Quote from SunBurstThis article is brilliant. Even though I've heard some of the numbers before for limited and the probabilities for curving out, I've never got it presented to me when it comes to cubing, especially not from a designer perspective. Things like these are so far from obvious.
Quote from Don DThanks for that! Great read!! Shocking numbers!!! My cube was already very "light" on mana demands (high number of colourless cards and a bias against colour demanding cards), but the fixing can definitely be increased. Cryptic Command was my only XXX spell left in my cube, and rotated already in and out a few times and will now definitely rotate out with my next update.
Quote from Reason...particularly aggro (because there's less untapped fixing for aggro). Perhaps there's a lesson to be learned here for all cubes that cards with multiple colored symbols are a significantly higher drawback than most people believe, doubly so when those cards must be played on curve to have effectiveness.
Quote from majikianPerhaps the only thing missing is Gold card analysis. e.g. what does my mana-base need to look like to reliably cast Putrid Leech on turn 2 etc.
Quote from majikian »Great article, thanks.
Quote from majikian »Perhaps the only thing missing is Gold card analysis. e.g. what does my mana-base need to look like to reliably cast Putrid Leech on turn 2 etc.
Quote from majikian »I think that change was probably the single best thing that's happened to my cube.
My cube's undergoing a bit of an overhaul at the moment but I think I'm probably going to end up with Strangleroot Geist as my only aggressive (non gold) double-color 2-drop (because it's still totally solid off-curve anyway). I'm going to take a long hard look at my double-color 3-drops too, which is something I've been intending to do for a while but this article has been a timely reminder of how important that can be.
Quote from calibretto »You should just be aware of them while you're drafting.
Or what about Cryptic Command? If your deck isn’t almost entirely mono-blue, your chances of playing it on turn 4 are close to nil.
You’ll need to build a 12/7 split to include something like Phyrexian Arena and Snapcaster Mage in the same deck.
Quote from Falcone1983 »I'll try to give some constructive criticism.
Quote from Falcone1983 »On one hand, I'm a bit surprised at your surprise. A lot of your math boils down to the following rule of thumb: "Try to avoid playing double-colored mana spells costing 4 or less in more than one color". I think a lot of Limited players, consciously or intuitively, follow such a rule of thumb. Of course, it's still very useful to see the math behind this rule.
Quote from Falcone1983 »On the other hand, I do think you're a bit too pessimistic, or rather, you're having unrealistic expectations of what a "functioning manabase" is, i.e. demanding that a spell can be cast "on curve" more than 75% of the time is a very high bar to set.
Quote from Falcone1983 »I see two (closely related) reasons for this:
(i) unless I'm misunderstanding your calculations, your percentages answer the question "Given a certain manabase and given I have card A in hand, what's the probability that I'll have the correct mana available on the turn number equal to A's CMC?" This question doesn't take into account the fact that if you fail to have the right mana on curve, you'll sometimes/often have other spells to cast (in your other color(s) or with easier mana requirements), mitigating the negative consequences of not casting A on curve. In other words, you don't necessarily lose the game or fall significantly behind in every case you fail to cast one specific card in your hand on curve;
Quote from Falcone1983 »(ii) many cards in Cube don't lose much efficiency when cast "off curve", or are even meant not to be cast on curve. You mention the best example, Balance, in your article, but the same goes for most spells, and even most creatures. For example, True-Name Nemesis is a relevant threat and/or defensive card at all stages of the game. Sure, your control deck can lose a game because you didn't have the 2nd blue mana to drop it on T3 against an aggro opponent, but often, you'll draw into the required mana later on and be fine (in the sense that your card will still have a relevant effect on the game).
Quote from Falcone1983 »That said, your analysis is absolutely very relevant for two specific sets of cards, cheap aggressive creatures and answers to cheap aggressive creatures. In an aggro deck, not curving out due to mana issues will often mean putting insufficient pressure on your opponent to win the game before his more powerful expensive cards take over. Conversely, not being able to cast your Wrath of God on T4 against a curved-out aggro deck will probably lose you the game on the spot. That is the nature of a match-up where the fundamental strategy of one deck is to cast all its spells as soon as possible and have the game end with the opponent having unused/unusable resources left over.
The same applies more or less to any other early threat that requires an immediate answer, for example a Tinker-ed robot or a T1/T2 Planeswalker via broken artifact mana.
Now, if you craft a Cube environment where the early turns are extremely important (because of aggro decks and/or early broken plays being common), then many of your games will end or be essentially decided on T4/T5, and the problem you identify will become more pronounced, so mana fixing will become more important. Conversely, if the "fundamental turn" of your Cube is increased by one, you get one more draw step to "fix" your mana, and the tables in the article show that this makes a significant difference.
Quote from Falcone1983 »I think that may be a useful additional point to take away from the analysis: the faster your Cube environment, the more mana fixing you need.
Quote from Falcone1983 »More in general, mana issues have always been a part of Magic, and many games are won and lost because of them. The question is: how big of a problem do you think this is and how far do you want to go to fix it? Overloading on mana fixing in the Cube itself may be a solution, but it leads to its own problems (one of which is "boring packs" with nothing but mana fixing). If you want to eliminate most mana problems, you could do something extreme like letting each player, after the draft, add up to four of any one dual land to his pool. That way, you can count on a reliable 2-color manabase without having to add a bunch of mana fixing to your Cube in the place of more exciting cards (which are why we play Cube, after all, right?).
Quote from Falcone1983 »In the first table, you say you need 18 mana sources to have a 77% chance of having 4 on T4; in the Phyrexian Obliterator example, you say you need 19 Swamps to have a >75% chance to cast it on T4. That can't both be correct.
Quote from Falcone1983 »Unnecessary hyperbole. Your own table shows that if you have 12 Blue sources (that's not "almost entirely mono-blue") you still have a better than 50% shot at casting it on T4 (that's not "close to nil").
Quote from Falcone1983 »Poorly chosen example, since you almost never play Snapcaster Mage on T2. Point still stands, obviously.
Quote from Falcone1983 »Hopefully you don't take this the wrong way; I think your article contains a lot of very insightful ideas on Cube design, drafting, and deck building, so it certainly is recommended reading!
Quote from dschumm »Thanks for writing this, I appreciated that you phrased it as the start of a conversation rather than reporting a conclusion.
Quote from dschumm »The main things it made me think about:
1. Number of Lands
In your own cube list you are try to push fixing lands above 10%. This is a fine way to handle it but I found my group started to get grumpy at about the 10% mark, and seeing packs with 4+ fixing in it is not something they liked. Part of this was how they evaluated cards vs. your group and another was outllook. My groups sees cube as super-drafting not randomized constructed. In all other draft environments we have played (including dragonmaze) there was an upper limit on fixing. And at that level and in my cube we still see the cut land strategies that take all fixing and crazy powerful 5 color stuff.
I also think that having a huge amount of fixing decreases the power level of green. On the flip side cubes without fixing make green the best color. This is one way I am going to calibrate my land inclusion.
Quote from dschumm »2. Drafting and Building
For cube we can get 45 playables every time often with 30+ being colorless or in our 2 colors. Compared to a normal limited environments, we have many picks to spare allowing us to take XX1 or XX2 cards in both colors and figuring out at that end what works. This finial decision process is what wins games in a fast high-level setting. Your article will have a big impact on me here.
Quote from dschumm »3. Type of Lands
I think that some lands will be raised in my estimation. You pointed out how challenging the splash can be while maintaining a decent 2 color manabase. So lands that can splash in more directions are increased in value. The fetches of course, but there are others that are regularly dismissed by the group at large. The mirage fetches, terramorphic expanse, and evolving wilds can play a big roll getting that double mana or the splash on turns 2-4. The vivid lands can do this as well. Unfortunately this leaves only the pain lands as a place to look for untapped lands. Another often disregarded land type is the triple lands. While not helping with the double cost they are a great option for those that want to support larger gold sections. Ancient Ziggarat is a bone I might throw my aggro decks to compensate for the lack of great fixing for aggro decks. Cavern of Souls is also one i have my eye on. With the rise of human tribal in my cube those cards seem like they might have a home. Reflecting pool also goes up in my estimation.
As a side note I think Gemstone mine is creeping closer to a pick 1 in my mind. I also want to hear if people have tried Archeological Dig, Tengo Ice Bridge, Mirrodin's Core, or Lotus Vale.
Quote from dschumm »4. Gold Section
The reaction I see most many managers have is to go after gold cards. The most important thing is to make your group happy. Many groups like the author's just want their decks to run smooth as silk with tons of flexibility. Golds cards really aren't for them. There are more than enough mono colored powerful cards to completely remove this section. Other groups love the power level and evocative nature of the gold cards. Many of them are egendary, or planeswalkers, one-of-a-kind effects and seem irreplaceable. I fall in the middle. The point of the author is something that has caused my to halve my gold section. But this article actually made me not want to cut anymore. Why? It looks like XX2 and YY2 cards are very hard on a deck evenly split in two colors. Comparatively XY2 is going to be very easy in that same deck. I may not want Koth and Thrun in the same deck, but I want Huntmaster in, no matter if I am R/G, R/g, or r/G. As others have said the math would be interesting to see in detail but I think it would hold throughout the curve. Interestingly wizards also views XY2 as more difficult than XX2 or X3 and is more comfortable increasing the power level. So while a certain gold card only goes in 10% of decks this article tells us we should not assume that all colored cards go in 20% of decks. More likely double cost cards should only be going in X/y decks which is 10%ish.
Quote from dschumm »5. Hybrids
I have started sliding these in my main colors which is very controversial, but I like to cheat a little on my gold section. What is interesting from this article is that something like Boros Reckoner is probably better for my boros deck(mana wise) than RR1 or WW1 cards because I will always be dropping it turn 3 (not assuming colorless lands). I have been and will continue to be on the lookout for more hybrids and ways to get them in cube.
Quote from dschumm »6. Other Fixers
The analysis assumes just the lands for fixing with half a point credited to signets and such. I think there is a lot more too it than that and for me will be the biggest way this article impacts my cube. Each color has ways of dealing with this that need to be managed in addition to number of fixing lands in the cube.
Green: The most obvious. Card like birds of paradise are staples but there is a significant resistance to Sylvan Caryatid which I think should be a 360 staple. And for those 540 and above Utopia Tree an the like should be looked at. The Cultivate cards are obviously strong in this area. Mana elves allow GG1 and GG2 cards to be much easier than cards in other colors. I have had great success with Avacyn's Pilgrim and Noble Hieracrh and will be trying to find room for Elves of Deep Shadow and Deathrite Shaman. The enemy pairs are notoriously weak for lands and I actively encourage people to try Elves of Deep shadow instead of a subpar G/B land.
Blue: Card drawing and filtering. Why am I playing the Clique with double mana cards of other colors? I look at a heck of a lot more cards than the generic table above. Having recently acquired some new high power cards I have trimmed back on cantrips. Also I have added cards to go with Master of Waves. This article reminded me that I was being greedy and need to correct this.
Black: Black gets the occasional ritual or swamp fetcher but it mostly needs card drawing to pull into more fixing. With the heaviest color requirements I will continue to try to find ways to reward heavy black decks for thier dedication. I will also take a second look at unusal fixing like Crypt Ghast. Where black really shines is tutors, although not sexy, how many of us Demonic Tutor for a dual on turn 2 or 3 to fix the rest of our plays?
Red: Ut-Oh. The fastest, most aggro color in cube really doesn't have anything that can help it. I really wish wizards would print some better looters. But I think I just need to be upping the mountains in my deck. What I want is a a R1 2/2 haste looter once when it comes into play.
White: Not much better than red but it gets Land Tax like cards on occasion. The biggest advantage white has is the depth in it's aggro choices, allowing for the removal of WW cards.
Quote from dschumm »7. Mono-Colored Decks
I really like drafting mono decks for the exact reason the author is here. I get to play all my cards! Now they are less sexy and very hard with a smaller table of drafters, I think putting rewards in for players who go that way is a good idea. Not only have they made a risky choice, but the data aboves says it could be a smart move many times. Additionally it has a side benefit of leaving more fixing in the pool for others. Not sure exactly what I mean by rewards, but I am already running Thassa and Porphorous. Other options like Crypt Ghast and Nythos, Shrine to Nix would excite my group.
Quote from dschumm »8. Land Destruction
It strikes me that if fixing is so tenuous for players than disrupting that is something I want to be doing. In a general sense this makes me want to include a couple cards I removed because I didn't think the 'Land Destruction.dec' was coming together. Also Fulminator Mage is now one of my highest priority acquisition targets. And as we see increased fixing is important does it make other cards better(Blood Moon, Molten Rain, Spreading Seas, Kiora's Follower, Tutors).
Quote from dschumm »9. XX and Aggro
The author posited that the forum is dismissive of fixing concerns but in the last year (I am new here) there has been a big house cleaning on XX cards, particularly in white. By removing those cards it has certainly helped aggro. Another thing besides cutting is making sure that I have enough redundancy. That way I can have opening hands with W,R,W1,R1,RR so that I can drop threats uninterrupted . Dropping the Ash Zelot on turn 3 or 4 is not ideal in the abstract but isn't a big deal if I have another quality 2 drop which came down in its place.
Quote from dschumm »10. Cuts
The author is going after gold, and that is certainly fine for his group I am going to look at my XX cards again but more importantly my XX1 and XX2. These cards really need to be better than X2 and X3 cards. They also need to be cards that are designed to be in the primary color combination than in the secondary or splash. Changes will probably be subtle but I need to realize that many/most of the time these cards will not come down on curve. I also need to reexamine the purpose of heavy requirement cards. Geralf's Messenger stays for me as my group likes it as more of a combo/reach card, cheating it, reanimating and all forms of goofy. But for groups that are on black aggro, you better see a lot of Mono-B otherwise it is gone. Cards with a back up plans are great. My hearld of torment didn't come down on 3, but I get a second black at turn 5, no problem.
Quote from dschumm »11. Colorless cards and Phrexian Mana
I just don't have the room for colorless lands that are not amazing. This math reinforces something I have felt for a while. If someone told me they didn't put the colorless manlands in their 360 I would understand. Scary right? These cards are either taking up spell slots or you have specific decks that can handle the colorless mana. For me they are frequently a good 18th land but I see many people playing 8/8/Mutavault decks and cringe.
Dear wizards, make more aggro artifact creatures. The flip side of the above argument is the increased value of cards that don't care what lands you have. Porcelain Legionnaire and Phyrexian Metamorph are great in aggro decks because they really fill holes well. I will be doing another sweep of unassuming completetly functional cheaper artifact creatures to include. Epocrocite is a card I can't recommend enough.
Quote from dschumm »In conclusion (if you have read this far). The math is an important thing to keep in mind as a cube designer and player. The demands from the author's group bring it into high clarity but it all reinforces that fundamental cube concepts "What kind of experiences do you want players to have?" and "What challenges should they face?"
Quote from MunchieeZExcellent article. Thanks for this. I've gotten some of my new cubers to prioritize drafting fixing so those cards are already extremely hard to come by. I'm looking forward to your results of testing Tendo Ice Bridge. I'm already considering if maybe I should add another cycle of guild lands to my cube. Eesh!
Quote from eidolon232 »As requested, I am not going to get into detail about all the wrong numbers in your results here.
Quote from eidolon232 »I highly doubt that somebody is designing (as in not just throwing together the cards that they have lying around/ copying a list) a cube without understanding the basic math for this game.
Quote from eidolon232 »How about your life total? ;-)
No seriously, are you saying that you drawing the right amount of lands changes the chance of winning more than the deck choice? For reference, in Constructed, the win percentage of a deck against its best matchup is about 70% (if you are only considering mainstream decks).
Quote from eidolon232 »Sure, but the mana fixing, be it from lands, direct mana fixing or indirect mana fixing in the form of card draw and library manipulation plays a big part in most cubes, allowing players to look at color intensive cards in a very different light.
Imo Inquisition of Kozilek and Soltari Priest are relatively bad examples for cards that you want to play immediately, since when it is important what you hit instead of if you hit, it is often correct to save up discard spells for later in the game (e.g. casting IoK on turn 3 otp against a blue deck to increase the chance of hitting a counter that is threatening to deny you from casting a 4 drop that they can't handle.) As an "unblockable" creature, the Soltari has a constant damage/turn output, while unevasive low drops often can't get in for damage anymore as the game progresses.
Quote from eidolon232 »You are only talking about casting a single card on curve. If you want to really curve out (as in casting a 1,2,3,4 drop in consecutive turns) adding more than 17 lands is actually worse. Anyway, can you explain what the table is supposed to display, since the numbers don't really match what I have in mind.
Quote from eidolon232 »5+ mana cards (and to a lesser extend the 4s as well) are not designed to be reliably cast on curve, since the effective mana cost of cards doesn't increase linear. This is why the 4 mana cards are so overrepresented in cubes: they can be really devastating if cast an curve, because people not being able to do so all the time has been considered when they were designed, but you still get the opportunity to do so a high percentage of the time - especially with all the acceleration available. In the most basic model, you simply assume that you run out of lands on turn 4 and everything else has to be cast with the mana sources that you naturally draw, meaning that a cmc 5 mana card effectively costs 6 mana, a 6 costs 8 mana, and so on.
Quote from eidolon232 »When you are including nonland mana sources to ramp out some expensive cards / have a mana advantage over your opponent, it is only logical that they don't compete with your regular mana sources.
Quote from eidolon232 »I don't think that you should only present the data for being on the play, since decks are getting adjusted preboard.
Quote from eidolon232 »It would surprise me if anybody would look at Geralf's Messenger as anything other than a 5 drop (outside of mono).
The color requirements of a cards are less important the turn you can cast it according to their cmc, but more when you actually reach that much mana: A Vampire Nighthawk gets worse in relation to other 3 drops in a deck with a bunch of mana elves, while you have more time than 6 turns to find 3 black for Massacre Wurm.
Quote from eidolon232 »I would focus more on the percentage of the time where the color requirements (and you have enough mana to cast a card with the same cmc) are an issue when you have the card in your hand.
Quote from TheGebI have been playing Magic since beta and I have never thought about deck construction in quite this way. It really makes you analyze what you are pulling from the pack. That being said, my group hardly ever plays more than a two color deck so I like most of the cards in my cube despite the hard realities. Maybe my group, which is totally casual, is more for the greedy splashy spells that the on curve players.
It does make me happy that I made the right choice with including the vivid lands. They really shored up my mana fixing since foil fetch lands are still too expensive for their inclusion -- c'mon reprint!
-- Geb --
Quote from SalmoI like the article a lot. These are good numbers, for sure. There have been cases where I've skirted past mana restrictions because engines and archetypes within the strategy allow me to try and tackle both (say, a dedicated reanimator deck that can also cast Phyrexian Obliterator, or one that has multiple ways to pitch your off-color creatures and only cast them in the rarest of cases) but it certainly is important to keep mana restrictions in mind for the fairer decks and the ones that only have the cheating aspects as a feature and not a main component.
This article also makes me a little sick since I realize how lucky some opponents--and myself--have gotten with mana bases in the past. Sonnuvaguns "splashing" cryptic command grumble grumble...
Quote from rantipoleLoved this article. It made me really think about how I draft and construct decks. I sent a link out to my whole cube group.
Quote from trogdoor »One thing I kept coming back to was how this would be affected by mulligans. 1-, 6- and 7- landers will always get sent back into mulligans, and a lot of other hands too. Did you think about this?
Quote from trogdoor »...what composition of cards will make our deck the best...