With the release of Dominaria and Magic's 25th anniversary, the call of nostalgia is being heard. Maybe not by all players, but most definitely by those who were there when we first set foot on Urza's plane in the early 90's. While there's a lot that could be said about the difference between the game today and back then, one particular set of rules has only changed twice in the last 25 years: Mulligans.

Mulligans have been part of the game since Alpha and were included in the original rule book, but not quite like the one you know today.

It read:

"No land Mulligan": At the beginning of the game, if your starting hand doesn't contain any land card, you may reveal it to your opponent, shuffle your cards in your library and draw another hand of seven cards."

"All land Mulligan": At the beginning of the game, if your starting hand contains seven lands, you may reveal it your opponent, shuffle your cards in your library and draw another hand of seven cards."

You may only use a Mulligan once a game.

Before I learned that Mulligan was a term used in golf, I always thought it was someone's name, like "Callaghan." You would show your hand to your opponent and say "No land, Mulligan!" like you would say "No Luck, bro!"

With these rules, some cards were extremely risky to play: Maze of Ith, Kjeldoran Outpost or Lake of the Dead; basically any land that wouldn't produce mana on its own. If you had one in your hand and no mana-producing land, you were forced to keep…and you were screwed.

One of the common practices among casual players (at least where I'm from, and I'm sure it was the same all over the world), before the DCI enforced tournament rules, was to "weave" their decks, basically making two piles – one of lands, one of spells – and putting them back together two spells, one land, two spells, one land… ensuring you would draw a decent amount of lands every game and to keep deck building simple (40 spells plus 20 lands). This is of course illegal, but not everyone played by the rules and just wanted to have a good game that involved both players playing spells. Where I was, the practice was so common that asking your opponent if you could shuffle their deck before the game was considered an insult. Cutting once was the norm.

The guy pile-shuffling you in three piles had a good chance of having you draw no lands or all lands. In our circles, that guy was blacklisted. There was also the Millstone guy, who would wait till he milled two active cards and let you draw a land every turn. Good times! Anyway, these seem like practices from another time, but I'm sure in communities of casual players or even at your local prerelease you can find the two-spells-one-land shufflers, or the lands on one side, spells on the other riffle-shufflers.

All that to say that I don't remember the first mulligan rule to be so harsh, since we never really randomized our decks enough and often ended with playable hands (or at least when we started playing).

Once we started to play competitively, and had to play by the rules (as in not weaving and shuffling our opponents' decks), we added a couple of lands decks as 20 lands wouldn't cut it every time to cast Serra Angels. That time also came with a part of frustration that we had to live with: Magic had a terrible flaw. Or was it really a flaw? At times, you never really got to play the game. You would open a one-lander and wouldn't cast a spell.

The necessity for a fairer Mulligan rule became stronger as the game grew and it was in April 1997 at Pro Tour Paris (my first Pro Tour) that the mulligan rule changed for the first time. If you ever heard "I'm going to Paris" from a player taking a mulligan, they were referring to the "Paris Mulligan."

You would no longer need to show you hand to your opponent to draw another opening hand, you just had to shuffle your hand into your library to draw another one with one less card. The rule was implemented at the Pro Tour, but it took a couple of more months or even a year to be implemented as the rule in all tournaments. Some strategies took advantage of the all-land, no-land mulligans to win tournaments. My deck of choice back then, as many of you should know by now, was Legion Land Loss, a mono-green land destruction deck that was super stable and would punish any land-light draws.

The Paris Mulligan was well-received after its testing period at the Pro Tour and was universally adopted soon after, a rule that would survive for 17 years before a new rule was tested at Pro Tour Vancouver. A player would now get to Scry once after taking a mulligan, increasing the chances of having a game where both players actually play spells.

Magic was designed with a unique – at the time – system of resources. Cards would be used as spells, others as lands (mana, therefore resources) and you would build your deck trying to find the right balance. Of course, the risk of any player not drawing a land is ever-present and that is not fun for any party involved. Good players want more skill involved, and less randomness; they never want to lose to worse players. But Magic wants to give a chance to worse players to beat better players. And that's what makes the game so awesome. There's been a debate for years that Mark Rosewater addressed a million times. Without getting into details, he was saying that yes, "mana screws" are good for Magic for the reason mentioned above.

Mulligans address the problem of drawing an unplayable hand without completely breaking the game. If Magic didn't have mana screws, for example, if you always got to start with two or three lands in your opening hand, it would leave too much room for deckbuilders to abuse the rules (playing 57 spells and three lands for example). A lot of other card games, inspired by Magic, used other resource systems that prevented you from not having resources. VS System for example allowed you to turn any of your cards into "a land" (or something similar). The result was that the game was easily breakable, with decks completely unbeatable due to their stability.

Overall, the evolution of mulligans went in the right direction. Magic still punishes players with greedy mana bases while giving a chance to players with bad opening hands.

One of the main concerns with the Paris Mulligan was that combo decks would benefit too much from it, allowing them to mulligan into their combo pieces, almost for free. This didn't prove to be true as it didn't advantage them that much (if at all) over time. Sending back a hand light on disruption against combo decks proved to be just as good.

Ironically, I could see a lot of combo decks today that would benefit a lot from the old mulligan. Off the top of my head, Mana-less Dredge in Legacy or even Cheerios in Modern. Running less than 15 lands gives you a good chance to not have one in your opener. In Mana-less Dredge, you would have a free mulligan every game if you don't like your first hand.

It is actually pretty likely that if the mulligan rule had never changed, we would see more a lot more aggro and combo decks played. While I don't have the numbers, I can say with a good level of certainty that decks with more than 20 lands would almost never have the chance to mulligan over the course of a tournament (how often do you draw zero or seven lands in 14 rounds?), while decks with a lower land count (20 or less) would draw a no-lander once in a while and wouldn't mind drawing only one. With a low curve or a deck that could play with only one or two lands, it would be worth the risk to even go down to 18 lands to score the free mulligan once in a while.

All in all, the original mulligan rule would have pushed aggro and combo strategies and would have given almost no chance to any slow decks opening a one-lander. It basically wouldn't have helped Magic to become the diverse game it is today.

Can We Improve on the Vancouver Mulligan?

As far as fairness is concerned, I don't see any archetype that benefits more than another with the Vancouver Mulligan. There are decks that suffer a lot from mulliganing, but it's mostly in their design and it's usually decks that can't afford to lose cards from their starting hand (like combo decks), but they are supposed to be hard to put together, so I guess It's also just fair.

As far as fun is concerned, I believe we can do better. I'm all for punishing greedy mana bases and giving a chance to the worse of the two players. However, I hate to see a game where a player plays no spells. As a player, I don't like to not draw my third land in Limited (and I feel a little bad when my opponent doesn't either). As a commentator in the booth of the GP coverage, it's always a little awkward to have to explain that a player is losing because they can't play a spell and that there's nothing they could do.

As the game is right now, it feels like mulliganing still costs too much: starting with one less card, taking the risk of not drawing lands in the next hand… I'm on the team that would rather lose than mulligan, I just believe that people mulligan too often in general and underestimate the actual risk.

I loved the introduction of the scry after the mulligan, especially in Limited. It helped to address one of the factors you have no control over – the place of the third land on the top of your deck. Scrying a spell to the bottom helps get you closer to the land you need. It doesn't impact the way you build your deck and benefits everyone equally.

For mulligan number one, I think the Vancouver Mulligan is fair game. However, I think the second mulligan should be fixed – a double mulligan on either side of the table makes the odds too low for the player in question and makes for poor games in terms of fun, frustration and game play, as the odds of having that player do nothing are too high.

In casual formats, I've seen players try out many things – starting with two basic lands in hand, having free mulligans, sending a set number of cards from their hands to the bottom and drawing as many minus one. All these rules make for fun Magic but aren't compatible with competitive Magic. To address the issue that I mentioned above, I'd like to submit a new mulligan rule:

"A player can choose to shuffle their hands into their library and draw that many cards minus one. A player can repeat this process. After that player decides to keep their hand, scry 1 for each time they shuffled their hand back into their library."

Basically, you scry 1 twice if you mulligan down to five (three times if you mull down to four). It's easy to implement, can't be abused (not much more than the current Vancouver Mulligan), and helps give a chance to the unfortunate player who has to mulligan twice. The fact that you would scry 1 twice instead of scry 2 once prevents game sequence where a player would know too much of the top of their library and abuse that knowledge.

What do you think? Is the Vancouver Mulligan the fairest way to mulligan in Magic? Will there be other mulligan rules changes in the future? Did other games find the right solution to avoid the "mana screws" and keep their game balanced and fun?

Let me know in the comment below and if you like my proposed rule, and let's make it a reality!