Thank you David Schwimmer for explaining this to us (lol sorry I couldn't resist). Although as some people said, all you're talking about here is probability of card draw, not really how to build a deck, like using mana curves or anything. Generally when I build a deck I'm not hoping for one card to win it for me. If I need one or two cards to really pull the deck through for a win, I just consider that a fun deck, because it may also fail miserably- especially if my opponent has a response to whatever threat I have concocted. Also I think what you said about people recommending 21 lands is true, when I started playing, people told me 20 lands is enough. However this just isn't true, 24 is really what I hear a lot of better players recommending these days. I used to play 20 lands but me and all my friends mana-weave and just play casually with a quick shuffle. Since I started playing on Magic Duels, however, where the game will automatically give your deck a thorough shuffle; I've found for a deck with an average curve, 24 lands works much better. 20-22 is okay for a cheap, highly aggressive deck and 26 for a slower deck with many costly spells. Also my last tid bit is that I do encourage players to play test their new decks, preferably against a friend, but by yourself works too. When I finish making a new deck I play test it to see: how fast can it deal 20 damage unimpeded? How well do the cards synergize? Your formula is good to figure out ahead of time how well I can assemble a combo, but unless you play against an actual opponent to test your deck, you're still not 100% sure if your proposed combo is powerful enough to win. You may find out that it's fragile and easily countered and now you have to tweak your deck to protect the combo, or evolve your deck to be able to handle different situations.
when deck building with myself or teaching a new player how to build I tell them to start with the base deck of 20/20/20. 20 lands, 20 creatures, 20 spells (sorcery, instants, artifacts, enchantments) shuffle up and play it till the tires fall off, but keep in mind that you will have to find out what's working and what's not, and we can adjust accordingly.
+Dan Acton The widely accepted rule-of-the-thumb is to start with 24 land, 20 creatures, 16 spells and tweak from there; even the old deckbuilder's toolkit mentions this. 20 lands is the lower fringe for decks that top the curve at 3 mana. Personally, I usually start with 23 lands and at least 6 draws/selectors/accelerators for mana heavy decks and 21 lands for most aggro decks. (Making a deck outside the 20-24 land range needs a lot of experience that no new player can comprehend. Also, it is better to include an extra land in deckbuilding, as players usually underestimate the deck's mana needs.)
This actually helped a lot. I still need to learn how to get the synergy with the cards now and know a win condition, but at least the core building of the deck isn't broken. lol
The way to really screw with people is to make a deck that exploits the pace of the game. Green and red are obnoxious because of speed. But stalling and then wiping amd taking over the board is fun and leaves them with almost no answers
as for the number of lands greater out equal to 2, you need to also remember that this includes hands of 5+lands which in most decks is too many lands.
I noticed I was having a problem with a spellslinger deck I was making for standard and low and behold I was having either mana problems or problems getting specific cards for the combo. This helped out a bit which really helped. Thanks!
This is a great video, but it is a little simplistic. The deck archetype you showcase might function well with 21 lands (and I'm sure it does, I'm not doubting you), however a UW control deck will need closer to 26 (sometimes 28 depending on which blocks make up standard/if you play modern). The issue I have with the video is that is fails to accurately depict lands, and their various functions. Some lands essentially act as spells (halimar depths, kessig wolf run, Valakut molten pinnacle) while some only perform basic functions (basic lands). In an aggro deck, let's say RG just so we can use a land I mentioned, it might be better to run an extra land to compensate for running a kessig wolf run. In control, conversely, you might want to run several more lands because they function a hand-sculpting (desolate lighthouse) or win-conditions unto themselves (celestial colonnade). Overall I love the video, and it's perfect for new players.
your on point Desolator, it is a game of mathematics. I've made some decks exactly like this method and some on "feel". I think this way is perfect for competition (money on the line) events, but I also like playing a deck I've made without knowing the exact probable outcome for casual. keep making videos. cheers
I have been building decks in a similar fashion for 20+ years without ever knowing this formula. I think as a player you begin to understand these concepts of multiples of cards, ratio of card types, and mana needs intuitively with experience. I still enjoyed this video immensely. :)
This is what I'm doing in math class right now, the greater the sample size for your experimental probability, the closer it matches your theoretical probability.
thanks man!!! i always have fun ideas that should work but i also dont usually only run 1 of a card unless it isnt important .. i usually have trouble getting some of my cards and you pointed out that draw and scry may be just what i need... i never really use them which is ironic considering i built deck that relied on Scry to get what you need for my friend...
This video is perfect in so many ways its good for new players and old its engaging it has cats and its super helpful having a formula for deck building will save me a lot of time trialing my decks you should remake this video and post it again and do the same thing for arena like plug the numbers into the mtga algorithm
My advice for deck building is to make the deck in an online deck builder sight after you get the probabilities to determine land, creatures, and spells to get a visual.(Reason why is to let you build decks for free) Then I recommend making proxies of what you want to play with before getting all the cards you need. This allows you to test the deck and gives the person a sense of if they would enjoy playing the real deck before they spend money. Final advice is test for the main 60 and only worry about the sideboard after you are confident with main deck. Other than that some good general advice is look for cheaper card that can do what you need for the deck, both in mana cost and cash wise, and to trade for cards with other players to help build the deck. Just remember to use a well known mtg card selling website for prices when trading cards.
Hypergeometric Probability is honestly pretty easy to learn. It seems overwhelming when you look at the formula, but after a day or two of teaching yourself what everything means, it's pretty easy. Still time consuming, though. And you have no way of doing it without a calculator unless you're okay with doing factorials up to 60!. x! = x(x-1)(x-2)...(1). So 60 would be 60*59*58*57 and so on. Pretty crazy. Also, great video, helped me out a lot when I first started making decks!
The only thing that is different here is a play / draw difference. The opening 7 would be the same, but if you are on the play, you only have 7 draws by turn 8 instead of 8 draws by turn 8. Minor difference there, but it might be worth pointing out.
12:25 Unless you want to be on the draw (which most players deck doesn't want to be - or at least they want to prevent other decks from being first) you would only have ten cards at turn 4 not eleven since you don't draw on turn 1 when you go first.
I think it is worth noting that this method assumes that you have a perfectly random shuffle every time. if someone uses this method, but doesn't Shuffle properly, the method goes immediately out the window. So, before you decide to just up and do this, watch videos online about how to shuffle properly. even then, slight variations in your shuffle will throw the probabilities off.
It also assumes you're playing a deck where you have to draw specific cards. Not all decks require you to draw specific cards. All that said, I'm liking this video simply because you can become a much better game player by understanding probability.
You can just shuffle and draw 10 times and see what % of the hands are usable for the goal of your deck. If it is 7 or 8 out of 10, then you are solid. If you tell yourself you need this card on your first hand or the deck doesn't work, see how many times you get it in 10 mock starting hands
The problem with using the formula % for 1+ or 2+ of a card is that you don't want more than say 4 (maybe less) for example a hand with mostly land or mostly the creatures.. The probabilities overlap and what we really need is a specific ratio of certain creatures and lands.
That would actually make the calculation useless because your number of successes would skyrocket. It would give you an extremely high rate but it would also consider only lands a success
I have one reserve about this and you said it. It cannot calculate stuff like multiple series of probability, like when I dig my cards out with aggressive draw and scry effect. Because it's basically how combo deck works :D I play some stupid 3-card combo and the probability are not useful in this case. You just have to test again and again and just follow your instinct ! :D
This is a simple formula for deck building. I make 60 to 90 card decks. 1/3 is mana and if its multi colored then i make half of that mana 1 color and the other half the other color. Then i do 1/3 creatures and 1/3 enchantments artifact sorceries and instants. Unless it is a creature or spell based deck then I might do (for creature heavy deck) 40 creatures and 20 sorceries instants enchantments and artifacts. But if it is multi colored then same goes for creatures and other things (half one color half the other.) Hope you found this helpful if you wanted a simple formula.
learning hypergeometric distributions for stats brought me here, but I'm glad I did. I don't play MTG, but I can immediately see how I can apply this to gwent which I do play, thanks.
Surely you need to multiply the probability that you get your required creatures by the probability that you also get the necessary lands to cast them, since you need BOTH to avoid a mulligan.
The "very complicated" reason that the cumulative probabilities don't add up (10:40) is that you are rounding both down. They should and do add up to 100%.
Great video! I use hypergeometric probabilities to design all of my decks. The playtesting is supposed to be in order to get a feel for when cards should be played optimally and to get familiar with a deck. All your probabilities and theorycrafting should be locked in with hypergeometric probability first. One question though, because its something I've been debating on a lot. I run a RB agro/burn deck and I use 20 lands. Coming off the calculator that puts me at a 75% chance of getting 2 lands in the opening hand. Is it worth dropping a card for an extra land to get to 78%?
I would absolutely add another land to bolster your mana base. That 3% might not seem like much, but across a dozen games it adds up. It also increases the likelihood of hitting land drops in the early game. The drawback is, hitting a land turn 7 is just a dead draw.
I have a ton of papers covered in data sets for mutual exclusivity for tempo based on the even plus even equals even, odd plus odd equals even, even plus odd equals odd. matching it up to the draw values to the different possible tempo outputs. omfg etc.
I just began the game about a week ago, and I see now through this video that my deck is weak, because the cards I need just aren't going to be pulled often. The thing is, I can't fix it because I have such a small collection.
+Seth Brown I made the same mistake :D It's easy when you're new to the game to just focus on guarded single purpose decks or 20-20-20 decks (which are actually 20-19-21 but that sounds stupid). Guarded single purpose are basically my aircraft carrier deck. 100% flying creatures plus 3 spells to protect them and prevent large damages to your life total. You do one thing then put in about 12 cards that one thing works. 20-20-20 is like 20 creatures with heroic, 20 spells that target your own creatures, 20 lands. These tend to be pretty reliable.
So I wont know if my B/G eldrazi is good because even with only 6 cards that win me games I have 12 cards that draw and scry for me so the probabilities are complex
I don't play MTG but my kids got me into Pokemon TCG almost a year ago and I like it so much I wish I had started way back in the 90s. I have an embarrassing amount of cards after less than 12 months. But I'm still a noob and I'm still figuring out how to build a deck that works. I've had some hits and misses. I actually built a deck online that I beat the typical Volcanion EX standard deck with. But I was wondering if this formula can be used for building a PTCG deck? Or is there a different formula?
rojoshow13 pretty much the same, as long as the currency system (mana for magic) is similar, I don't know how pokemon works, also the base hand and base card card in the deck would make a diff...also the maximum amount of one certain card allowed in the deck. ...adjust those numbers and it should work...but a mathematical formula probably wouldn't be as effective as just building a basic version of your proposed deck and play play play whilst tweaking for what you need...also, I too wish I had started this back in the 90's ...wish I had those old 5 dollar beta packs!
What about legendary cards? how many of hose should i run? i found the perfect number to be two or three, but i havent ever won with my homebrew, so im beginning to wonder.
Call me crazy but unless I'm running ring colors in the deck the most land that I run is between 20 and 22 because of the balance that I put in my decks that works perfect for me
+NegativeZeroh I meant cards that do the same thing. Like 8 cards that draw a card or kill a creature or grant protection. Like put in 4 gods willing and 4 feat of resistance. Or do murderous cut and hero's downfall.
Mandalorian1001 - The problem is that feedback like "This wasn't helpful" isn't helpful as feedback. You didn't explain anything until after he insulted you. Also, you're quite simply wrong. Using math, he's showing new players how to build consistent decks. That's the whole point of the video. You can use this website and look and see if you can consistently pull the cards you need to win. While there are plenty of other things somebody needs to know to successfully build decks, this is undoubtedly helpful advice to gain a more consistent deck. Let's address a couple of your points. You said that you don't need a calculation to know what will and what will not regularly show up in a game. You're right, near the end, he gave an example of what your chances are based on how many cards you have. Using that as a basis, without having to do any math at all, you'll get an idea for the consistency of your deck. Of course, you're dead wrong if you think it's reasonable to assume that the average person will have the slightest clue how often something will and will not show up. They'll know the obvious, like if they put 1 of a card in, they're not going to get that card very often, and if they put 10 cards in, they'll probably get that card pretty often. Beyond that, it's a guessing game, and you'd have to play many games to actually figure out your consistency. You then said that the method of having multiple cards is common, and not revolutionary. This is true, but this video highlighted what having multiple cards actually means. There are plenty of people who would think that having 4 win condition cards in there deck would be plenty, but this math shows that that isn't really the case. If you build a deck with only 4 win condition cards, you're hamstringing your ability to win consistently. By bumping it up to 8, you've increased your odds significantly. This alone would be quite helpful to new players for sure, and even some older hands who had never bothered to do the math. Finally, you say that using this math is needlessly complicated. How so? He gave a link to a site that will do all the math for you, and even gave a decent basic example based on how many of a type of card you have. This is only complicated if you're below average intelligence. I don't wish to exclude those people from the game, but honestly those people would probably find this game needlessly complicated. Finally, you mentioned that it requires logic, skill and care to build a deck. Sure, without doubt. Of course, he can't teach any of those things, so why would you expect him to do so? If he tried to cover every single thing that goes into making a good deck, the video would be hours long. If this wasn't what a person was looking for, they can look elsewhere, and they'll likely find the advice they needed. Regardless, if they paid attention, there were some good points to be taken from this video. Keep in mind, just because certain things are obvious to you, and thus isn't helpful to you, doesn't mean that the advice isn't helpful.
@10:38 you state that the probabilities of drawing less than two cards and drawing two or more doesn't add up to 100% and that it shouldn't anyway, but it actually does. If it wouldn't I'd tell you your caclulator doesn't work...
I multiplied the first part and got 046,790,195,362,944, then Divided and got 121,153.17706618... Basically, the odds of getting the right materials to win would be X/121,153.17706618*100 for a Percentage
12:41 Your sample size has to be lower because you said you need at least 2 lands and one double striker. So out of 7 starting cards 3 spots are already taken. So the real sample size to get the buff spell by turn 4 is 8. Of course this is just one out of few possible scenarios, so the real probability of getting of your combo gets a bit more mathematical, but nothing too complicated though...simply tedious.
+Golden Indian That's a really good question. Basically one. I mean in a combo deck it's classified as a "glass cannon" if you have 1 win condition but if your win conditions is beat them in the face with creatures then that's acceptable. It really depends on the deck's overall strategy. If it's narrow and preventable, you need more than one. Like "I'm going to get out willbreaker and steal all their creatures" is a bad single win condition because what if he dies?
+scatlauncher yes and no, any deck can technically win by attacking with small creatures, but maybe it isn't the strategy the deck intends to focus on. In my eyes, a wincon is a card, or a bunch of cards, that are meant to support the strategy with which the deck seeks to win a match. In delver, it's delver of secrets. in Kiki pod, it used to be birthing pod, Kiki jikki and the creature with which Kiki becomes an infinite combo. In the case of the latter, several creatures can help you finish the game without necessarily having to do the combo, but they wouldn't be wincons.
+Pyr Gynt Phytoburst is not as good because sorcery speed. means it's predictable. if he swings with two creatures, and you can block one, then he buffs the other one for game at instant speed with titanic. if he has to cast phytoburst beforehand you can just decide to block that one instead. it's all about playing your spells as late as you can, but before they can respond.
I hate math, but this intrigues me... I understand the math behind it, and I understand deck building a little better because of this video, so good on you, but what kind of difference mathematically would cards like Collected Company or Commune with the God's (I know they do different things to the probability considering Commune with the Gods actually pitches cards into the graveyard) do to the probability of pulling the cards you need?
Using the advice he put forth, it would tell you how consistently you would pull (we'll use commune with the god's as the example) commune with the gods within the first 2 turns, or 3 if you started. I'm saying this assuming you have no way to gain more land more quickly. Then, you could actually add the five that it pulls into the sample size. Essentially, if you played it the first moment you could, assuming you can consistently get it by turn 2/3 you could actually treat turn 2/3 as turn 7/8. It would almost increase your chances of getting the card you needed by turn 2/3 by 2. Assuming you had an 80% chance of getting the card you needed by turn 2/3 that would be boosted to like 90%. I used the calculator, and with 8 of the cards you're looking for in a 60 card deck, without this card, you have a 75% of getting the card you want. By using this card, it jumps up to 90%. To clarify, when I put down 2/3 that means turn 2 if you went second, because you drew from the your first turn. The three means if you went first, because then you didn't draw on your turn. Oh, and you might be questioning, how is 80% to 90% doubling your chances? Well, your chances can't go above 100%, so it obviously doesn't go from 80% to 160%. What you actually do is look at the remainder percentage after the 80%. So, after the 80% you have 20% left over. Well, when you do this, it drops that in half to 10%. And halving the chances you don't get it, is effectively doubling your chances you do get it. Hopefully this made sense and it helped. :)
"I know it doesn't seem to add up to a hundred, but it's not supposed to, very complicated as to why" It does add up to 100, it HAS to add up to one hundred and its not complicated why. The probability of having either less than two or two or greater than two has to be one because there is no other choice ! The only reason it didn't seem to add up was from your roundings. Good video though ^^ Just as a math student that physicly hurt me ^^
Desolator, what is the best number of lands for a 40 card booster draft deck? I used your calculator and it seems to be 15 or 16 (83%-87%). What would be your recommendation?
I'd run this ratio --> 21/60 times x/40 X is lands 21/60 is 21 lands(the most common amount in an average deck) and 60 total cards. You get the rest I'm sure.
Joe S You're welcome! If you have the time it's always good to do a few mock games where you shuffle and draw 7, and see what you would be able to do from there
i can try to help you out the best i can, which would be to tell you it depends on the type of deck. In aggressive decks, at least 50% of your deck should be win conditions, while control decks require more board control in order to get away with their win condition which could just be something as simple as Aetherling. If you have a deck idea, try and find the win condition first, so that you may properly place other cards that help the win condition(s) shine. I hope this helped.
Remy Tetreault well currently I'm trying to work on an orzhov control deck in standard, it runs Thalia's, kambals, and authority of consuls. I'm thinking aetherflux reservoir, but not sure.
Think of something that will win the game if you do it, then figure out how to do it. A good X spell + a metric asston of mana A huge expensive creature + a slow game A bunch of cheap creatures + anthems Double strike/infect + power boosting instants
one of the reasons I like Squadron Hawk...once I cast it lets me go through my deck and find the other 3 and put them in my hand....also...at the risk of sounding like a total noob...why fetchlands? why sacrifice a mana that represents two colors and 1 life to find another of the same type or one color mana?
That way you don't have to draw more land than you need. Basically you get a higher chance of drawing a land in the beginning but because most of them are fetches you lower your chances as the game progresses. Trench-Gorger Leviathan and Harrow are also good at this, but don't work until later.
Beacause Fetchlands don't tap for mana, they are used to Thin out your deck and get more responses instead of lands, think of it like, 2 cards for 1 land
this equation becomes sooo much harder for graveyard decks... i like graveyard decks. every time you mill you would have to redo the math EXCLUDING the cards in the graveyard. cool video though
blue cheese cult i find it hard to get hyped over new sets and i just started playing less than a year ago. I'm not into putting down hundreds of dollars every few months just to have a deck rotate out and become illegal. Plus Wizards just takes old mechanics and puts them on semi new cards or copies of old ones and calls it a "new set". The only new mechanics they have suck. Take cohort for example. It sucks balls.
+blue cheese cult That's perfect for the casual player. There's nothing wrong with that if you don't want to win. I know there are amazing budget decks out there that can even counter net decks but for players like me its a different story. like to go to tournaments and win as well.
+blue cheese cult Magic is a buy to win game most of the time. you cant spend less than 100 dollars in standard and expect to win. in modern maybe yes but standard... no. and im pretty damn sure wizards is coming out with new fetch lands for SOI to replace the old ones that are rotating out. why do you think they came out with expedition lands?
great video changed the way i build my deck, but this is not how probability works. for example lets say i role a 6 side die and i get 6 it does not change the probability for me to role a 6 next time same in magic, lets say i have 80% to draw a copy of a card in my opening hand it does not mean that i will not draw it in every five games. still thank you so much for the video
+ytn999 I concur. But when you klick on the link, the site says "teach yourself statistics" so I would assume, that this video was made with those statistics in mind. Now I am not very good at maths (as well as in mtg, I assume^^) but it is my understanding, that probabilty and statistics kinda rule each other out. Like one of my books keeps telling me: Coincidence has no memory^^ Great video nonetheless and a good read in the comments!
Yes and no, remember that your population (your deck) decreases with each draw. You have less cards in your deck and hence a higher chance to draw any of the cards still in your deck. In this case, if you have 4 copies of a card and 3 copies of another one, the probability that you'll draw the one you have 4 copies of, increases more than the one you have 3 copies of.
Yeah, you're right. However, there still is no way of drawing a certain card "for sure". One can only increase his chances of drawing a certain card, but there is no "guaranteed draw" of a certain card, even if the statistics would "predict" that. It's kinda like roulette^^ But in MTG the player can actually do something in order to have a better chance of drawing a certain card. That's why Tutors and Fetchlands are so popular (as well as so expensive^^) That being said, I know, what you mean.
Yeah, you're right. However, there still is no way of drawing a certain card "for sure". One can only increase his chances of drawing a certain card, but there is no "guaranteed draw" of a certain card, even if the statistics would "predict" that. It's kinda like roulette^^ But in MTG the player can actually do something in order to have a better chance of drawing a certain card. That's why Tutors and Fetchlands are so popular (as well as so expensive^^) That being said, I know, what you mean.
I think the "one in five games" was being used as a way for some people to think about what those numbers mean. And maybe more so a way to see if your strat is workable, idk.
Hi Des, this is slightly miss leading. Your example with 16 creatures resulting in only have to mulligan 10% of the time, is not really true. You Need X > = 1 and X < 6, and moreover you need at least 2 lands. So your 90% success includes X = 6 and X = 7, ie a handful of creatures with no land, which does not work and you would have to mulligan.
this is just as relevant for edh as anything... it's just a bigger sample size so you'll need more cards of a certain type to have any kind of consistency...
That table cloth though. Looks like clouds, but looks like popcorn. At the same time.
+Jacob Good I know right!?
Im not high enough for this shit.
but, what if they are indeed popcorn clouds? mind = blown
+Dealio09 Dude lol. I still can't believe that this comment is No. 1 lol.
Lol
awesome use of cat pic to keep people engaged
A secret to not getting land screwed as much: run some extra cycle lands, I do, and they are cheap too. Got too many lands? Cycle them.
MiiTRI or play them so you can do more on those sweet sweet extra turns from expropriate.
Thank you David Schwimmer for explaining this to us (lol sorry I couldn't resist). Although as some people said, all you're talking about here is probability of card draw, not really how to build a deck, like using mana curves or anything.
Generally when I build a deck I'm not hoping for one card to win it for me. If I need one or two cards to really pull the deck through for a win, I just consider that a fun deck, because it may also fail miserably- especially if my opponent has a response to whatever threat I have concocted.
Also I think what you said about people recommending 21 lands is true, when I started playing, people told me 20 lands is enough. However this just isn't true, 24 is really what I hear a lot of better players recommending these days.
I used to play 20 lands but me and all my friends mana-weave and just play casually with a quick shuffle. Since I started playing on Magic Duels, however, where the game will automatically give your deck a thorough shuffle; I've found for a deck with an average curve, 24 lands works much better. 20-22 is okay for a cheap, highly aggressive deck and 26 for a slower deck with many costly spells.
Also my last tid bit is that I do encourage players to play test their new decks, preferably against a friend, but by yourself works too. When I finish making a new deck I play test it to see: how fast can it deal 20 damage unimpeded? How well do the cards synergize? Your formula is good to figure out ahead of time how well I can assemble a combo, but unless you play against an actual opponent to test your deck, you're still not 100% sure if your proposed combo is powerful enough to win. You may find out that it's fragile and easily countered and now you have to tweak your deck to protect the combo, or evolve your deck to be able to handle different situations.
dude this brings me nostalgia i watched this when i first started out brings great memories
"now you don't have to leave". Love that
"You'll go to turn 8 that's pretty much guaranteed". Red aggro: hold my beer.
Put em to the flame!
Awesome desolator!
Been playing magic since unlimited. But you've shown me quite a few new tricks. Keep up the good work.
Thank you so much. The chance for me to get the combo I needed for my Death Star deck was only 22%. Gonna go and improve it.
when deck building with myself or teaching a new player how to build I tell them to start with the base deck of 20/20/20. 20 lands, 20 creatures, 20 spells (sorcery, instants, artifacts, enchantments) shuffle up and play it till the tires fall off, but keep in mind that you will have to find out what's working and what's not, and we can adjust accordingly.
thanks I'm new and I'll use this to help me out.
+Dan Acton The widely accepted rule-of-the-thumb is to start with 24 land, 20 creatures, 16 spells and tweak from there; even the old deckbuilder's toolkit mentions this. 20 lands is the lower fringe for decks that top the curve at 3 mana. Personally, I usually start with 23 lands and at least 6 draws/selectors/accelerators for mana heavy decks and 21 lands for most aggro decks. (Making a deck outside the 20-24 land range needs a lot of experience that no new player can comprehend. Also, it is better to include an extra land in deckbuilding, as players usually underestimate the deck's mana needs.)
I usually do 20 lands 15 spells 25 creatures
This is the single most helpful deck building video I have ever seen.
Note that with this method you need to calculate both when you're on the draw and on the play. If you're on the play, sample size is reduced by 1
This actually helped a lot. I still need to learn how to get the synergy with the cards now and know a win condition, but at least the core building of the deck isn't broken. lol
The way to really screw with people is to make a deck that exploits the pace of the game. Green and red are obnoxious because of speed. But stalling and then wiping amd taking over the board is fun and leaves them with almost no answers
I love Collected Company, Search for Azcanta, draw spells, and tutors. You can tool box creatures or spells, and remain pretty safe.
as for the number of lands greater out equal to 2, you need to also remember that this includes hands of 5+lands which in most decks is too many lands.
I noticed I was having a problem with a spellslinger deck I was making for standard and low and behold I was having either mana problems or problems getting specific cards for the combo. This helped out a bit which really helped. Thanks!
This is a great video, but it is a little simplistic. The deck archetype you showcase might function well with 21 lands (and I'm sure it does, I'm not doubting you), however a UW control deck will need closer to 26 (sometimes 28 depending on which blocks make up standard/if you play modern). The issue I have with the video is that is fails to accurately depict lands, and their various functions. Some lands essentially act as spells (halimar depths, kessig wolf run, Valakut molten pinnacle) while some only perform basic functions (basic lands). In an aggro deck, let's say RG just so we can use a land I mentioned, it might be better to run an extra land to compensate for running a kessig wolf run. In control, conversely, you might want to run several more lands because they function a hand-sculpting (desolate lighthouse) or win-conditions unto themselves (celestial colonnade).
Overall I love the video, and it's perfect for new players.
your on point Desolator, it is a game of mathematics. I've made some decks exactly like this method and some on "feel". I think this way is perfect for competition (money on the line) events, but I also like playing a deck I've made without knowing the exact probable outcome for casual. keep making videos. cheers
I have been building decks in a similar fashion for 20+ years without ever knowing this formula. I think as a player you begin to understand these concepts of multiples of cards, ratio of card types, and mana needs intuitively with experience. I still enjoyed this video immensely. :)
This is what I'm doing in math class right now, the greater the sample size for your experimental probability, the closer it matches your theoretical probability.
Pretty awesome video, never considered magic from the mathematical side, will really use these tips!
Very nice guide. I think the deck I'm gonna really use that on is my reanimator deck. That really needs a lookup and boost in consistency.
thanks man!!! i always have fun ideas that should work but i also dont usually only run 1 of a card unless it isnt important .. i usually have trouble getting some of my cards and you pointed out that draw and scry may be just what i need... i never really use them which is ironic considering i built deck that relied on Scry to get what you need for my friend...
hazzah for the laws of probability! I knew math would come in handy some day ;-)
This video is perfect in so many ways its good for new players and old its engaging it has cats and its super helpful having a formula for deck building will save me a lot of time trialing my decks you should remake this video and post it again and do the same thing for arena like plug the numbers into the mtga algorithm
The calculations could even work with EDH, if you group cards with the same purpose.
My advice for deck building is to make the deck in an online deck builder sight after you get the probabilities to determine land, creatures, and spells to get a visual.(Reason why is to let you build decks for free) Then I recommend making proxies of what you want to play with before getting all the cards you need. This allows you to test the deck and gives the person a sense of if they would enjoy playing the real deck before they spend money. Final advice is test for the main 60 and only worry about the sideboard after you are confident with main deck. Other than that some good general advice is look for cheaper card that can do what you need for the deck, both in mana cost and cash wise, and to trade for cards with other players to help build the deck. Just remember to use a well known mtg card selling website for prices when trading cards.
Hypergeometric Probability is honestly pretty easy to learn. It seems overwhelming when you look at the formula, but after a day or two of teaching yourself what everything means, it's pretty easy. Still time consuming, though. And you have no way of doing it without a calculator unless you're okay with doing factorials up to 60!. x! = x(x-1)(x-2)...(1). So 60 would be 60*59*58*57 and so on. Pretty crazy.
Also, great video, helped me out a lot when I first started making decks!
The only thing that is different here is a play / draw difference. The opening 7 would be the same, but if you are on the play, you only have 7 draws by turn 8 instead of 8 draws by turn 8. Minor difference there, but it might be worth pointing out.
12:25 Unless you want to be on the draw (which most players deck doesn't want to be - or at least they want to prevent other decks from being first) you would only have ten cards at turn 4 not eleven since you don't draw on turn 1 when you go first.
I think it is worth noting that this method assumes that you have a perfectly random shuffle every time. if someone uses this method, but doesn't Shuffle properly, the method goes immediately out the window. So, before you decide to just up and do this, watch videos online about how to shuffle properly. even then, slight variations in your shuffle will throw the probabilities off.
It also assumes you're playing a deck where you have to draw specific cards. Not all decks require you to draw specific cards. All that said, I'm liking this video simply because you can become a much better game player by understanding probability.
how would you use this to calculate the probability of all 3 tron lands and maps/scrys?
I used to do a simplified version of this in my head when I was younger... but for some reason I forgot how and I can't do it anymore?
You can just shuffle and draw 10 times and see what % of the hands are usable for the goal of your deck.
If it is 7 or 8 out of 10, then you are solid.
If you tell yourself you need this card on your first hand or the deck doesn't work, see how many times you get it in 10 mock starting hands
I clicked on this video because the "A'int nobody got time fo' dat" lady was in the thumbnail.
Evil Gary who didnt?
The problem with using the formula % for 1+ or 2+ of a card is that you don't want more than say 4 (maybe less) for example a hand with mostly land or mostly the creatures.. The probabilities overlap and what we really need is a specific ratio of certain creatures and lands.
I play 24 lands minimum and keep only if 3 lands. I find that a good ratio
Same except I play 24 lands but keep if only 2 but that's bc I run an insane amount of low cost-but effective cards so it works :p
I usually add up the cost of my spells and divide by 2.5 Usually lands in between 21 and 24
my aggro deck has 20 and sometimes i sideboard 2 lands out
Wouldn't a success be having the creatures AND the lands to cast them? Without the land part the successes includes hands with 0 lands.
That would actually make the calculation useless because your number of successes would skyrocket. It would give you an extremely high rate but it would also consider only lands a success
I have one reserve about this and you said it. It cannot calculate stuff like multiple series of probability, like when I dig my cards out with aggressive draw and scry effect. Because it's basically how combo deck works :D
I play some stupid 3-card combo and the probability are not useful in this case. You just have to test again and again and just follow your instinct ! :D
There's a video with Patrick Chapin on a boat. It's a video I watch at least every few months. Perfect advice, very interesting.
This is a simple formula for deck building. I make 60 to 90 card decks. 1/3 is mana and if its multi colored then i make half of that mana 1 color and the other half the other color. Then i do 1/3 creatures and 1/3 enchantments artifact sorceries and instants. Unless it is a creature or spell based deck then I might do (for creature heavy deck) 40 creatures and 20 sorceries instants enchantments and artifacts. But if it is multi colored then same goes for creatures and other things (half one color half the other.) Hope you found this helpful if you wanted a simple formula.
@ 10:47
They do add up to 100%
Once I understand this, Im sure it is amazing advice, so thanks!
18:07 princess bride
finally, the math behind the probability, not what I was looking for, but very helpfull.
learning hypergeometric distributions for stats brought me here, but I'm glad I did. I don't play MTG, but I can immediately see how I can apply this to gwent which I do play, thanks.
Gwent and mtg are like baseball and softball.... Basically the same game
Surely you need to multiply the probability that you get your required creatures by the probability that you also get the necessary lands to cast them, since you need BOTH to avoid a mulligan.
The "very complicated" reason that the cumulative probabilities don't add up (10:40) is that you are rounding both down. They should and do add up to 100%.
Subscribed because of this video!
Thanks man, this is exactly what I was looking for
With my deck I only need to make a second pull 5% of the time, and that is my first deck :)
Does anyone know what the formula looks like in Excel?
Any tips for a thallid deck
How do you play with more than four of a card?? He says to play with eight?
One thing what about mana types
And the card types such as swamp creatures to your swamp mana
so why does the probability of having either 1 lands not equal 100%? Did I misunderstand something?
Great video! I use hypergeometric probabilities to design all of my decks. The playtesting is supposed to be in order to get a feel for when cards should be played optimally and to get familiar with a deck. All your probabilities and theorycrafting should be locked in with hypergeometric probability first.
One question though, because its something I've been debating on a lot. I run a RB agro/burn deck and I use 20 lands. Coming off the calculator that puts me at a 75% chance of getting 2 lands in the opening hand. Is it worth dropping a card for an extra land to get to 78%?
I would absolutely add another land to bolster your mana base. That 3% might not seem like much, but across a dozen games it adds up. It also increases the likelihood of hitting land drops in the early game. The drawback is, hitting a land turn 7 is just a dead draw.
I have a ton of papers covered in data sets for mutual exclusivity for tempo based on the even plus even equals even, odd plus odd equals even, even plus odd equals odd. matching it up to the draw values to the different possible tempo outputs. omfg etc.
Hello, do you have any advice for a splashed green and mostly white cat tribal deck that top decks by like turn 4?
I just began the game about a week ago, and I see now through this video that my deck is weak, because the cards I need just aren't going to be pulled often. The thing is, I can't fix it because I have such a small collection.
+Seth Brown I made the same mistake :D It's easy when you're new to the game to just focus on guarded single purpose decks or 20-20-20 decks (which are actually 20-19-21 but that sounds stupid). Guarded single purpose are basically my aircraft carrier deck. 100% flying creatures plus 3 spells to protect them and prevent large damages to your life total. You do one thing then put in about 12 cards that one thing works.
20-20-20 is like 20 creatures with heroic, 20 spells that target your own creatures, 20 lands. These tend to be pretty reliable.
Never rely on 4 copies of a card? I thought the rules stated a MAX of 4 of a card. Please elaborate.
To sum it all up- Get a bunch of cards that do the same thing so that the probability of pulling what you need is as high as it can be.
So I wont know if my B/G eldrazi is good because even with only 6 cards that win me games I have 12 cards that draw and scry for me so the probabilities are complex
I don't play MTG but my kids got me into Pokemon TCG almost a year ago and I like it so much I wish I had started way back in the 90s. I have an embarrassing amount of cards after less than 12 months. But I'm still a noob and I'm still figuring out how to build a deck that works. I've had some hits and misses. I actually built a deck online that I beat the typical Volcanion EX standard deck with. But I was wondering if this formula can be used for building a PTCG deck? Or is there a different formula?
rojoshow13 pretty much the same, as long as the currency system (mana for magic) is similar, I don't know how pokemon works, also the base hand and base card card in the deck would make a diff...also the maximum amount of one certain card allowed in the deck. ...adjust those numbers and it should work...but a mathematical formula probably wouldn't be as effective as just building a basic version of your proposed deck and play play play whilst tweaking for what you need...also, I too wish I had started this back in the 90's ...wish I had those old 5 dollar beta packs!
lol I'm taking ap stats right now and we literally just learned this formula
i allways find it odd when it comes to lands because i usually run 16-18 but almost never get land screwed
What about legendary cards? how many of hose should i run? i found the perfect number to be two or three, but i havent ever won with my homebrew, so im beginning to wonder.
Call me crazy but unless I'm running ring colors in the deck the most land that I run is between 20 and 22 because of the balance that I put in my decks that works perfect for me
And it's only 20 if it's one color
I appreciate this greatly. Thank you for sharing.
When you say don't put four cards of a card put 8.. I thought you can't put more than four of one card in a deck. Or am I wrong?
+NegativeZeroh I meant cards that do the same thing. Like 8 cards that draw a card or kill a creature or grant protection. Like put in 4 gods willing and 4 feat of resistance. Or do murderous cut and hero's downfall.
Mandolorian1001
Then you must be some new kind of stupid.
Mandalorian1001 - The problem is that feedback like "This wasn't helpful" isn't helpful as feedback. You didn't explain anything until after he insulted you. Also, you're quite simply wrong. Using math, he's showing new players how to build consistent decks. That's the whole point of the video. You can use this website and look and see if you can consistently pull the cards you need to win. While there are plenty of other things somebody needs to know to successfully build decks, this is undoubtedly helpful advice to gain a more consistent deck. Let's address a couple of your points.
You said that you don't need a calculation to know what will and what will not regularly show up in a game. You're right, near the end, he gave an example of what your chances are based on how many cards you have. Using that as a basis, without having to do any math at all, you'll get an idea for the consistency of your deck. Of course, you're dead wrong if you think it's reasonable to assume that the average person will have the slightest clue how often something will and will not show up. They'll know the obvious, like if they put 1 of a card in, they're not going to get that card very often, and if they put 10 cards in, they'll probably get that card pretty often. Beyond that, it's a guessing game, and you'd have to play many games to actually figure out your consistency.
You then said that the method of having multiple cards is common, and not revolutionary. This is true, but this video highlighted what having multiple cards actually means. There are plenty of people who would think that having 4 win condition cards in there deck would be plenty, but this math shows that that isn't really the case. If you build a deck with only 4 win condition cards, you're hamstringing your ability to win consistently. By bumping it up to 8, you've increased your odds significantly. This alone would be quite helpful to new players for sure, and even some older hands who had never bothered to do the math.
Finally, you say that using this math is needlessly complicated. How so? He gave a link to a site that will do all the math for you, and even gave a decent basic example based on how many of a type of card you have. This is only complicated if you're below average intelligence. I don't wish to exclude those people from the game, but honestly those people would probably find this game needlessly complicated.
Finally, you mentioned that it requires logic, skill and care to build a deck. Sure, without doubt. Of course, he can't teach any of those things, so why would you expect him to do so? If he tried to cover every single thing that goes into making a good deck, the video would be hours long. If this wasn't what a person was looking for, they can look elsewhere, and they'll likely find the advice they needed. Regardless, if they paid attention, there were some good points to be taken from this video. Keep in mind, just because certain things are obvious to you, and thus isn't helpful to you, doesn't mean that the advice isn't helpful.
1v1 him bro since your deck is effective lol
@10:38 you state that the probabilities of drawing less than two cards and drawing two or more doesn't add up to 100% and that it shouldn't anyway, but it actually does. If it wouldn't I'd tell you your caclulator doesn't work...
This was a very good video and i wasnt thinking about it this way percentages helps alot thx
10:40 you should have said 4/5, since .78 ~ .80. Then it would add up to one hundred easy.
Thanks for uploading! i learnd from it
Love this channel
((12 choose 1)* (48 choose 6))/(60 choose 7)
Looks good man. 0.38129
12*12271512/386206920
Yeah no-one's going to do that stuff by hand.
I multiplied the first part and got 046,790,195,362,944, then Divided and got 121,153.17706618... Basically, the odds of getting the right materials to win would be X/121,153.17706618*100 for a Percentage
I love these kind of videos, you should collaborate with Travis Woo
you should make a video explaining the math behind it
12:41 Your sample size has to be lower because you said you need at least 2 lands and one double striker. So out of 7 starting cards 3 spots are already taken. So the real sample size to get the buff spell by turn 4 is 8. Of course this is just one out of few possible scenarios, so the real probability of getting of your combo gets a bit more mathematical, but nothing too complicated though...simply tedious.
you sound like ross geller from friends
makes sense but how many win cons do i need for it to be solid
+Golden Indian That's a really good question. Basically one. I mean in a combo deck it's classified as a "glass cannon" if you have 1 win condition but if your win conditions is beat them in the face with creatures then that's acceptable. It really depends on the deck's overall strategy. If it's narrow and preventable, you need more than one. Like "I'm going to get out willbreaker and steal all their creatures" is a bad single win condition because what if he dies?
It's less about how many you have, and more about how reliably you can get one out.
My deck never really uses a win condition. I just build up my creatures and while I do that slowly do damage with flying
Then your creatures are your win con. Everything has a wincon.
+scatlauncher yes and no, any deck can technically win by attacking with small creatures, but maybe it isn't the strategy the deck intends to focus on. In my eyes, a wincon is a card, or a bunch of cards, that are meant to support the strategy with which the deck seeks to win a match. In delver, it's delver of secrets. in Kiki pod, it used to be birthing pod, Kiki jikki and the creature with which Kiki becomes an infinite combo. In the case of the latter, several creatures can help you finish the game without necessarily having to do the combo, but they wouldn't be wincons.
Why not phytoburst instead of titanic?
+Pyr Gynt Phytoburst is not as good because sorcery speed. means it's predictable. if he swings with two creatures, and you can block one, then he buffs the other one for game at instant speed with titanic. if he has to cast phytoburst beforehand you can just decide to block that one instead. it's all about playing your spells as late as you can, but before they can respond.
Whats a double striker card?
Blue Night It attacks twice, once before your opponent and once at the same time as your opponent
I hate math, but this intrigues me... I understand the math behind it, and I understand deck building a little better because of this video, so good on you, but what kind of difference mathematically would cards like Collected Company or Commune with the God's (I know they do different things to the probability considering Commune with the Gods actually pitches cards into the graveyard) do to the probability of pulling the cards you need?
Using the advice he put forth, it would tell you how consistently you would pull (we'll use commune with the god's as the example) commune with the gods within the first 2 turns, or 3 if you started. I'm saying this assuming you have no way to gain more land more quickly. Then, you could actually add the five that it pulls into the sample size. Essentially, if you played it the first moment you could, assuming you can consistently get it by turn 2/3 you could actually treat turn 2/3 as turn 7/8. It would almost increase your chances of getting the card you needed by turn 2/3 by 2. Assuming you had an 80% chance of getting the card you needed by turn 2/3 that would be boosted to like 90%. I used the calculator, and with 8 of the cards you're looking for in a 60 card deck, without this card, you have a 75% of getting the card you want. By using this card, it jumps up to 90%. To clarify, when I put down 2/3 that means turn 2 if you went second, because you drew from the your first turn. The three means if you went first, because then you didn't draw on your turn. Oh, and you might be questioning, how is 80% to 90% doubling your chances? Well, your chances can't go above 100%, so it obviously doesn't go from 80% to 160%. What you actually do is look at the remainder percentage after the 80%. So, after the 80% you have 20% left over. Well, when you do this, it drops that in half to 10%. And halving the chances you don't get it, is effectively doubling your chances you do get it. Hopefully this made sense and it helped. :)
why did he use titanic growth and not phytoburst?
Growth = instant, phyto = sorc
DesolatorMagic ok makes sense
"I know it doesn't seem to add up to a hundred, but it's not supposed to, very complicated as to why"
It does add up to 100, it HAS to add up to one hundred and its not complicated why. The probability of having either less than two or two or greater than two has to be one because there is no other choice ! The only reason it didn't seem to add up was from your roundings.
Good video though ^^ Just as a math student that physicly hurt me ^^
Desolator, what is the best number of lands for a 40 card booster draft deck? I used your calculator and it seems to be 15 or 16 (83%-87%). What would be your recommendation?
I'd run this ratio --> 21/60 times x/40
X is lands 21/60 is 21 lands(the most common amount in an average deck) and 60 total cards. You get the rest I'm sure.
+Captainnoobface Thanks. I'm going with 16 lands for a 40 card deck.
Joe S depends on the average cost, I've found for anything less then 4 13 or so
Anna Summers thank you!
Joe S You're welcome! If you have the time it's always
good to do a few mock games where you shuffle and draw 7, and see what you would be able to do from there
I love making decks, but I can never think of a win condition, any tips?
lol ok
i can try to help you out the best i can, which would be to tell you it depends on the type of deck. In aggressive decks, at least 50% of your deck should be win conditions, while control decks require more board control in order to get away with their win condition which could just be something as simple as Aetherling. If you have a deck idea, try and find the win condition first, so that you may properly place other cards that help the win condition(s) shine. I hope this helped.
Remy Tetreault well currently I'm trying to work on an orzhov control deck in standard, it runs Thalia's, kambals, and authority of consuls. I'm thinking aetherflux reservoir, but not sure.
TheUnprofessional Sanguin bond + Tainted Remedy. Phyrexian Altar + gravecrawler + any card that damages players when you sac.
Think of something that will win the game if you do it, then figure out how to do it.
A good X spell + a metric asston of mana
A huge expensive creature + a slow game
A bunch of cheap creatures + anthems
Double strike/infect + power boosting instants
great info man, i didnt even know this was a thing. thank you for the assistance! :D
OMG someone else knows how to do math thank god
this is what i need in my boggles deck :P
one of the reasons I like Squadron Hawk...once I cast it lets me go through my deck and find the other 3 and put them in my hand....also...at the risk of sounding like a total noob...why fetchlands? why sacrifice a mana that represents two colors and 1 life to find another of the same type or one color mana?
That way you don't have to draw more land than you need. Basically you get a higher chance of drawing a land in the beginning but because most of them are fetches you lower your chances as the game progresses. Trench-Gorger Leviathan and Harrow are also good at this, but don't work until later.
Although they're still pretty overrated. That's why Temporal Expanse and Evolving Wilds are great
Beacause Fetchlands don't tap for mana, they are used to Thin out your deck and get more responses instead of lands, think of it like, 2 cards for 1 land
Matheus Bravo oh ok, so it's less about easy land access and more about emphasize your play cards, so you can draw your creatures sooner?
math goes totally haywire when you add synergies instead of exactitude's
this equation becomes sooo much harder for graveyard decks... i like graveyard decks. every time you mill you would have to redo the math EXCLUDING the cards in the graveyard. cool video though
blue cheese cult i find it hard to get hyped over new sets and i just started playing less than a year ago. I'm not into putting down hundreds of dollars every few months just to have a deck rotate out and become illegal. Plus Wizards just takes old mechanics and puts them on semi new cards or copies of old ones and calls it a "new set". The only new mechanics they have suck. Take cohort for example. It sucks balls.
+blue cheese cult That's perfect for the casual player. There's nothing wrong with that if you don't want to win. I know there are amazing budget decks out there that can even counter net decks but for players like me its a different story. like to go to tournaments and win as well.
+blue cheese cult Magic is a buy to win game most of the time. you cant spend less than 100 dollars in standard and expect to win. in modern maybe yes but standard... no. and im pretty damn sure wizards is coming out with new fetch lands for SOI to replace the old ones that are rotating out. why do you think they came out with expedition lands?
+blue cheese cult So no... magic will only be as expensive as a players ambition. 💀
great video changed the way i build my deck, but this is not how probability works. for example lets say i role a 6 side die and i get 6 it does not change the probability for me to role a 6 next time same in magic, lets say i have 80% to draw a copy of a card in my opening hand it does not mean that i will not draw it in every five games. still thank you so much for the video
+ytn999 I concur. But when you klick on the link, the site says "teach yourself statistics" so I would assume, that this video was made with those statistics in mind.
Now I am not very good at maths (as well as in mtg, I assume^^) but it is my understanding, that probabilty and statistics kinda rule each other out.
Like one of my books keeps telling me:
Coincidence has no memory^^
Great video nonetheless and a good read in the comments!
Yes and no, remember that your population (your deck) decreases with each draw. You have less cards in your deck and hence a higher chance to draw any of the cards still in your deck. In this case, if you have 4 copies of a card and 3 copies of another one, the probability that you'll draw the one you have 4 copies of, increases more than the one you have 3 copies of.
Yeah, you're right. However, there still is no way of drawing a certain card "for sure". One can only increase his chances of drawing a certain card, but there is no "guaranteed draw" of a certain card, even if the statistics would "predict" that.
It's kinda like roulette^^ But in MTG the player can actually do something in order to have a better chance of drawing a certain card. That's why Tutors and Fetchlands are so popular (as well as so expensive^^)
That being said, I know, what you mean.
Yeah, you're right. However, there still is no way of drawing a certain card "for sure". One can only increase his chances of drawing a certain card, but there is no "guaranteed draw" of a certain card, even if the statistics would "predict" that.
It's kinda like roulette^^ But in MTG the player can actually do something in order to have a better chance of drawing a certain card. That's why Tutors and Fetchlands are so popular (as well as so expensive^^)
That being said, I know, what you mean.
I think the "one in five games" was being used as a way for some people to think about what those numbers mean. And maybe more so a way to see if your strat is workable, idk.
you never draw during you upkeep unless a card tells you to. your draw for turn is during your draw step
now thats my kind of math!
Hi Des, this is slightly miss leading. Your example with 16 creatures resulting in only have to mulligan 10% of the time, is not really true. You Need X > = 1 and X < 6, and moreover you need at least 2 lands. So your 90% success includes X = 6 and X = 7, ie a handful of creatures with no land, which does not work and you would have to mulligan.
Nice video, very informative.
I play EDH this is a bit useless for me :( other than for basic lands.
this is just as relevant for edh as anything... it's just a bigger sample size so you'll need more cards of a certain type to have any kind of consistency...
putting statistics to use!