In the early days, Magic had crazy, subpar mulligan rules: you could shuffle up and draw a new 7 if your hand had no lands or all lands, otherwise you were out of luck. A long time ago the Paris Mulligan rule was introduced and it seems to have stuck. If you don't like your hand, throw it back and draw 6. If you don't like that one, try again drawing 5, and so on.

This seems like a pretty good rule and I didn't question it, but recently I've been playing "Duels of the Planeswalkers" on XBox (incidentally, if anyone else has that and wants to challenge me, let me know!). The mulligan rule there is a little different. You can throw your hand back and draw a new 7 once for free. After that Paris rules are in force: every subsequent time you mulligan your hand size goes down by one.

It seems to me that this "lenient mulligan" rule is really good for the game. Magic is a game with a large element of randomness from the shuffle. Quite often players will draw a 7-card hand that's basically unplayable and have to throw it back, unless they're crazy gamblers with nothing left to lose. If your opponent keeps and you mulligan to 6, you're at a significant disadvantage already. If you have to mulligan to 5 or fewer, you'd better hope somebody up there likes you.

Does anyone know (or have an informed idea) why Magic retains a mulligan rule that allows unbalanced games to happen as often as they do? It just feels like the XBox mulligan would allow players to play their decks "as they were meant to work" more often, and have to battle against long odds much less often. What is the defence for the continued predominance in the Magic world of the Paris mulligan rule?

  • magic.wizards.com/en/articles/archive/news/… Commented Oct 30, 2016 at 16:00
  • Any answer that is accepted or upvoted should address the fact that permissive mulligan rules allow combo decks to win faster and more consistently. This means, especially in faster formats like vintage and legacy, that there will be more games won in 1-3 turns and fewer skill-intensive games where many decisions are made.
    – John
    Commented Oct 4, 2018 at 21:47
  • As a note, MtG no longer uses the Paris mulligan, it currently uses the Vancouver mulligan (which adds a scry at the end if you are at less than 7 cards) and is now testing the London mulligan, redrawing a full hand then sending cards to the bottom of the library based on the number of mulligans taken.
    – Andrew
    Commented Mar 1, 2019 at 16:16
  • 2
    And now the London mulligan has indeed replaced the Vancouver mulligan as the standard type of mulligan.
    – theosza
    Commented Jan 22, 2020 at 15:43

12 Answers 12


A few points:

  1. "Good for the game" is drastically different if you are playing competitively or casually. Competitive Magic should encourage good deck building and make it as fair as possible to each player. Paris mulligan vs. lenient paris mulligan isn't going to drastically change the fairness between the two players at the table and adds more complexity to an otherwise already complex game.

  2. Learning when to mulligan is actually a very useful skill in Magic. Making it easier to stomach doesn't necessarily make it a better mechanic. Designing (and testing) decks with 7, 6, 5 card hands is very important. Likewise when playing against particular decks. Some decks struggle against a 1 card advantage. Some decks shrug it off and keep pounding.

  3. Real life Magic mulligan problems can often be solved by shuffling properly. People can get lazy in their shuffles and tournament nerves can make it worse. Take the time to adequately shuffle your deck each time you draw a hand. This is especially true after a mulligan: Simply cut-shuffling a few times will result in you drawing the exact same cards you just threw back.

  4. When playing casually, I recommend players just draw back up to 7 after a mulligan. I would rather play against a deck's best chance than roll over someone with a 5 card start. Casual games should be fun more than fair; if the lenient paris mulligan is more fun for your play group then that is how you should play.

  • I agree with the spirit of your point 1, but I disagree in specific with Paris mulligan vs. lenient paris mulligan isn't going to drastically change the fairness insofar as lenient mulligan rewards decks that are explosive but inconsistent (for example, two-card monte, and to a lesser extent, dredge)
    – Daenyth
    Commented Apr 5, 2016 at 15:16
  • 4
    Strategy bias isn't the same thing as unfairness.
    – MrHen
    Commented Apr 5, 2016 at 15:21

I'm old and tired, and I haven't played Magic much in the last 15 years. With that as a caveat, my recollection is that the harsh penalty for declaring a mulligan was to prevent excessive "gaming" of declaring a mulligan. Part of the point of M:tG was to create a deck that was playable despite the fact that the order of the cards would be random. Being able to mulligan out of a "bad" hand lets players create a less balanced deck that needs a specific start hand to succeed.

In other words, playing decks "as they were meant to work" should take into account the fact that sometimes things won't come out the way you expect.

Related link of possible interest: http://www.wizards.com/magic/magazine/Article.aspx?x=mtgcom/daily/af3


The old "you can mulligan if you don't have lands in hand" rule was killed fairly early on, because players would build decks with no lands (Moxes and other artifacts for mana instead), and then legally mulligan over and over until they drew their perfect hand.

  • 3
    Under Wizards' official no-land/all-land mulligan rules, you only got one mulligan; you couldn't 'legally mulligan over and over'. I can't dig up a copy of the old floor rules ATM, but look at archive.wizards.com/Magic/magazine/article.aspx?x=mtgcom/daily/… for confirmation of this. Commented Jan 23, 2015 at 23:08
  • 1
    Interesting. I'm sure I've heard stories of repeat mulliganing, but the ye olde days are vague in memory. Good link! Commented Feb 3, 2015 at 21:21

Informed idea: The problem with a more "open" mulligan rule is striking a balance between "fair randomization" and "how a deck is supposed to work." It's fairly easy to construct a deck in which the correct 6-8 cards being in your starting hand gives you a near-guaranteed win in 1-3 turns. Preventing this is the reason decks have a minimum size, and individual cards have a quantity limit. But if players could freely reshuffle and re-draw, they would. Over and over, until they got that perfect hand which won the game. The concept of "how a deck is supposed to work" would morph into perfect, 7-card combos. To prevent this, and create more options in deck design, mulligans are very, very costly.


One advantage the xbox version has over RL is that it can shuffle the deck instantly, whereas in RL it take a bit of time to shuffle.

Other card games take alternate routes to make it work with less mulligans needed.

Universal Fighting System: To mulligan you remove your hand from the game and draw a new one. Only one Mull is allowed.

Magi-Nation: Depending on your starting Magi you search through your deck and pick out 2-3 cards and then draw the rest of your hand afterwards.

Commander (MTG Variant): Partial Mulligans. You set aside any number of cards and draw that many from your deck. You may do this again drawing one less card than you set aside, until you are happy with your hand or run out of cards. Then you shuffle all of your set aside cards into your deck. Note that Commander (or EDH) has a 100 card deck size, and no duplication rule so this might not work as well for regular magic.

Now, I'm not suggesting that time saving is the only reason for less lenient Mulligan rules, but I would posit that it is a factor.


One element that I haven't seen in any of the answers yet: a 'lenient' mulligan rule advantages some deck archetypes over others. Some decks are built on the concept of redundancy: for a mono-Red aggro or burn deck, for instance, there may be minor differences amongst its various one-mana creatures or its burn spells — maybe this spell does an extra point of damage but only hits players, maybe creature X has a modest downside but creature Y has one less point of toughness — but for the most part, they're indistinguishable; just (e.g.) 12 copies of 'guy who costs 1 mana and has 2 power', etc. For a deck like this, absent issues with drawing the right number of lands, one 7-card hand is going to look like any other.

On the other hand, a combo deck — let's say the Splinter Twin decks in Modern — is looking for a very specific set of cards, often to the point where almost all the rest of the deck will be some combination of card draw and selection, with just enough control elements to keep the deck alive while it assembles its combo. This sort of deck wants a very specific set of cards in its opening hand (it would, for the most part, much rather have its combo pieces than have to find them) and so it often has to mulligan aggressively to find them; getting a 'free' seven-card mulligan is worth much, much more to a deck like this than a red aggro deck, and so decks like this become much more viable in the presence of a free-mulligan rule.

This can actually be seen in action in many casual EDH/Commander environments, where people tend to be very lenient in their mulligan rules and so wacky combo decks are often the order of the day.

  • 1
    So, this does explain one consequence of a free mulligan rule, but you still need to link that back to why Magic chose to use paris mulligan and not free mulligan. Commented Jan 24, 2015 at 0:20
  • @doppelgreener I thought that was implicit in the above - one reason (and there are certainly several) is that it's a game-design decision made to balance the power level of different archetypes. Commented Jan 24, 2015 at 1:24
  • @doppelgreener Paris mulligan was selected first. As a general principle, I support not changing rules unless there is something to be fixed. As this answer points out, a more lenient mulligan rule doesn't fix a problem, it just changes some aspects of the metagame.
    – Hao Ye
    Commented Jan 24, 2015 at 3:35
  • Another way to look at this that might make it more obvious to some people: the more free the mulligans, the closer you'll get to the best possible hand with the deck. For some decks the best possible hand is incredibly good, way better than average, and so lenient mulligans favor those decks and make them "unreasonably" good.
    – Cascabel
    Commented Jan 24, 2015 at 4:15
  • 1
    @doppelgreener I second your concern. Stating how the metagame would be affected is not enough to answer the question. The questions was not why you would choose one over the other, but why Wizards chose one over the other. To answer that, you either need to be able to read minds or you need to read this article that does a great job explaining why Wizards settled on Paris mulligans. I am having trouble figuring out when that article was posted. It may have been after the question was asked.
    – Rainbolt
    Commented Jan 26, 2015 at 15:37

There is an excellent article up by one of the Magic developers that discusses the issues with various mulligan types, and why they upgraded to the new "Vancouver" Scry 1 mulligan.

A quick summary:

  • The "cheaper" it is to mulligan, the more likely you are to do it. On digital, this is instantaneous. In paper, properly shuffling takes a lot longer, as does deciding whether to keep the hand.
  • The more control you have over your mulligan (e.g. the way Hearthstone works, where you select cards to replace), the more likely it is that certain powerful decks that are currently kept in check by their reliability issues will simply become broken
  • They want to have the same mulligan rules for all the major 1v1 formats; stuff that's fine in Limited can be bonkers for constructed.

One interesting point in the article is that it agrees with you that it would make casual games a lot more fun. So if you want to casually agree to do it outside of ranked events like FNM, go for it! But be aware that it's not how cards are balanced, and so as you progress, certain decks may be more powerful than they should be.


In a well-designed deck, the possibility of a truly unplayable hand is small. The possibility becomes vanishingly small after a single mulligan.

Thus, there really should never be a need to mulligan more than once (unless you're fishing for your golden starting hand).

Being down a single card is NOT a "significant disadvantage" - it's relatively minor. The current mulligan rules are established in such a way to discourage fishing for gold while not preventing a relatively fair start for each player.

  • 2
    Actually, I'd argue this isn't true. For example, in a 40 card deck with 17 land (above average) there is about a 10% chance of drawing only one land or less. Assuming a single mulligan back to 7 cards you've got a 1% chance per game of pretty much losing at the start. In a 16 round tournament you'd play between 32 - 48 games. So basically half the players would be hosed with a single mulligan at least once. Without a mulligan 10% loses due to poor luck is unacceptable.
    – aslum
    Commented Mar 20, 2015 at 18:20

First - this question and the answers to it are out of date. The Paris mulligan was first replaced with the Vancouver mulligan (Paris, but anyone who mulligans to six or less gets a scry 1 before the first turn starts), which has since been replaced with the London mulligan (Draw 7 cards and put 1 on the bottom of the deck for each time you took a mulligan when you accept a hand).

The digital versions of Magic the Gathering and the formats that have a free mulligan in paper have a big difference from paper tournament play that no one has commented on, these are games and formats where a match is played "best-of-one" where events in other formats are played "best-of-three".

Commander is a multiplayer best-of-one format that allows a free mulligan. Two-Headed Giant is a best-of-one two-on-two format that allows a free mulligan. Originally, MtG Arena, like the Duels of the Planeswalkers games before it, was a best-of-one only game, this game does something very different* with the starting hands, but does not allow a free mulligan. The more common formats, Modern, Standard, Draft, Sealed, Vintage, Legacy are all meant to be played in a best-of-three match setting, and these do not allow for a free mulligan, one single bad hand in one of the three games doesn't have near as much impact as a bad hand in a best-of-one format.

*Magic the Gathering: Arena shuffles up 2 copies of each player's deck and draws 2 starting hands then shows only the deck which has the better balanced starting hand to the player - the other copy is discarded. This effectively is an automatic "free mulligan" where the game, not the player, decides.


In any card game, a person who draws a bad hand will always be at a disadvantage against someone who draws good cards. That's just the nature of the game. An exception might be someone who draws a "mixed" hand (a technically inferior hand with good chances of improving). Such a hand might be four to a flush or straight versus one or two pair in draw poker.

If you get a "redraw," you actually have an advantage, maybe not against the good-drawing player, but against a "random" player. Magic tries to compensate for this fact with the Paris Mulligan rule; the second draw costs you a card, the third, two cards, etc. Then it's up to you to determine whether your starting hand is so bad that it's worth paying the penalty for a redraw.


The best solution would be to increase starting hand size. The % of games that are decided by initial bad draws/mulligans is very high, especially when considering how much meticulous thought and planning deck building requires. This is why having an 'auto-lose' from bad luck is frustrating to people, because most of the effort is acquiring cards and building the deck, and then they get to the 'actual game' and someone's fate in a match or tournament can (and will, given enough games) be completely sealed by luck of the draw. Even the pros do not have that high of a win%(although still higher than everyone else by at least several %), what they do have is the patience and temperament to keep chugging along despite the bad beats that no one can avoid no matter how clever or skilled they at this card game.

There is only a very small number of competitive decks in standard at any given time, this is because deck options are extremely limited due to the incredibly high level of consistency needed. This also makes competitive magic very expensive, as the need for consistency leads to only several % of the most powerful, efficient cards worth playing, and it also makes mana bases extremely expensive as every 'free' card that gives you more consistency will lead to more wins.

In summation, Mulligan-ing well isn't just about hand analysis or deck building skill. Avoiding the pitfalls of bad starting hands is more about having the small clump of cards that will actually give consistent, powerful results given that a deck size is pretty large and any one non-basic land can only make up 6.7% of the cards in the deck(note how powerful tutors of almost all kinds are, just because they let you bypass randomness for that single instance).

  • 1
    -1: Increasing the hand size actually improves the chances of extremely strong situational cards, which can in fact reduce the number of competitive decks (because once you remove inconsistency, fast decks that work because they take advantage of their opponent's inconsistency go away, and slower decks become stronger). For example, with 10 cards, many combo decks have the reliability to break the format, so you end up with just those decks in the format.
    – deworde
    Commented Oct 12, 2015 at 13:28
  • 1
    I agree with deworde. With larger starting hand sizes, you will see combo decks winning more consistently and faster. The only thing Storm decks need to win is a critical mass of solid cards in hand. If you give them an extra card or two (and therefore eliminate the games where they take mulligans, so they get a secondary increase in hand size), you're going to see a lot of turn 2-3 combo kills that can't necessarily be stopped by a Duress. That makes it harder for any other deck to compete. New players with weak decks would get blown-out in extraordinary, almost traumatic fashion.
    – John
    Commented Oct 5, 2018 at 17:51

Truly the best solution to poor opening hands is creating a balanced "deck within a deck."

Every card that searches for other cards or draws a card(s) can be considered a subtraction from the deck count if the mana cost is low enough. The card itself, mathematically, is replaced with the card(s) it draws/searches for, thus creating the external deck. Now, the internal deck should consist of the cards you need to play, such as Mana sources (basic lands, for example) or versatile, useful cards.

Imagine, now, that you have a 60 card deck. Opening hand, you draw a fetch land. Play it and pass your turn. On your next turn, it will untap and allow you to search for a land of the appropriate color. That means the fetch has been replaced by a basic land, and a basic land has been removed from your deck, lowering your chances of drawing it. If you run 12 fetch lands (pretty high number, but this is the extreme example), 20% of your deck count has been shaved. This correlates to 1 fifth of your deck, leaving you with 48 cards left to amass. Now, 4 in 48 (1 in 12) chance of drawing any given card, as opposed to 4 in 60 (1 in 15) you would have had without the fetches, and that doesn't include all the other searchers you might fit in there. It is possible in Modern format to make a deck with 30 searchers/draw cards, 20 basic lands and 10 cards that produce various desired effects and be an efficient killer.

  • 3
    This answers "What is deck thinning?" The question was "Why does Magic: the Gathering use the Paris mulligan?" To stay on topic, your answer needs to find a way to relate deck thinning to the Paris mulligan. I can kind of see where you were going with it, but I think you stopped writing before you established your point.
    – Rainbolt
    Commented Jan 23, 2015 at 15:14
  • It was meant to be in response to what was said previously. People say that decks are 60 cards, and that large number can validate a more generous mulligan, but with the large number of cards that allow for low cost searching, at least in modern format, the mulligan would be over powered if given more generosity. I suppose if I were to add on to my answer, I would add the following: "And this is why the Paris Mulligan was chosen over the Free Mulligan." Commented Jan 25, 2015 at 21:38
  • 1
    You're overvaluing the importance of deck thinning. This is almost on-par with saying that the Hindenburg blew-up partially because of a butterfly beating its wings one town over. One fetchland would reduce the landiness of each of your draw steps by about 2%. More negligible in faster formats than slower ones, because fewer draw steps. A tiny benefit when you don't want to mana flood, and a tiny liability when you don't want to mana screw. Add the benefit and the liability together; the overall net benefit is microscopic. The Paris mulligan also predates the Onslaught fetches.
    – John
    Commented Oct 5, 2018 at 18:58

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .