Welcome to another edition of Keep or Mulligan! The idea is pretty simple: I present some hands, their context, and you choose Keep or Mulligan. After that, I’ll give you my answer and my reasoning.

Hand #1

You’re playing an aggressive, fairly average RG deck in a Dragons of Tarkir/Dragons of Tarkir/Fate Reforged draft. Your mana is 9 Mountains, 8 Forests, and you’re on the draw. Your opening hand is:

Keep or Mulligan?

Solution

Mulligan.

It’s tempting to keep this hand because you have two plays if you draw a land, one of which gets you back in the game depending on what they have, but I think that ultimately this land is just too much risk for no payout. You’re going to flat-out lose the game a fair portion of the time when you don’t draw a second or third land (or green mana), and the times you do draw it, the hand is not even that good, it’s just an average hand. I think you can do better with six.

[collapse]

Hand #2

You’re playing Modern Junk against Zoo. This is your mana base:

You’re on the draw and your opening hand is:

Keep or Mulligan?

Solution

Keep.

This hand is basically a math exercise, but there’s no way for you to calculate all the possible permutations during the time you are allowed for a mulligan. If you are playing competitive Magic, then this has to basically be instinctive. What you need to do is try to have a rough assessment of whether a) you’re likely to hit or not, b) how good your hand is when you hit, c) how good your hand is when you don’t hit, and d) how good the matchup will be if you mulligan. This hand makes things quite complicated, however, since there are multiple opportunities for a “hit”: drawing Forest on turn one is perfect, but drawing it on turn two is likely still a hit.

If I am playing in a tournament and I have this opening hand, this is my thought process: “If I hit in the first two turns, then this hand ranges from perfect to good enough, depending on what I hit and on what turn. I’m playing against a punishing deck, and I’m on the draw—that means there’s no opportunity for me to recover from a stumble, but it also means that the range of hands that I would consider optimal in the matchup is very low. If I go to six cards, on the draw, then the likelihood of having a hand that wins is small. I think the likelihood of just hitting a land with this hand is better, and that gives me almost perfection. Keep.”

Considering all of this, I would keep the hand. I wouldn’t try to do math in my head because I know I wouldn’t have time to do that. I would try to analyze how good the matchup is and whether the hand is good enough if I hit. In the end, I would deem this hand good enough, and I’d think that I’d need to get considerably lucky to get a hand that is good enough on the draw on a mulligan to 6.

Since this is not the tournament and we actually have some time (and a calculator), we can try to figure out the odds of these things happening. I’d say there are four different outcomes:

  • You draw a land that comes into play untapped on turn one. There are 17 of those, which gives you roughly a 30% chance that it will happen. If that is the case, then the hand is excellent and you are in very good shape since you’ll have a turn-two Thoughtseize + ‘Goyf either way. If this happens, you’re a big favorite.
  • You draw a tapped land on turn one, or an untapped land turn two. Unless my math is wrong, the chance of this happening is about 27%. If that happens, you’re in still in good shape—you play turn-two Seize + Hierarch, and on turn three you have two big blockers. In this scenario, I’d still consider you a favorite.
  • You draw a tapped land on turn two or an untapped land by turn three. The chance of that happening is roughly 20%. If this happens, then you can play a turn-three Thoughtseize + Tarmogoyf, which is mediocre but better than many six-card hands can do. In this spot, you’re not a favorite.
  • You don’t draw a land by turn three and lose on the spot. There’s roughly a 25% chance of this happening.

Overall, I’d prefer taking my chances that scenarios 1 and 2 happen, since I think you’re not a favorite to win on a mulligan on the draw to begin with.

[collapse]

Hand #3

You’re playing the UB control mirror match in Standard. You’re on the draw and your opening hand is:

Keep or Mulligan?

Solution

Keep.

The control mirror has historically been about making your land drops, and the UB mirror is no different—most of the turns consist only of someone playing a land and saying “go” until they have to discard and get rid of a useless removal spell. By the time the action starts, you’ll have drawn a bunch more cards and will likely be ready for it. If you mulligan this hand, there’s a chance you’ll get a 2-3 lander and you’ll be stuck on five lands while your opponent has nine, and that’s a situation you do not want to risk because then it becomes very hard to fight for anything.

It’s also worth noting that, barring any Thoughtseize shenanigans, your opponent is not going to know that you have nothing, so you don’t actually need to have anything. He won’t play into your counterspells, so you won’t be casting them even if you have them. Even if your opponent knows you have nothing (such as with an early Thoughtseize), there’s very little that they can do to punish you—the only thing that comes to mind is Ashiok, and most UB versions aren’t even playing it. In the end, you’re assuming the risk of losing to an Ashiok that they might not have or might not draw for a hand that is otherwise great.

[collapse]

That’s what I’ve got for today! If you have a hand that you think is interesting and would like to see featured in the coming weeks, just leave it in the comments and I’ll take a look. Some conditions:

  • You must be playing a competitive deck, because I want other people to be able to relate and I also don’t have enough experience with all the rogue decks to be able to have an opinion. The format must be Standard, Modern, Legacy, or Limited. If it’s Limited, I’m going to need a good explanation of the deck.
  • You must give me all the relevant information—what you’re playing, format, which game it is, sideboarding or not, play or draw, whether you know your opponent’s deck or not, whether there is anything unusual with your deck list (I don’t need the whole deck list, I just need a general idea of what’s going on that might impact the decision).
  • The hand has to be at least interesting. Don’t submit a 0-land hand that’s obviously unkeepable, for example, and don’t submit a hand that is clearly great but “didn’t get there.” Something you’re genuinely unsure whether you should keep or not.

See you next week,
PV