If I had to guess why most games are won and lost in Magic,  the first stop on the blame train would be mulligans. A good number of players have a hard time knowing what is keepable and what would be better as a random six.  An even larger number of players don’t know when to go down to five. While a chunk of this is based on metagame, matchup, format, decks being played, and other variables that I can’t address without 8,000 different 10-page articles, there are some common rules to live by. Mulliganing shouldn’t be based on feelings and intuition—instead, a purely information-based outlook should be used to make decisions. If what you “feel” always works out, then you don’t have to change a thing. But if seven-land starting hands just repetitively do poorly for you, let’s bring in my favorite guest star: math.

Mulligans the Right Way

For today’s exercise, we are going to use Limited, the most “LUCK”-based format  that exists. Take a normal 40-card deck with 17 lands, 16 creatures, and 7 spells. The CMC breakdown looks like this:

1CMC = 1

2CMC= 4

3CMC = 7

4CMC= 6

5+CMC= 4

Lets draw a few hands with this deck.  First on the list is the five-land hand. It contains a three-drop and a five-drop. The three-drop is a  creature  and the five-drop is a kill spell (welcome to Theros block Limited). The question is: do I mulligan to six?  Our first thought might be, “There are only 12 lands left, so odds are that I don’t flood here.”

Let’s let our friend math take over from here. You are on the draw. On turn one, you have a 36.4-percent chance of drawing a first land and a 12.5-percent chance of drawing a second land. That means that more than 10 percent of games you start with five lands, you will get flooded out.  Yes, 63.6 percent of the time, we will draw either a land or something costing four or more. For your second draw, you are looking at 39.75 percent of the time where you will not have drawn anything relevant to play in your first three turns.

If our opponent is playing a two-drop on turn two, a three-drop on turn three, a four-drop on turn four, and either a five or a combination of two- and three-drops on turn five, it will feel as though we got flooded. If that turn two is [card]Vanguard of Brimaz[/card] and they bestow onto it turn three or four, or if their turn three is a [card]Wingsteed Rider[/card]  into a turn-four strive card, you can’t play around it almost 40 percent of the time.

I don’t know about you, but I’m not willing to keep a hand that only has a 60-percent chance at having a shot at winning. If I mulligan to a six-card hand with three lands and three spells, my chances of flooding go down drastically. I only have a 42.4-percent chance of drawing a land turn one and a 17-percent chance of getting two land draws in a row. This means on turn two I will have five lands and three spells, which is still better than the over-60-percent chance that I will draw into six lands plus three spells or seven lands plus two spells by keeping a hand of five lands. I like 73 percent way better than 40 percent for my ability to play a game of Magic. I’m not saying that there are not exceptions to the rule, but in general, five land are too many for me.

How about the one land hand?  Your odds of not drawing a land in your first two turns are 25 percent. That means that by the time you draw a land, 25 percent of the time it will be too late. At this point, it is not appropriate for you to tell me that you got mana screwed.

I think that it is safe to say that any one-land hands or five-or-more-land hands are almost an automatic mulligan in a Limited environment. This doesn’t mean that you can’t get there with them, but the odds of getting there are against you. At that point, Magic is only based on skill in 75 percent of games. If you are equally matched to the field, that means that you will win  fewer than 40 percent of your games. I’m not a Magic pro, but I would much rather play a game based on skill than luck. The rest of the time is a bit less clear.

Some Questions to Ask Yourself

When deciding whether to mulligan or not, there are a few questions that we need to ask ourselves.

Can this hand get me through the first five turns of the game? If the answer to that is no, then we really need to look at taking a mulligan. In most formats, I need to be present in a game within the first five turns. If we are  playing Legacy, we need to determine if we can survive if our opponent is running a combo deck. Do I have a way to deal with a turn-three [card]Batterskull[/card]?A turn-two [card]Heritage Druid[/card]? A turn-two [card]Show and Tell[/card]? Do I stand a chance against most of the field with this hand? If the answer is no, then you are probably ready for a mulligan. By the way, if you can combo off before they get a Batterskull, that is most likely an okay hand.

Can I survive a mana flood or mana screw? In Magic the true test of a player isn’t if she can play her deck when it’s working optimally. The true test is whether she can power through the bad hands as well. That’s the difference between the folks that get a top eight once at an SCG Open and Raymond Perez, Jr. I would bet if you asked any pro, they could tell you a story about how they were getting mana flooded or mana screwed and still won that game.

We should also define what a mana flood or a mana screw is. There have been plenty of situations where I have had an opponent complain about mana flooding with eight lands and twelve non-land cards. Just because you didn’t draw all spells doesn’t mean you’re getting flooded. Similarly, I’ve seen someone start with one land in their opening hand and get stuck on three until turn seven or eight. Statistically, this is supposed to happen in both cases. Neither are what I would call a severe mana flood or a severe mana screw. Just because your plays weren’t as good as your opponent’s, or the last thing you drew before you died was a land, doesn’t mean that you got screwed or flooded.

It all begins with your starting hand.  If my current hand cannot survive a few lands or a few non-land cards off the top, is it really worth keeping? Take into account the cards in your hand. If you have four lands, a six-drop creature, [card]Divination[card], and  [card]Wrath of God[/card], you are going to be fine most of the time. If you draw five lands off the top, you will live. If you draw five non-land cards off the top, you’ll be good. If you have four lands and three one-drops, it’s time to Mulligan.

Would an random hand of six (or five) look better than this? If the average six-card hand is better than your current hand, consider taking a mulligan. If the average five is better than your current seven, ship that hand as fast as you can. If your current hand of seven isn’t great, but it’s better than any five card hand you could get, it might not be worth the risk. But think intellectually about it—don’t make decisions based on that one time when you kept a one-land hand and ripped everything you needed off the top of your deck.

The only way to know what the average hand of five or six looks like is to look at them. It is much easier to make the decision to mulligan if you know what you are getting into. The scariest part of mulliganing is not knowing what you are going to get. The fear of the mulligan goes away once you know that you can take a hand of six cards and still win.

There’s an app for that! With technology where it is, we now have apps for our phones that can figure out odds and statistics. They make these for Magic: The Gathering if you look hard enough, or you can download a statistics app on your phone and start plugging in numbers yourself. When I’m playtesting, I love being able to see what my odds of drawing lands, creatures, spells, etc. are. Once you play with it enough, you start to intuitively know what the probability of any given play is at any given moment. It probably won’t be as accurate as a calculator, but if your brain says 40 percent and it’s actually 41, it shouldn’t change your decision.

Take the blame.  I don’t discount games where I got flooded or screwed like some do. It may or may not be my fault, but if I don’t take responsibility for it, I can’t fix it. Get some practice taking mulligans and you will get better. There are times when you should be playtesting hands of six against someone with seven.  If you know what to expect out of a six or five card hand, mulligans aren’t nearly as scary. Put the labor in to get the correct knowledge so that you can start getting LUCKy.