, though flawed formula that you can use if you take care to make manual adjustments afterwards.
For each color in your deck count up every colored mana symbol in a spell's mana cost. Take the total number and divide by 2, round up. That is the minimum number of lands capable of producing that color of mana that your deck will require to cast its colored spells consistently.
What's wrong with this formula? It doesn't take into account turn priority. Maybe you have only a few green mana spells in your deck but they are very time sensitive and must be cast on turn 1 or 2 to be maximally useful (Birds of Paradise, for example). In order to account for this you must increase your green mana producers above and beyond what is strictly needed based on color symbol count. What else is wrong with this formula? it doesn't make any statements about lands that enter the battlefield tapped. some decks can support many EtB tapped lands and can play many more non-basic lands than usual because of it. control decks often have this advantage. Some decks have a much greater need to play their cards on the curve (meaning a 2 drop on turn 2, a 3 drop on turn 3, etc.) and cannot support many EtB tapped lands and must use basic lands instead or limit themselves to duallands that have a mechanism of entering play untapped. The above formula can be used to get a good baseline but it must be adjusted manually to account for the effects mentioned.
Other Concerns
Many decks include non-land mana sources, cantrips, card drawing spells, or other effects that assist with mana development in one way or another. This topic can get very complicated and sophisticated and further discussion is beyond the scope of this article. Just be aware that the rabbit hole is deep on this one and there are many exceptions to the rules of thumb given above. Each exception can be justified with good reasons though. If you don't have a good reason, you shouldn't break the rule of thumb.
If you're ever in doubt and feeling tempted to cut a land, don't do it. If anything you should add one more land then you think you need. You always lose when you get mana screwed but sometimes you win anyway if you get mana flooded. More is better. Don't shave lands just to fit extra cool spells in. You'll regret it.
Good decks respect their opponents
Magic is a 2 player game. Your opponent is not a goldfish. A common mistake is to assume that your opponent is less capable than he really is. This is inviting disaster. It is better to assume your opponent is MORE capable than he really is and to build for the most hostile environment you can think of. If your deck remains functional even in the face of considerable hostility then it is probably a good deck. If your deck crumples to enemy resistance then it probably wasn't good to begin with.
There are two main principles we can use to think about how a deck fairs against its opponent: Interactivity and Resilience.
Interactivity
Interactivity is a measure of a deck's defensive strength. When we "interact" with our opponent, what we mean really is that we're trying to screw up his plans by countering his spells, discarding his hand, killing his creatures, and otherwise just being a pain in his butt.
Why should you interact with your opponent? Isn't it better to just focus on the kill? Sometimes it is. There are some aggro decks and some combo decks that are at their best when they try to just kill as fast as possible and let the opponent do whatever they want. How often is that the best plan? Rarely. Sometimes even just a small amount of interactivity is enough to secure a win. A single crucial removal spell or counterspell is the difference between winning and losing a huge amount of the time. Interactivity is so important that its possible to base an entire archetype (control) around it.
The reason we should almost always attempt to interact with our opponent is because of variance. We cannot guarantee that we always get our best draws that will definitely win faster than our opponent. Interacting with the opponent smooths out this variance. If we can't win as soon as possible we must instead make sure our opponent does not win first. All we have to do is win before he does, not win before some arbitrary turn has elapsed in game.
There's a pitfall here though. Some players get into the habit of playing "answers for answers". they see that their opponent has some card that is particularly effective against their deck (a hoser basically) and think they can overcome it by playing a spell that defeats the hoser. Example: Dismember as answer to Kor Firewalker. This is not optimal strategy. An answer for an answer only works if you're holding it first. Its very time sensitive. If you draw your answer for answer too late it was ineffective and the damage was already done. Further, there's the risk of it being redundant. If you draw your answer for answer but your opponent does not draw his hoser then your draw was dead. A better strategy is to reinforce your game plan with resilience.
Resilience
Resilience is the compliment of interactivity. It is a measure of how difficult your deck is to effectively interact with. An effective interaction is one where the answer was efficient (in terms of mana AND cards spent) and timely (played in a relevant time frame). Resilient threats are hard to effectively interact with. Your opponent's ability to answer them is limited.
A good case study in Resilience is the issue of creatures vs. removal spells. This is a very common theme on the forums here and many players make the mistake of playing creatures that have no resilience against the most effective forms of removal. The fact is that most removal costs little mana (just 1 or 2) or if it costs more mana it is very card efficient (a sweeper, for example). For a creature to be resilient against this effective removal one or more of the following must be true
1) the creature must be cheap and individually unimportant. it is just one threat among many and there is another one right behind it.
OR
2) the creature must provide value regardless of whether or not it is immediately removed.
OR
3) the creature must itself be unusually difficult to remove. the most common effective removal spells must be of limited (or nil) effectiveness against it.
if a creature fails to meet at least one of those conditions it is not resilient against removal and is probably not a good choice, since most standard formats are filled with effective removal spells. These resilience concerns are the reason why Titans outrank just about any other large creature you might play in Standard. they are also the reason why aggro decks based on lots of small creatures are far more effective then aggro decks based on a smaller number of large creatures. there are always exceptions of course, but in general thats what we observe.
This is just one case study, the case of creatures vs. removal. similar analysis could be done for any complimentary pair of threat + answer. the best threats are usually the most resilient ones. it is often more important to be resilient then it is to be efficient. as mentioned previously, you only have to win before your opponent does, not before some arbitrary turn has passed.
Good Decks Have 75 Cards
For competitive play, one cannot ignore the sideboard. The sideboard is relevant in more games than not, at minimum half all tournament games include sideboarding, and realistically its closer to 60% (since more matches are 3 games than 2). The maindeck has to be good but if your sideboard isn't good then you're not really ready for competitive match play.
Specific sideboard analysis is the most valuable kind and it requires specific decklists and metagames to do, which is beyond the scope of this article. I'm going to focus on general principles and pitfall avoidance in this section so you can get the right idea and go from there.
Here's the basic, fundamental idea of correct sideboard building: you cannot board a card IN without also boarding a card OUT, so you must build your sideboard with this in mind. Often it is more important what you board out than what you board in. You might not have a dedicated hoser for a matchup but you're incredibly likely to have at least a couple of maindeck cards that are subpar in the matchup. If the only thing you use your sideboard for is replacing subpar cards with something with better performance then you're using your sideboard effectively.
Pitfall #1 - using too many hosers
A common pitfall is to jam your sideboard full of hosers against what you think the best decks are but you don't really have a plan for how to use them. Hosers are just hate cards and sometimes they work, but often they aren't even good enough because they don't mesh with your plan. If you took out cards that were useful and synergistic with your strategy and replaced them with cards that do nothing in particular for you other then make your opponent's life miserable, then you are not optimally using your sideboard. The goal is to win the game, not to grief your opponent. When choosing cards for your sideboard keep in mind that its more important that the card makes sense with your strategy than it is that the card is harmful to a specific opponent. You can't guarantee what opponent you'll face, but if you have cards that are good in your deck in general you can always use your sideboard to tune up the deck between games and make it slightly better than it was game 1, based on what your opponent is playing.
There are exceptions of course. Sometimes a deck is very dominant in the metagame and the best bet is to just hose it. This is rare though. It usually requires a kinda degenerate deck with an unusual strategy, usually combo decks of some kind (graveyard combo decks like Dredge or Reanimator are notorious).
A pretty good way to build a well integrated sideboard is to sketch out multiple alternative maindeck configurations that you'd be happy with. Each configuration might differ by 4-6 card choices. Let the overflow live in the sideboard and swap between configurations between games, based on which configuration you think will be best. If you build a sideboard like this you'll avoid having narrow cards that don't mesh well with your deck. Every card in the sideboard will be independently useful so you'll always be able to board at least a few cards in. This is better expected value then sometimes boarding in a really nasty hoser, but often just having nothing to board in.
Pitfall #2 - fighting the last war
There's another pitfall thats really important to avoid. Always remember that you are boarding against your opponent's game 2 deck, not his game 1 deck. You never face his main deck with your sideboard. you always face his maindeck + sideboard with your maindeck + sideboard. This pitfall is best illustrated with an example. Here's one I see people make ALOT.
Example: You are playing UW Control vs. opponent playing RDW
In game 1 you managed to stabilize the board around turn 5 by resolving a Day of Judgment to sweep his creatures. You were at 6 life when this happened. Unsurprisingly your opponent won the game by burning you out with a Bolt and Incinerate he had been saving in his hand.
Here's the mistake: you're playing Leyline of Sanctity in your sideboard and you want to board it in for game 2 because you lost to direct damage game 1 and figure you would have won if only your opponent couldn't have finished you off with burn.
This is a mistake because you're fighting the last war. What did your opponent do with his sideboard? He boarded out his burn for haste creatures. From his perspective the game looked like this: I was beating my opponent down something fierce but the DoJ landed just in time. I lucked out on that one and had the burn, but thats kinda loose. A Hero of Oxid Ridge would have been perfect though. Incinerate out, Hero in. This is good strategy from the red player.
From his perspective the burn spells are mostly used to remove blockers so his men can connect for damage. They can also be useful to damage players but a persistent threat like a creature is even better, its worth more damage for the card and is far preferable against a control player who might be running countermagic or lifegain which would completely invalidate the burn. Furthermore, the control player has few blockers anyway so the burn just sits around in his hand. Why did he have 2 burn spells waiting? because there was nothing to use it on before anyway.
Your Leyline whiffs. Your opponent boarded out the thing the Leyline protects from. The Leyline doesn't do anything to help win the game and produces no value, its just a narrow hoser, and now you're caught short because your opponent went the other way on you. He boarded out the burn and brought in haste creatures. You know what you should have boarded in? Creatures that can block. He boarded out cards that can remove blockers and boarded in more cards that are stopped by blockers. Furthermore, creatures wouldn't just be narrow hosers, they'd be useful no matter what your opponent draws. What you should have boarded in was something like Kor Firewalker, or Timely Reinforcements, or Wall of Omens or Spellskite or whatever.
I hope the example illustrates the point. If you use independently useful cards in your sideboard, and keep in mind that you are facing your opponents game 2 deck (not his game 1 deck) then you will have an effective sideboard.
Sometimes Good Decks Are Bad Choices
Its important to recognize that all decks have limitations and no deck can do everything. The best you can hope for is to have a deck that is powerful and consistent. It can execute a game winning plan that is difficult for the opponent to disrupt, while interacting enough with the opponent to make sure your plan wins before your opponent's plan does. Thats all you have to do to win in magic. You don't have to handle every situation, you just have to execute your plan before your opponent executes his plan.
What if its still not good enough? What if your plan doesn't win even when you execute it successfully before your opponent executes his? Yes. this is entirely possible. You can seriously just get trumped outright by a deck that has a strategy that utterly defeats yours. In well balanced metagames it doesn't happen very often but there are some classic historical examples. Here's one:
You are playing a mono-green Elves deck and your opponent is playing RDW. Your elves deck is pretty good in its own right. Its fast, its consistent, you've steamrolled multiple opponents and tend to do particularly well against certain types of control decks that can't keep up with your rapid board development. However, you've got a bad matchup. I mean a truly terrible awful shamefully bad matchup against Red. You lose this matchup like 80% of the time. The red player burns up our elves and attacks with his guys. Your early board development is completely thwarted and your normally awesome Overruns are deadweight in your hand, you don't have enough mana to cast them (because your mana guys got burnt) and you don't have enough creatures on the board anyway (again, burnt). You're getting bashed for 4 or 5 points a turn and don't have anything big enough to block. What do you do?
You might think you should try to address this terrible matchup by changing your deck. The elves are too puny and too vulnerable to burn so clearly you need to add bigger guys to the deck to fight back with, right? No. Wrong. Your manabase can't support bigger guys because you've tuned your deck to run on just 20 lands (and about 12 mana elves). Your deck is designed to swarm with 1 and 2 mana creatures and then cast Overrun for the win. The whole plan is supposed to execute between turns 1 and 5. What are you gonna do with 4 or 5 mana creatures? Even if you draw enough land to cast them (you won't, but lets just assume) what are you gonna do with them on turn 5 anyway? Nothing. Its already over by then. Too little too late. And now that awesome matchup you had against control sucks too because Control can easily deal with the big mopey 4 mana creature. You managed to ruin your good matchup without fixing your bad matchup.
There are many similar examples. The moral of the story is that a deck can only do what its strategic design and mana base allow it to do. You can't do everything and shouldn't try. The most important thing is to maintain strategic focus and let your deck be good at what it does. You'll be rewarded when your opponents are not able to deal with your strategy. However, opponents can easily deal with a weak version of any strategy just by overpowering you with their superior efficiency, consistency, and strategic focus. Don't sacrifice these fundamentals to cover your weaknesses. Stay focused and you'll win when you ought to. The point of deck building is to maximize strategic advantages to produce game wins. Focus. Keep your advantage maximized. Focus.
If the metagame is incredibly hostile to the strategy of an otherwise good deck then the correct choice is usually to just play a different deck. There's no shame in that. Good decks are still good even when they aren't the right choice. If you want to win though you need to make the right choice. Choose the best good deck for the metagame to maximize your wins.
Sometimes Your Good Deck Isn't
This section will act as a kind of punctuation mark on this article. There's one more thing I have to say about deck building, and I think its the most valuable advice I can give. I hope you take this to heart: most ideas don't pan out. The failure rate of deck ideas is exceptionally high. I throw away close to 90% of my skeched out decklists. I throw out most of them before even sleeving them up for playtesting. Of the decks I do get around to actually playtesting I still throw out most of them. There are times when I throw out ALL of the deck ideas I have and just play an established deck because its the right choice.
How do you know a deck idea is a failure? Start with the theoretical analysis.
Go through the deck and ask yourself "are these cards good enough? are there any weak cards in here that shouldn't be? are too many of the cards weak?" thats a stopping point right there. if the deck is full of weak cards its probably not gonna work. Remember that efficiency is not the only measure of power. Consistency is probably an even more important measure. Many decks have incredibly powerful best case scenarios but are on average extremely weak because they are inconsistent. Keep consistency at the front of your mind when evaluating how powerful your decklist looks.
Go through the deck and ask yourself "is it obvious how this deck wins?" If not, scrap it. A random pile of cards (even good cards) with no plan is not a good deck. Remember that many good plans are pretty simple. But you have to be realistic. If your plan is unrealistic the deck is probably a failure.
Go through the deck and ask yourself "does the mana work?" If not, you just can't play it. Sorry. Mana base is the most important constraining factor on deck construction. If you can't cast your spells, nothing else even matters.
Go through the deck and ask yourself "will this deck be able to actually win 2 out of 3 games in a match?" Some decks might do very well in game 1 but not actually be viable for competitive play because they are SUPER vulnerable to commonly played sideboard cards. You get nowhere winning every game 1 and losing every game 2 and 3. If a deck has a distinctive game 2 vulnerability, is its game 1 advantage HUGE to make up for it? If its not the deck is probably a failure.
And finally, ask yourself "does this deck work in the expected metagame?" As mentioned in the previous section, even decks that are good on all the fundamentals are often bad choices because the metagame is filled with bad matchups for them.
Remember that there's no shame in abandoning an idea. Its a skill in itself to know when something isn't working out and when your time would be better spent doing something else. Most ideas are flawed and its important to go and look for the flaws. Don't get caught up in the initial enthusiasm when you think of something cool. Let it simmer. Sleep on it. Take a look at it again tomorrow with a more critical eye.
Very nicely done. I've been waiting patiently for this. I will look forward to it being stickied.
The information is very useful for people who don't understand some of the basic deck building ideas like consistency and making sure that you have the mana to make plays.
It's reading articles like this one that gives me hope for the magic community as a whole. While some people are flaming and writing stupid comments, you take time out of your busy day to write this amazingly constructive article. I applaud you, good sir :smileup:.
Good work so far. I especially like your resilience section, for illustrating the old dies to Terror issue. The first point "1) the creature must be cheap and individually unimportant. it is just one threat among many and there is another one right behind it." is often missed. More so than just cheap it has to be efficient/consistent (basically your first section) and one of many. This redundancy argument is often lost and people focus on the latter 2. This argument takes a more specific look at the format and it's why a deck of hasty 2/2's can win at a certain threshold.
Any time we go back and look at efficiency and consistency on why a card is good requires that look at the format which I'm glad you preface to your points in the first section. Basically look at what you'd expect in a typical Standard Environment as a starting point. Every time of card evaluation always comes back to this so this type of discerning of the card pool might be the greatest skill that comes with experience from deck builder's perspective.
Private Mod Note
():
Rollback Post to RevisionRollBack
Current Decks: GWUKnightfall Modern UWTempo Legacy UGRBurning Wish Cobra Vintage
Very good thread, great read. The one thing I feel that should be added to your section on whether or not a creature is playable is whether that creature can simply win the game on it's own in a timely fashion if it's not immediately removed. Creatures like these generally force your opponent to spend all of their resources dealing with that threat and prevent them from executing their game plan for a turn or two in order to find a removal spell, and if they can't, they will die to it or the impact it left on the game will be enough to end it. Creatures that fit into this category that don't fit into your other categories include Hero of Bladehold, Steel Overseer, Fauna Shaman, etc.
I know that I am building a lot of decks you would consider BAD (I remember you banishing any possibilities of playing Mono Blue Control), still I like your article a lot. I'd only say you could write something more about philosophy of gaining card-advantage, many people don't understand this concept.
Not to change the subject, but please tell me what you gain from playing Mono Blue Control in Standard at the moment as opposed to Blue/White or Blue/Black or even Blue/Red?
I really like your Mana-Curve section. It helps out a lot to someone who just runs 21-22 lands for every deck. Aggro, Control, Jank whatever... now I kinda understand how it works.
Private Mod Note
():
Rollback Post to RevisionRollBack
To the people that say that a card needs to be a higher rarity because of Limited... I hate you guys so much. I present to you with this.
Very good thread, great read. The one thing I feel that should be added to your section on whether or not a creature is playable is whether that creature can simply win the game on it's own in a timely fashion if it's not immediately removed. Creatures like these generally force your opponent to spend all of their resources dealing with that threat and prevent them from executing their game plan for a turn or two in order to find a removal spell, and if they can't, they will die to it or the impact it left on the game will be enough to end it. Creatures that fit into this category that don't fit into your other categories include Hero of Bladehold, Steel Overseer, Fauna Shaman, etc.
This is actually a good point. Fauna Shaman is the standout in that list. Steel Overseer and lord effects in general require redundancy usually as well. Tempered Steel has it's namesake, Overseer, and Signal Pest.
Hero of Bladehold is sort of in the middle. As it requires other bodies and redundant pressure to be able to do anything other than die to the first removal spell. Some people would consider Hero only borderline playable.
Lotus Cobra is a creature that can provide zero value if removed immediately but represents and impending threat. I suppose though it has redundancy in ramp spells and other ramp creatures.
Fauna Shaman though works fine without redundancy and is very threatening. It's goodness seems to be directly proportional to the speed of the format, the power of creatures worth fetching, and the average CMC of removal spells that see play that can remove it being 2 or greater (or 1 and provide card advantage). It's a lot more like Jace or Birthing Pod in that it is a CA filter.
I think what's missing is an adaption of the 2nd rule.
It should read something like:
2) The creature must provide value for it's time in play that exceeds the tempo loss of it's removal.
Like if you untap with certain cards they are devastating. But then look at Consecrated Sphinx. It's sort of middle ground. Does nothing if it is removed immediately does something if it survives to your opponents main phase. It doesn't really fit into 1 or 3. Definitely it's value is it's ETB even if it doesn't have an ETB exactly. It continues to provide even greater value the longer it lives. Fauna Shaman is like this as well but it takes longer to see value. This is the key difference between a creature like this and Primordial Hydra or even Hero of Bladehold as a removal spell largely sets back the value even a turn or so later.. you are only gaining value off the opponents life total or if they attack into it and their creature dies (bladehold does leave some soldiers around assuming you are attacking). I still think is factor when looking at these creatures as there is some value there even if it is significantly less (probably below the threshold).
The one thing I really enjoyed about your post, above all over things, is that you worke hard to demonstrate a point in each section, WITHOUT bein conceding or smugish in your typing. Very good work, and I hope others will read.
Fantastically done. people really forget that Magic is a 2 player game. Interaction in essential in order for a strategy to be successful. You also don't mention land destruction, which is another form of interaction, punishing decks with greedy manabases as well as rewarding decks for being heavy basic. However, these basic manabases lack manlands and other utility lands which impacts the resilence of the deck.
Private Mod Note
():
Rollback Post to RevisionRollBack
EDH Commander WUBSharuum the HegemonBUW UGRRiku of the Two ReflectionsRGU
Amazing read, even though I consider myself a pretty good deckbuilder. The sideboarding section is INCREDIBLY helpful. I look forward to its completion.
posted section "Sometimes Even Good Decks Are Bad Choices". probably will write the last section tomorrow. link this article to your friends! i'm glad people are finding it helpful. help me get more readers for it.
Honestly, I feel as though this should receive a sticky from the Moderators. I've played in PTQs, but I'm by no means a "pro" Magic player. I don't really build decks. I put together what I feel is the best deck out there and then tweak it to whatever sort of meta I expect to face. I stay away from trying to innovate myself because I don't think I really have the experience or know-how to really build a competitive deck from the ground-up and make it work. Consequently, I feel as though I understand less about Magic than I otherwise might by exploring real, intense deck-building.
So, thank you for this post. Really valuable information for me.
The guide is fantastic! Definitely gonna show this to the new(and probably old!) players in my LGS. I certainly going to agree with the others here that this thread is sticky-worthy.
Private Mod Note
():
Rollback Post to RevisionRollBack
"For every complex problem there is an answer that is clear, simple, and wrong."
- H. L. Mencken
This is intended as a summary guide laying out fundamental principles of good deck building. This is not intended to be comprehensive nor is it intended to introduce cutting edge concepts, just summarize established wisdom. A player reading this ought to be able to learn the established best practices of deck building.
Who is this for?
This is for players with some magic experience who are looking to take their deck building skills to a higher level.
This will probably be a good read for true novices also but there is an assumption of familiarity with core game concepts (like Tempo and Card Advantage and Mana Curve) that a true beginner might not have. This might be a good read for experienced players also, but it will not teach you anything you don't already know. You may enjoy reading it anyway because sometimes seeing something written in somebody else's words makes it easier to think about.
The real target audience for this guide are players who are familiar with the the game but have not mastered deck building. My assumption is that a player who wants to build a better deck wants to build a deck that wins more and is more competitively viable. If that is not your goal then this guide is not for you.
Who am I?
I'm Metamorph. No I will not tell you my real name or DCI number. Thats for my real life friends only. I'm an experienced magic player with almost 17 years under my belt. When I started playing the core set was Revised Edition and the two expansions you could buy in stores were The Dark and Fallen Empires. The first box of cards I ever bought was Ice Age, and I bought it because I coveted the awesome new artwork on Icy Manipulator in that set.
I first started playing competitively when Mirage/Tempest was the Standard environment and I've played competitively off and on ever since. I've had some mixed success but I'm certainly not a pro. I've Top 8e'd PTQS twice, a $5K open event once, and back when I was a kid the Junior Super Series multiple times. I've top 16ed more than 20 PTQ/$5K level events though so I'm quite familiar with what competitive play is like and how to produce a winning record, though I'm not quite good enough to play at the highest level.
I play lots of different types of Magic and I play on a budget most of the time so I absolutely do understand casual play and budget issues. But I won't be talking about that here. Or anywhere really. I stick to pure strategy. Thats what this guide is.
A Note on Format
This is intended for Standard Constructed Magic. Format provides context and context is the framework of meaning. Most of what I will say here is not at all true of formats besides Standard. If you are interested in learning how to build decks for Legacy, Vintage, Commander or any other format then this post will be of only minimal usefulness. Finally, the orientation of this post is for Standard in its modern form (meaning effectively Standard since about Invasion block onwards). Standard from the bad old days was very different and should not be used as a context for this discussion.
Summary
1. Good Decks Play With Good Cards
2. Good Decks Have Good Plans
3. Good Decks Have Good Mana Bases
4. Good Decks Respect Their Opponents
5. Good Decks Have 75 Cards
6. Sometimes Even Good Decks Are Bad Choices
7. Sometimes Your 'Good Deck' Isn't
Good Decks Play With Good Cards
The concept is simple but the idea itself has many wrinkles to it. The power level of a deck is directly related to the power level of the individual cards that the deck is composed of. Synergy enhances this further, but at the most fundamental level the standalone power of each card in the deck is extremely important. While it is possible to build decks where the Synergy of card interactions allows for otherwise weak cards to become powerful this is an exceptional case and must be justified independently. The most common case is that independently powerful cards are the best choice for building a deck.
This raises the crucial question: how can I tell when a card is independently powerful?
Thats a deep question in its own right and there is no easy answer when it comes to card evaluation. However, there are some very useful and well established guideline principles we can use to get it right most of the time. Not always though. Sometimes things aren't as they seem for one reason or another, but if we make our usual assumptions and consider the card in a typical Standard environment.
Efficiency
Efficiency is a rough measure of how much bang you get for your buck. How much effect does a card provide given the resources you spend on it. Every card requires resource expenditure, there are no exceptions. Even if a card costs zero mana and zero alternative costs it still costs the card itself. This is something that is frequently overlooked by inexperienced deck builders. Always remember that at minimum a card costs the card itself and there are many cards whose effects simply don't justify that cost.
How do we measure efficiency? There's not standard method. Its very contextual. A card is efficient if it produces more then other similar cards in the metagame. Sometimes we get "strictly better" scenarios where it is obvious that Runeclaw Bear has a better body for its mana cost then Goblin Piker. But how do we know that neither one of them is efficient enough for constructed? No easy answer. Generally its because the 2/2 body itself is not particularly worth spending a whole card on. This is a very good example of why considering the card cost is so crucial. We are readily willing to play other 2/2 for 2 mana cards if they provide an additional effect. Grand Abolisher or Leonin Arbiter or Kalastria Highborn or Goblin Wardriver are all readily playable for the same converted mana cost. The additional effect on these creatures is arguably the more valuable part. Would you play that effect if it was just on an enchantment with no body? Probably not. So really its the value we're looking at. We want to have it all. We want the additional effect AND the body.
I realize its not exactly useful to tell people they shouldn't play Runeclaw Bears. Lets extrapolate from the example and try to form a general principle: when evaluating a card for efficiency consider how much value the card provides. If a card provides just a body, could you instead play something with an effect AND a body? If a card provides just an effect, could you instead play something with an effect AND a body? This value maximizing notion of efficiency is a good guideline. We prefer to play with cards that are almost like built in 2-for-1s. If you can get a body and an effect at the same time the card is almost doing as much work as 2 cards.
The other notion of efficiency thats important to consider is raw mana efficiency. This is not a hard concept to measure. Lightning Bolt is the king of mana efficiency for burn spells. Nothing is better. Creatures like Leatherback Baloth are the kings of mana efficiency for creature body size. Mana efficiency comparisons are straightforward. Just look at the numbers. There's not much more to say except to emphasize that often it is incorrect to favor mana efficiency over card efficiency. In standard it is frequently the case that card efficiency is the more useful measure. Simple Example: Squadron Hawks is a 2 mana 1/1 (very mana inefficient) but is a 4 for 1 in terms of cards. Its one of the best creatures in the format.
Consistency
The other measure of card power that is useful to consider is consistency. Consistency refers to how likely a card is to be able to perform as desired under normal play circumstances. A card like Phyrexian Obliterator has very high mana efficiency but its restrictive mana cost makes it extremely inconsistently castable except in almost mono-black decks. Mana consistency is always related to what your mana base is capable of, so keep this in mind when including cards. You must be able to cast them consistently or they will be useless.
Mana cost consistency is not the only form. Some cards have functional dependencies that raise consistency issues. Any aura or equipment or creature enhancing effect (like Honor of the Pure) is functionally dependent on there being one or more creatures in play on your side of the table. This is a pretty soft consistency requirement (you usually do have creatures) but it still must be considered. A common mistake is to underestimate even soft requirements. If you are playing an equipment, could you have played another creature instead? That would add more power to the board without being functionally dependent on another creature. Does the equipment provide a powerful enough effect to justify using it instead of another creature? It probably has to be about as powerful as a Sword of Feast and Famine to be worth it. Thats a high bar to clear. Most creature enhancing effects are not worth it for this reason. I don't want to get caught up on this one example. There are many strategies that utilize synergy to make creature enhancement truly worth it (Tempered Steel decks, for example) but you should be aware that these synergy based strategies suffer from consistency issues because of the nature of the strategy. Its sometimes worth it anyway though.
There are other forms of functional dependency too. Any card that relies on you having draw another card first has a functional dependency. Be very wary when playing cards like this. Only the softest functional dependencies are consistent enough to be worthwhile. Very hard dependencies require ENORMOUS support to achieve consistency. If you're going to be running a hard dependency be prepared to build a dedicated combo deck, otherwise its probably not worth it.
To clarify the terminology: a soft dependency would be between categories. Equipment + Creature, Ramp Cards + Bombs, etc. A hard dependency would be between specific cards such as Joraga Treespeaker + Myr Superion.
Niche Cases and Role Players
There are times when it is not possible to only use the most efficient and consistent cards in the format. Maybe your colors simply can't do something efficiently. Maybe the format card pool simply doesn't have any consistent options. There are too many exceptions to list so my purpose with this brief section is just to make it clear that there are good exceptions to the general guidelines. Sometimes you just need a certain job done and have to work with the tools you have, even though by historical standards they are inefficient or inconsistent. We do what we must. However, if you find that your deck is including a large number of inefficient or inconsistent role players (cards that serve a highly specific purpose) then you might want to seriously question whether or not its worth it. If your strategy requires too many inefficient/inconsistent cards its probably not good enough. It might have worked in past environments but is not well supported in the current card pool. It might just not be a good idea, period.
Good Decks Have Good Plans
Magic is a strategy game. Strategy is about planning as well as about execution. If your plan is bad then you are at a huge disadvantage. Make sure you have a good plan before you show up to the fight. The classic way of defining your plan is by strategic archetype. Most players have heard of these before, we throw the terms around all the time without having a clear idea about what they really mean and the permutations of them are: Aggro, Control, Combo.
Mission Statement
A Mission statement is one or two sentences that describe in the most concise terms how your deck plans to win a game of magic. Every deck should have its own mission statement that is based on its archetype. Here are the general purpose archetypal mission statements, and a few examples of deck specific versions of them.
General Aggro Archetype: "play too many threats too quickly for my opponent to deal with them all before his life total has been reduced to zero."
specific example: "Deal as much damage as possible with my 1 and 2 mana red creatures. Finish the game with a second wave of red creatures or direct damage spells."
General Control Archetype: "prevent my opponent from executing his game plan. Then, when my opponent cannot defeat me anymore play a threat that my opponent cannot answer and win with it."
specific example: "utilize countermagic and creature removal to deal with early game threats. use card drawing spells to make sure I have more answers than my opponent has threats. then, when my opponent is exhausted play a Grave Titan and kill him with it in 2 turns."
General Combo Archetype: "assemble a combination of two specific cards as quickly as possible. the two cards, in combination, will produce an effect so overwhelmingly powerful that my opponent will lose to it."
specific example: "utilize cheap blue cantrips and card drawing effects to get Deceiver Exarch and Splinter Twin into my hand. Then, cast these two spells in combination to produce infinite Exarch tokens with haste and attack for the win."
Mission statements are very important. If you cannot produce a succinct mission statement for your deck then you probably don't have a good enough plan. How can you expect to win a game of Magic if you don't even know how you're going to win it?
There's more to archetypes and strategic plans then just what I've covered here. It could be a whole article series discussing the various hybrid archetypes (aggro-control, mid-range, ramp, aggro-combo, combo-control, multi-combo, etc.) but for the sake of preventing this already sprawling article from getting out of control i'll leave that discussion for another day.
Have a Plan B
No plan survives first contact with the enemy fully intact. Sometimes things just don't work out according to plan. Maybe your opponent had all the right answers for your attackers. Maybe your deck's own internal consistency issues screwed you and you just didn't draw what you needed in time. A good deck should be able to win even when things go a little wrong. If things go alot wrong even the best deck will probably lose, but having a fallback plan to cover some hiccups in your main plan is a very good idea.
What constitutes a good Plan B? Totally specific to each deck. No general way to characterize this. What makes a good plan B? Its not vulnerable to the same things Plan A is. If your Plan A is to attack with creatures then your Plan B should not also be vulnerable to creature removal. If your Plan A is to assemble a combo then your Plan B should not be vulnerable to the same things that disrupt your Plan A combo. Plan B doesn't need to be complicated or independently viable. Simple things will suffice. Maybe you'll just use Burn spells to finish the job. Totally valid plan B. Maybe you'll just get 20 points in with Manlands. totally valid plan B. maybe you've got a second, weaker combo in your deck to supplement the first. Totally valid plan B. Doesn't really matter what it is or even that its particularly good. What does matter is that you have a backup plan that still works even when your primary plan has failed.
Good Decks Have Good Mana Bases
This section can be short and sweet. The topic itself is deep and interesting but fortunately can be summarized in a few simple words.
On Mana Curve and Land Count
Mana screw will kill you every time. So will color screw. You can't play spells that your mana base cannot support. The single most common error I see from any deckbuilder is playing insufficient land. It is absolutely imperative that you play enough land for your deck, based on what the deck's mana curve is.
In general we can think of there being three typical mana curves: low, medium, and high. In practice there's a huge amount of variation but at minimum we can categorize any mana curve into these ranges.
Example of a low mana curve:
CMC 1: 8-12
CMC 2: 8-12
CMC 3: 4-6
CMC 4: 2-4
CMC 5+: probably none, maybe 1-3
Example of a medium mana curve:
CMC 1: 0-4
CMC 2: 8-12
CMC 3: 6-8
CMC 4: 4-6
CMC 5 or 6: 4-6
CMC 7+: probably none, maybe 1 or 2
Example of a high mana curve:
CMC 1: 0-4
CMC 2: 6-10
CMC 3: 4-6
CMC 4: 4-6
CMC 5: 3-5
CMC 6 or 7: 3-5
CMC 8+: probably none
most aggro decks have low mana curves, most control decks have high mana curves, most ramp decks have high mana curves. alot of decks end up with medium mana curves just because its fairly consistent. some decks are called midrange for no reason other then they have a medium mana curve.
here's a very basic rule of thumb for assigning land count to your deck. 24 lands is the default number for medium mana curve decks. if your deck has a low mana curve you can justify playing fewer then 24 lands. 22 works for most aggro decks. a very slim one (with lots of 1 drops) can get away with 20 or 21. if your deck plays fewer then 20 lands it has to have a curve that is lower than low. lots of free spells, maybe a mana cheating mechanic. if your deck has a high mana curve you definitely need more then 24 lands. 26 is the most typical number for control decks. ramp decks often play as many as 27.
On Colored Mana Balance
additionally we must ensure that colored mana costs are covered. it is usually correct to play as many quality dual lands as exist in the format that can fit into your mana base. some of the best decks of all time were good in no small part due to their mana base being composed almost entirely of dual lands. UB Faeries (a previous standard boogeyman, one of the best standard decks of all time) played 16-20 dual lands in its mana base. the advantages of this are profound. many have tried to come up with a mathematical formula for computing color ratio but there's usually some flaw with whatever formula you might come up with. here's a potentially useful, though flawed formula that you can use if you take care to make manual adjustments afterwards.
For each color in your deck count up every colored mana symbol in a spell's mana cost. Take the total number and divide by 2, round up. That is the minimum number of lands capable of producing that color of mana that your deck will require to cast its colored spells consistently.
What's wrong with this formula? It doesn't take into account turn priority. Maybe you have only a few green mana spells in your deck but they are very time sensitive and must be cast on turn 1 or 2 to be maximally useful (Birds of Paradise, for example). In order to account for this you must increase your green mana producers above and beyond what is strictly needed based on color symbol count. What else is wrong with this formula? it doesn't make any statements about lands that enter the battlefield tapped. some decks can support many EtB tapped lands and can play many more non-basic lands than usual because of it. control decks often have this advantage. Some decks have a much greater need to play their cards on the curve (meaning a 2 drop on turn 2, a 3 drop on turn 3, etc.) and cannot support many EtB tapped lands and must use basic lands instead or limit themselves to duallands that have a mechanism of entering play untapped. The above formula can be used to get a good baseline but it must be adjusted manually to account for the effects mentioned.
Other Concerns
Many decks include non-land mana sources, cantrips, card drawing spells, or other effects that assist with mana development in one way or another. This topic can get very complicated and sophisticated and further discussion is beyond the scope of this article. Just be aware that the rabbit hole is deep on this one and there are many exceptions to the rules of thumb given above. Each exception can be justified with good reasons though. If you don't have a good reason, you shouldn't break the rule of thumb.
If you're ever in doubt and feeling tempted to cut a land, don't do it. If anything you should add one more land then you think you need. You always lose when you get mana screwed but sometimes you win anyway if you get mana flooded. More is better. Don't shave lands just to fit extra cool spells in. You'll regret it.
Good decks respect their opponents
Magic is a 2 player game. Your opponent is not a goldfish. A common mistake is to assume that your opponent is less capable than he really is. This is inviting disaster. It is better to assume your opponent is MORE capable than he really is and to build for the most hostile environment you can think of. If your deck remains functional even in the face of considerable hostility then it is probably a good deck. If your deck crumples to enemy resistance then it probably wasn't good to begin with.
There are two main principles we can use to think about how a deck fairs against its opponent: Interactivity and Resilience.
Interactivity
Interactivity is a measure of a deck's defensive strength. When we "interact" with our opponent, what we mean really is that we're trying to screw up his plans by countering his spells, discarding his hand, killing his creatures, and otherwise just being a pain in his butt.
Why should you interact with your opponent? Isn't it better to just focus on the kill? Sometimes it is. There are some aggro decks and some combo decks that are at their best when they try to just kill as fast as possible and let the opponent do whatever they want. How often is that the best plan? Rarely. Sometimes even just a small amount of interactivity is enough to secure a win. A single crucial removal spell or counterspell is the difference between winning and losing a huge amount of the time. Interactivity is so important that its possible to base an entire archetype (control) around it.
The reason we should almost always attempt to interact with our opponent is because of variance. We cannot guarantee that we always get our best draws that will definitely win faster than our opponent. Interacting with the opponent smooths out this variance. If we can't win as soon as possible we must instead make sure our opponent does not win first. All we have to do is win before he does, not win before some arbitrary turn has elapsed in game.
There's a pitfall here though. Some players get into the habit of playing "answers for answers". they see that their opponent has some card that is particularly effective against their deck (a hoser basically) and think they can overcome it by playing a spell that defeats the hoser. Example: Dismember as answer to Kor Firewalker. This is not optimal strategy. An answer for an answer only works if you're holding it first. Its very time sensitive. If you draw your answer for answer too late it was ineffective and the damage was already done. Further, there's the risk of it being redundant. If you draw your answer for answer but your opponent does not draw his hoser then your draw was dead. A better strategy is to reinforce your game plan with resilience.
Resilience
Resilience is the compliment of interactivity. It is a measure of how difficult your deck is to effectively interact with. An effective interaction is one where the answer was efficient (in terms of mana AND cards spent) and timely (played in a relevant time frame). Resilient threats are hard to effectively interact with. Your opponent's ability to answer them is limited.
A good case study in Resilience is the issue of creatures vs. removal spells. This is a very common theme on the forums here and many players make the mistake of playing creatures that have no resilience against the most effective forms of removal. The fact is that most removal costs little mana (just 1 or 2) or if it costs more mana it is very card efficient (a sweeper, for example). For a creature to be resilient against this effective removal one or more of the following must be true
1) the creature must be cheap and individually unimportant. it is just one threat among many and there is another one right behind it.
OR
2) the creature must provide value regardless of whether or not it is immediately removed.
OR
3) the creature must itself be unusually difficult to remove. the most common effective removal spells must be of limited (or nil) effectiveness against it.
if a creature fails to meet at least one of those conditions it is not resilient against removal and is probably not a good choice, since most standard formats are filled with effective removal spells. These resilience concerns are the reason why Titans outrank just about any other large creature you might play in Standard. they are also the reason why aggro decks based on lots of small creatures are far more effective then aggro decks based on a smaller number of large creatures. there are always exceptions of course, but in general thats what we observe.
This is just one case study, the case of creatures vs. removal. similar analysis could be done for any complimentary pair of threat + answer. the best threats are usually the most resilient ones. it is often more important to be resilient then it is to be efficient. as mentioned previously, you only have to win before your opponent does, not before some arbitrary turn has passed.
Good Decks Have 75 Cards
For competitive play, one cannot ignore the sideboard. The sideboard is relevant in more games than not, at minimum half all tournament games include sideboarding, and realistically its closer to 60% (since more matches are 3 games than 2). The maindeck has to be good but if your sideboard isn't good then you're not really ready for competitive match play.
Specific sideboard analysis is the most valuable kind and it requires specific decklists and metagames to do, which is beyond the scope of this article. I'm going to focus on general principles and pitfall avoidance in this section so you can get the right idea and go from there.
Here's the basic, fundamental idea of correct sideboard building: you cannot board a card IN without also boarding a card OUT, so you must build your sideboard with this in mind. Often it is more important what you board out than what you board in. You might not have a dedicated hoser for a matchup but you're incredibly likely to have at least a couple of maindeck cards that are subpar in the matchup. If the only thing you use your sideboard for is replacing subpar cards with something with better performance then you're using your sideboard effectively.
Pitfall #1 - using too many hosers
A common pitfall is to jam your sideboard full of hosers against what you think the best decks are but you don't really have a plan for how to use them. Hosers are just hate cards and sometimes they work, but often they aren't even good enough because they don't mesh with your plan. If you took out cards that were useful and synergistic with your strategy and replaced them with cards that do nothing in particular for you other then make your opponent's life miserable, then you are not optimally using your sideboard. The goal is to win the game, not to grief your opponent. When choosing cards for your sideboard keep in mind that its more important that the card makes sense with your strategy than it is that the card is harmful to a specific opponent. You can't guarantee what opponent you'll face, but if you have cards that are good in your deck in general you can always use your sideboard to tune up the deck between games and make it slightly better than it was game 1, based on what your opponent is playing.
There are exceptions of course. Sometimes a deck is very dominant in the metagame and the best bet is to just hose it. This is rare though. It usually requires a kinda degenerate deck with an unusual strategy, usually combo decks of some kind (graveyard combo decks like Dredge or Reanimator are notorious).
A pretty good way to build a well integrated sideboard is to sketch out multiple alternative maindeck configurations that you'd be happy with. Each configuration might differ by 4-6 card choices. Let the overflow live in the sideboard and swap between configurations between games, based on which configuration you think will be best. If you build a sideboard like this you'll avoid having narrow cards that don't mesh well with your deck. Every card in the sideboard will be independently useful so you'll always be able to board at least a few cards in. This is better expected value then sometimes boarding in a really nasty hoser, but often just having nothing to board in.
Pitfall #2 - fighting the last war
There's another pitfall thats really important to avoid. Always remember that you are boarding against your opponent's game 2 deck, not his game 1 deck. You never face his main deck with your sideboard. you always face his maindeck + sideboard with your maindeck + sideboard. This pitfall is best illustrated with an example. Here's one I see people make ALOT.
Example: You are playing UW Control vs. opponent playing RDW
In game 1 you managed to stabilize the board around turn 5 by resolving a Day of Judgment to sweep his creatures. You were at 6 life when this happened. Unsurprisingly your opponent won the game by burning you out with a Bolt and Incinerate he had been saving in his hand.
Here's the mistake: you're playing Leyline of Sanctity in your sideboard and you want to board it in for game 2 because you lost to direct damage game 1 and figure you would have won if only your opponent couldn't have finished you off with burn.
This is a mistake because you're fighting the last war. What did your opponent do with his sideboard? He boarded out his burn for haste creatures. From his perspective the game looked like this: I was beating my opponent down something fierce but the DoJ landed just in time. I lucked out on that one and had the burn, but thats kinda loose. A Hero of Oxid Ridge would have been perfect though. Incinerate out, Hero in. This is good strategy from the red player.
From his perspective the burn spells are mostly used to remove blockers so his men can connect for damage. They can also be useful to damage players but a persistent threat like a creature is even better, its worth more damage for the card and is far preferable against a control player who might be running countermagic or lifegain which would completely invalidate the burn. Furthermore, the control player has few blockers anyway so the burn just sits around in his hand. Why did he have 2 burn spells waiting? because there was nothing to use it on before anyway.
Your Leyline whiffs. Your opponent boarded out the thing the Leyline protects from. The Leyline doesn't do anything to help win the game and produces no value, its just a narrow hoser, and now you're caught short because your opponent went the other way on you. He boarded out the burn and brought in haste creatures. You know what you should have boarded in? Creatures that can block. He boarded out cards that can remove blockers and boarded in more cards that are stopped by blockers. Furthermore, creatures wouldn't just be narrow hosers, they'd be useful no matter what your opponent draws. What you should have boarded in was something like Kor Firewalker, or Timely Reinforcements, or Wall of Omens or Spellskite or whatever.
I hope the example illustrates the point. If you use independently useful cards in your sideboard, and keep in mind that you are facing your opponents game 2 deck (not his game 1 deck) then you will have an effective sideboard.
Sometimes Good Decks Are Bad Choices
Its important to recognize that all decks have limitations and no deck can do everything. The best you can hope for is to have a deck that is powerful and consistent. It can execute a game winning plan that is difficult for the opponent to disrupt, while interacting enough with the opponent to make sure your plan wins before your opponent's plan does. Thats all you have to do to win in magic. You don't have to handle every situation, you just have to execute your plan before your opponent executes his plan.
What if its still not good enough? What if your plan doesn't win even when you execute it successfully before your opponent executes his? Yes. this is entirely possible. You can seriously just get trumped outright by a deck that has a strategy that utterly defeats yours. In well balanced metagames it doesn't happen very often but there are some classic historical examples. Here's one:
You are playing a mono-green Elves deck and your opponent is playing RDW. Your elves deck is pretty good in its own right. Its fast, its consistent, you've steamrolled multiple opponents and tend to do particularly well against certain types of control decks that can't keep up with your rapid board development. However, you've got a bad matchup. I mean a truly terrible awful shamefully bad matchup against Red. You lose this matchup like 80% of the time. The red player burns up our elves and attacks with his guys. Your early board development is completely thwarted and your normally awesome Overruns are deadweight in your hand, you don't have enough mana to cast them (because your mana guys got burnt) and you don't have enough creatures on the board anyway (again, burnt). You're getting bashed for 4 or 5 points a turn and don't have anything big enough to block. What do you do?
You might think you should try to address this terrible matchup by changing your deck. The elves are too puny and too vulnerable to burn so clearly you need to add bigger guys to the deck to fight back with, right? No. Wrong. Your manabase can't support bigger guys because you've tuned your deck to run on just 20 lands (and about 12 mana elves). Your deck is designed to swarm with 1 and 2 mana creatures and then cast Overrun for the win. The whole plan is supposed to execute between turns 1 and 5. What are you gonna do with 4 or 5 mana creatures? Even if you draw enough land to cast them (you won't, but lets just assume) what are you gonna do with them on turn 5 anyway? Nothing. Its already over by then. Too little too late. And now that awesome matchup you had against control sucks too because Control can easily deal with the big mopey 4 mana creature. You managed to ruin your good matchup without fixing your bad matchup.
There are many similar examples. The moral of the story is that a deck can only do what its strategic design and mana base allow it to do. You can't do everything and shouldn't try. The most important thing is to maintain strategic focus and let your deck be good at what it does. You'll be rewarded when your opponents are not able to deal with your strategy. However, opponents can easily deal with a weak version of any strategy just by overpowering you with their superior efficiency, consistency, and strategic focus. Don't sacrifice these fundamentals to cover your weaknesses. Stay focused and you'll win when you ought to. The point of deck building is to maximize strategic advantages to produce game wins. Focus. Keep your advantage maximized. Focus.
If the metagame is incredibly hostile to the strategy of an otherwise good deck then the correct choice is usually to just play a different deck. There's no shame in that. Good decks are still good even when they aren't the right choice. If you want to win though you need to make the right choice. Choose the best good deck for the metagame to maximize your wins.
Sometimes Your Good Deck Isn't
This section will act as a kind of punctuation mark on this article. There's one more thing I have to say about deck building, and I think its the most valuable advice I can give. I hope you take this to heart: most ideas don't pan out. The failure rate of deck ideas is exceptionally high. I throw away close to 90% of my skeched out decklists. I throw out most of them before even sleeving them up for playtesting. Of the decks I do get around to actually playtesting I still throw out most of them. There are times when I throw out ALL of the deck ideas I have and just play an established deck because its the right choice.
How do you know a deck idea is a failure? Start with the theoretical analysis.
Go through the deck and ask yourself "are these cards good enough? are there any weak cards in here that shouldn't be? are too many of the cards weak?" thats a stopping point right there. if the deck is full of weak cards its probably not gonna work. Remember that efficiency is not the only measure of power. Consistency is probably an even more important measure. Many decks have incredibly powerful best case scenarios but are on average extremely weak because they are inconsistent. Keep consistency at the front of your mind when evaluating how powerful your decklist looks.
Go through the deck and ask yourself "is it obvious how this deck wins?" If not, scrap it. A random pile of cards (even good cards) with no plan is not a good deck. Remember that many good plans are pretty simple. But you have to be realistic. If your plan is unrealistic the deck is probably a failure.
Go through the deck and ask yourself "does the mana work?" If not, you just can't play it. Sorry. Mana base is the most important constraining factor on deck construction. If you can't cast your spells, nothing else even matters.
Go through the deck and ask yourself "will this deck be able to actually win 2 out of 3 games in a match?" Some decks might do very well in game 1 but not actually be viable for competitive play because they are SUPER vulnerable to commonly played sideboard cards. You get nowhere winning every game 1 and losing every game 2 and 3. If a deck has a distinctive game 2 vulnerability, is its game 1 advantage HUGE to make up for it? If its not the deck is probably a failure.
And finally, ask yourself "does this deck work in the expected metagame?" As mentioned in the previous section, even decks that are good on all the fundamentals are often bad choices because the metagame is filled with bad matchups for them.
Remember that there's no shame in abandoning an idea. Its a skill in itself to know when something isn't working out and when your time would be better spent doing something else. Most ideas are flawed and its important to go and look for the flaws. Don't get caught up in the initial enthusiasm when you think of something cool. Let it simmer. Sleep on it. Take a look at it again tomorrow with a more critical eye.
The information is very useful for people who don't understand some of the basic deck building ideas like consistency and making sure that you have the mana to make plays.
UR Storm WUBRG Sunnyside Up W Death+Taxes BR 7->0:60 Discard
EDH
GU Vig RWU Zedruu BR Olivia
Play Magic online free with GCCG!
Download it here -- (My Username is ShinkyP)
Trade Thread: Here!
Any time we go back and look at efficiency and consistency on why a card is good requires that look at the format which I'm glad you preface to your points in the first section. Basically look at what you'd expect in a typical Standard Environment as a starting point. Every time of card evaluation always comes back to this so this type of discerning of the card pool might be the greatest skill that comes with experience from deck builder's perspective.
GWU Knightfall Modern
UW Tempo Legacy
UGR Burning Wish Cobra Vintage
Retired standard (rotated)
Aggro
:symg::symu: Splinterfright Midrange Aggro
:symb::symg: Birthing pod
Kuldoltha Goblins
:symu::symb: control (jace 1.0)
:symb::symr: vampires
Modern
Modern retired (Got tired of using)
smallpox
discard
Sphere of saftey
:symu::symg: Ghostly Flicker
RDW
:symr::symg: Gruul Zoo
Goblin tribal aggro
Burn
:symw::symg: auramancer
Not to change the subject, but please tell me what you gain from playing Mono Blue Control in Standard at the moment as opposed to Blue/White or Blue/Black or even Blue/Red?
http://www.youtube.com/watch?v=SY8h2vp5Xis
This is actually a good point. Fauna Shaman is the standout in that list. Steel Overseer and lord effects in general require redundancy usually as well. Tempered Steel has it's namesake, Overseer, and Signal Pest.
Hero of Bladehold is sort of in the middle. As it requires other bodies and redundant pressure to be able to do anything other than die to the first removal spell. Some people would consider Hero only borderline playable.
Lotus Cobra is a creature that can provide zero value if removed immediately but represents and impending threat. I suppose though it has redundancy in ramp spells and other ramp creatures.
Fauna Shaman though works fine without redundancy and is very threatening. It's goodness seems to be directly proportional to the speed of the format, the power of creatures worth fetching, and the average CMC of removal spells that see play that can remove it being 2 or greater (or 1 and provide card advantage). It's a lot more like Jace or Birthing Pod in that it is a CA filter.
I think what's missing is an adaption of the 2nd rule.
It should read something like:
2) The creature must provide value for it's time in play that exceeds the tempo loss of it's removal.
Like if you untap with certain cards they are devastating. But then look at Consecrated Sphinx. It's sort of middle ground. Does nothing if it is removed immediately does something if it survives to your opponents main phase. It doesn't really fit into 1 or 3. Definitely it's value is it's ETB even if it doesn't have an ETB exactly. It continues to provide even greater value the longer it lives. Fauna Shaman is like this as well but it takes longer to see value. This is the key difference between a creature like this and Primordial Hydra or even Hero of Bladehold as a removal spell largely sets back the value even a turn or so later.. you are only gaining value off the opponents life total or if they attack into it and their creature dies (bladehold does leave some soldiers around assuming you are attacking). I still think is factor when looking at these creatures as there is some value there even if it is significantly less (probably below the threshold).
GWU Knightfall Modern
UW Tempo Legacy
UGR Burning Wish Cobra Vintage
[Primer] Any deck in Type 2 Constructed
Kudos
Join the Poetry Running Contest!
Youtube: http://www.youtube.com/user/PsiandXenos?feature=mhee
i just added the section on sideboarding, so i hope you'll go back and read the new stuff.
eta on the rest of it is near future, possibly tonight.
EDHCommanderWUBSharuum the HegemonBUW
UGRRiku of the Two ReflectionsRGU
Clan Limited
Sig by Rivenor
Decks (All Budget)
Modern
WSoul SistersW
RGShamansRG
So, thank you for this post. Really valuable information for me.
Standard:
:symu::symb::symw:Solar Flare:symw::symb::symu:
Commander:
:symb::symw:Ghost Council of Orzhova:symw::symb:
Linger on death's door and risk being invited in.
- H. L. Mencken
French Duel Commander
WBR Kaalia of the Vast WBR
RUG Maelstrom Wanderer RUG