-
B Smith
United States
Boston
Massachusetts
-
The new year has arrived. It's a time for new beginnings. So let's talk about the easiest way to get a new beginning in a game of Magic: mulligans.
Actually, the real reason I want to talk about mulligans is because over the past year, I've been training myself to mulligan more and more aggressively. Now I'm starting to feel the pendulum in my brain start to pull back in the other direction. That tells me it's time to write down my thoughts and get them organized in my head.
What I'm writing here is general advice for Sealed Deck games. It's probably useful for Booster Drafts too, but you may have higher or more focused expectations considering that you presumably have a more focused deck. It may or may not be useful for Constructed formats, where hyper-focused decks may have very detailed mulligan considerations.
There are three big questions I try to keep in mind when I decide whether or not to mulligan.
What will the board look like after turn three?
Often people will boil this down to more specific questions about the number and types of lands in hand, having creatures you can cast, and so on. Those are all important factors to consider in this question, but I also try to look at the board state holistically—sometimes strengths in one area might outweigh a weakness in another, or clever play can fix what looks like a problem at first glance.
At the end of turn three, I ideally want to have access to all of my main colors (as opposed to colors I'm splashing—I should be able to do without those for a while). I can stand to be missing a color, but that makes it more important that I can cast the spells I have with the land that I have. Any spells that I draw in the missing color will be dead weight, and I really don't want to have a hand full of spells I can't cast.
I also ideally want to have a couple of creatures on the board. One creature and another useful permanent is usually acceptable. The further I dip below that benchmark, the more I have to be able to play control for a while: without creatures, I need to have spells that can hold off my opponent early, and a plan to overpower them later.
This means that, after game one, I need to consider what my opponent's side of the board can look like too. If one deck's fast while the other one's slow, the early game deserves more consideration—that's when the fast deck works to establish an early edge that it can ride to victory. The closer the decks are matched in speed, the harder it is to outpace your opponent. Finding other ways to dominate the game is more likely to bear fruit.
If I mulligan, is that hand likely to be better?
If your hand is wanting for some reason, next consider some of the objective reasons for that. What are the odds that you would draw a new hand with the same misfeatures as this one? The higher they are, the less likely it is that a mulligan will solve your problems.
This is one of those parts of Magic where a working knowledge of statistics comes in handy. Walking through the math would be a post in its own right, so I'm not going to do that here. For now, I'll just point out that the quick numbers people compile about decks—average mana cost, mana symbol counts, creature counts, land counts, on and on—are all reference points we use, in part, to help us make these decisions quickly. If you draw a hand where the average mana cost of the spells is 4, a mulligan is more likely to help you in a deck with an average mana cost of 2 than one of 3. Drawing a hand with no creatures is a lot more likely in a deck with ten creatures than one with fifteen.
Since you typically only have thirty minutes to build a deck in Sealed Deck, you probably won't have time to compile all these numbers for your reference. Fortunately, we're often building decks that are designed to hit common basic guidelines: seventeen lands, fifteen creatures, an average mana cost around 3, and so on. These numbers pull double duty by giving you a reference point in your mulligan decisions. You just have to remember if you've deviated from them in some significant way.
If a mulligan isn't statistically likely to help you, it's probably not worthwhile. That's not to say that you don't have a problem, but it's a problem that needs to be solved by fixing your deck more than any individual hand.
Does the potential reward justify the risk?
If a mulligan can help you, now you have to decide whether or not to plunge into the unknowable. If you shuffle up and deal again, your chances of drawing a bad hand are worse than they were before, since you'll be down a card. Do you stand to gain enough to justify the risk of losing a card to no benefit?
I don't think this question is fundamentally answerable. Sure, there are specific situations where we all agree: if your first hand doesn't have any land in it, go ahead and mulligan. But as the potential rewards shrink, the answers split more and more. This is why mulligans are an evergreen topic for strategy discussion, and why Luis Scott-Vargas can always spark discussion by putting mulligan puzzles at the bottom of his articles. Nobody can answer this question for you. You'll have to bring all your experience with the game to bear to make the best judgment you can.
While I think I've improved my ability to answer all these questions, probably the single biggest reason I'm trying to mulligan less these days is because I have more respect for the risk it entails. As I learned to care about the makeup of my hand beyond the land count, I would be quick to throw away borderline hands. This almost guaranteed I would be starting with five cards or fewer at least once per tournament—sometimes twice. That was more than I was willing to take, so now I'm more likely to keep hands that might make for a bad early game, as long as they let me survive and set me up to cast what I draw later.
As I play even more, I'm sure I'll gain even more experience about how to evaluate the value of a hand and the risk of pitching it. Maybe in another year the pendulum will swing again and I'll need to write down all my newfound wisdom in another blog post. If you want to speed up that process, please share your own Limited mulligan considerations in the comments. For now, I think this covers the fundamentals of what I know.
|
|
