Some weeks ago, other writers wrote about how sideboarding was the most important but forgotten art in Magic. Though I agree it’s very important and often neglected, I think there is something that is even more important, and even less talked about — mulligans. Many times you see people saying “I lost because I didn’t draw lands,” or “I lost because I mulliganed to five,” and that seems like a good excuse, but in those situations you have to ask yourself… were you really that helpless?
Talking about mulligans is very, very hard, because most of the time there isn’t a definite answer. How are you going to prove the hand should be mulliganed or kept? You aren’t very likely to draw the exact same hand once in a tournament, let alone the hundreds of times you’d need for empirical data. I suppose you could just draw seven and then draw the next cards and decide if you liked to keep it, and then repeat this process some hundred times, but that’s so situational — sure, you might even learn if that hand was a keeper or not, but what if that Tarfire was a Mogg Fanatic instead? You would have to try hundreds of hands again.
So, the point of this article is not to make you memorize what hands you should keep or not, but instead go give some insight on the process of mulliganing hands, so you might be able to decide for yourself when you encounter a situation you haven’t before — which will be most of the time with opening hands.
Let’s start with, for example, this hand on the play:
Do you keep it?
The answer yes, you keep it in a heartbeat. Why? Because this is the deck you’re playing:
This is, in fact, the best possible opening hand for this deck. My point with this example is that mulliganing outside of a context does not exist. You very rarely “mulligan this hand” — instead you “mulligan this hand, on the play, against this deck, when this is the list you are playing.”
The first thing to know about mulliganing is what a mulligan actually is. Mulligan isn’t “I don’t like my hand,” and it isn’t “I’m not feeling lucky today so I’m not going to gamble on this hand. Mulligan pretty much says “I believe that I have a higher chance of winning the game with a new hand of one card less than with this current hand.” That means some automatic mulligans in some situations are automatic keeps in others — just like the hand I just posted would be an auto mulligan in your average Mono Red deck, but it’s an auto keep in mine.
Though the main idea is always the same — trying to win the game — there are a lot of different reasons to think your hand has less chance of winning the game than your next. One thing you have to know is that I’m a pretty heavy mulliganer — if I think a hand doesn’t have what it takes, I have no qualms about shipping it back. In fact, I’ve only found one person in recent memory who mulligans as much as I do, and that’s Martin Juza.
Now let’s go through the most common situations:
– The one-lander – This is the most common mulligan situation, and most times it’ll be pretty obvious. Still, there are many situations in which people should mulligan because they don’t have enough lands and they simply don’t. I’ve seen many Pro Tour players keeping hands that just hope to get there when to me they are clear mulligans.
One mistake people make is that they always associate drawing a land to automatically winning the game when that is not true. Sometimes people will think “I have 47% chance to draw a land now, and since the matchup is worse than that there is no reason for me not to keep because my odds are better this way,” but, of course, you have to take into account the probability of actually losing the game even if you draw a land. I’m a notorious hater of keeping one-landers, and, though I’ve learned it’s correct with some Eternal decks, this is about as far as I would go. In the void, I’ll always mulligan hands with one land (by that you read one mana producer). Take this hand:
Assuming you’re playing your average Faeries deck, I’d always mulligan this, against any opponent, on the play or on the draw. Many people would keep it with the hopes of “getting there,” and that goes back to the decision of playing Dredge in Legacy — don’t you trust yourself and your deck, that you would want the outcome of the match to be decided on things that are not your playskill? With this hand, sure, if you draw a land you’ll be able to use all the deck/player advantages you have, but if you don’t, it doesn’t matter how good you are or how good your deck is… you are as good as dead. With this hand, on the play, the probability of drawing a land on turn two is 45%. Even if you assume you can skip one land drop, or if you are on the draw, the chance of drawing one land in your first two draws is 70% – that means there is a 30% chance you flat out lose the game on the spot (barring absurd scenarios, such as your opponent also keeping a hand like that). Even if you give yourself a 60% chance you win in case you do draw a land in the first two turns, it’s still only 42% chance you are winning this game — and 60% chance to win when you miss the turn 2 land drop is pretty optimistic.
– The one-lander six-card hand — this is a whole different world. I’m still very much against keeping them, but I reckon that in some situations it could be correct. Take the same hand as last time, with one less card:
I would probably keep this hand on the draw against an unknown opponent, and I’d mulligan it on the play.
One thing I see is that a lot of people have a mortal fear of going to five — they equate mulliganing to five to an automatic loss, and that’s far from true. Take, for example, LSV — he mulliganed to five both games in his first match in Kyoto and won both of them. I’ve won games with mulligans to four as quick as turn 4 myself. Mulliganing to five is part of the game, and saying you cannot win is like saying you cannot beat turn 2 Ravenous Rats, turn 3 Ravenous Rats.
Also, consider the scenario where you mulligan to six and keep a bad hand because you were afraid of going to five. Then your opponent mulligans to five himself after you, and beats you with his average five-card hand because your six just didn’t have what it takes — now you could have five cards in hand yourself, and fight a fair battle, but instead you decided to gamble.
You don’t have to look that far to see something like that (granted, not a one-lander, but same concept) taken from the GP: Kobe coverage:
“Takahashi sent his first two hands back and was quickly down to only five cards to work with. I had been expecting Tsuyoshi to mulligan as well – he was looking at six lands and a sideboarded Canonist — but he kept the hand and I am not sure if he would have done so if not for the five-card starting hand from Takahashi.”
The end result was that, even though Fujita drew the Spell Snare for the turn 2 Goyf, he just couldn’t deal with the Garruk that Takahashi followed up with, which won the game and the match. Not to say that Fujita would have won the game had he mulliganed that hand, but just to show that your opponent going to five, though relevant, is not an excuse for you to keep lousy hands.
One thing I’ve noticed is that people have some sense of entitlement. I used to be like that some years ago — nowadays I’m pretty much the opposite. People play against a worse opponent, with a bad deck, and they think “it’s not fair that I don’t draw the land I need AGAIN,” as if they just deserved to draw the cards they needed because of the situation. Well, it doesn’t matter how many times in a row you did not draw the land you needed — the chance to draw the land this time is exactly the same as it was before. Don’t be a victim of the Gambler’s Fallacy. It also doesn’t matter how much more you deserve to win, or you think you deserve to win — the top card is not going to change because of that. You should not be a believer in “the heart of the cards” — rather, believe in doing the right thing and then being rewarded for it, instead of being rewarded beforehand.
– The “this hand doesn’t do anything” — now we get beyond the obvious “no-lander.” A lot of times people will not mulligan hands that don’t do anything because they don’t look like mulligans to people who decided just to memorize a book on mulligans and keepers.
Say you are playing the Naya version from Saito, that won GP: Singapore, and your hand is this:
This hand doesn’t do anything — you should just ship it back. Now, most people would ship this hand back, because it obviously doesn’t do anything. Now, change the Umezawa’s Jitte for Mogg Fanatic and… well, it still doesn’t do anything! Yet a lot of people would keep it, because you have creatures, lands, and spells, and you can cast everything and you have access to all your colors.
There are two important things in mulliganing hands that don’t do anything — first is what you actually have to do, second is what your deck is capable of doing. Sometimes you just can’t help it if you don’t know what they are playing — for example, in Hanover I was playing Mono U Faeries and I drew the following hand —
This hand is one of those “doesn’t do anything” hands, so I mulliganed it. It turned out my opponent was playing Elves, and this hand would’ve been pretty good, but at that point I had no way of knowing. This hand is good against Elves and good against Zoo, but terrible against everything else and, again, I’m not a gambler. I was also in the two draw bracket by then, which made it more likely that he was playing something this hand would be awful against, but I’d mulligan this even if it was the first round of the tournament. Had I known what I was playing against, I’d know what I have to do — against Elves, this hand is not a “do nothing.’ It’s a “do everything you want done.”
The same happens with Zoo versions. A hand of Kird Ape, Gaddock Teeg, and five lands is a do-nothing against Elves, but a good hand against Desire. In general, if your hand is one of those “do-nothings” versus one kind of deck (not just “do nothing versus Elves” but, say, “do nothing versus control decks”), you should mulligan rather than hoping you’re playing against a deck against which your hand is good.
The other side is that you may know what you want done, but your deck might not be capable of it. You know that against Elves you want to kill their creatures to stop them from comboing, but if you are playing, say, Bant, you’ll mulligan to zero before you find a hand that has the necessary tools to beat them because, more often than not, those are not in your deck. This happens a lot in Limited — sometimes, your deck just cannot provide the kind of hands you want. I see people who decide to play five-color control decks in draft and then mulligan hands of four lands, two Obelisks and one big creature because they don’t do anything — but, in this situation, you have to ask yourself — can your deck do much better? Sometimes the answer is no — this is going to be your average hand with that deck. What are you going to do then? Mulligan until you draw the one Infest you have? Sometimes it’s better just to hope they don’t have a very good draw than to try to mulligan aggressively for a hand your deck will likely not provide.
This is especially common after sideboarding — for example, people side out their one- drops in Zoo and then mulligan a hand because it doesn’t have one-drops. Well, if you were going to mulligan all the hands without one-drops, maybe you shouldn’t have taken them out? When you are sideboarding, you have to take into account how this is going to impact your mulliganing decisions as well. You are playing a completely different deck now, against a completely different deck — why would your mulligans be the same?
One good example for that is the deck I played in GP: Chicago — the same as Nassif. With that deck, I was mulliganing most one-landers with a draw spell. Later on a friend pointed out that, if I was going to mulligan all those hands, I needed more than 21 lands. That made perfect sense — I was trying to get something that my deck would likely not provide.
– The impossible matchup — sometimes you are faced with matchups you can’t realistically win, if not for a card or a combination of cards. In that case, mulligan for them!
Let’s take it you are playing the Mono White Martyr deck against TEPS. You cannot win that game in fair terms. You do, however, have a couple of Runed Halos in your deck — there is a very big chance they just cannot beat a Runed Halo game 1. In this case, which hand would you rather have:
The three-card hand seems much better in this matchup. You have to know when you have hopes of winning and when you don’t. Playing turn 1 Martyr of Sands and attacking twenty times does not count as a hope of winning — you simply need that Runed Halo. Don’t be afraid of mulliganing to 3, or to 2, or even to 1 if your hand has no non-lands, Runed Halo cards — you might as well treat everything in your deck that is not Runed Halo, a way to find it, or a way to cast it a complete blank.
– The hand that already has a mulligan — sometimes people will draw hands with the likes of two Legendary lands, or a card that is completely useless (like a maindeck Naturalize against a deck you know has no targets) — in this case, it is like you are already one card down. People are more likely to mulligan those hands, because they “already have a mulligan” — which might not be true because they might play, say, Blightning and then you’ll be glad you had that Naturalize, but let’s assume it is — but in this case you have to take that card out and then think if you would be happy with the next six cards if they were your hand. You are basically being given a free mulligan, EDH style, but you don’t have to take it — your next hand might as well be worse, and then you’ll have to go to five.
Some hands, however, are already mulligans and people don’t realize it. A Zoo hand with five lands is as good as mulliganed already — take out one of those lands and see if you’d like that hand on six. The same goes for multiples of cards you could play but don’t really want to — like three Bitterblossoms in a BW deck, for example.
– The Limited mulligan — most of the situations I described happen in Limited too, but almost always to a lesser extent. Mulliganing in Limited is a completely different animal, though. Some people just keep any hands with lands and spells — I’m not one of those, but I still keep a lot more hands in Limited than I do in Constructed. The reason for that is that in Limited you have a lot more room for mistakes, in Sealed more so — in the current Constructed formats, you can’t just keep a mediocre hand to see what’s going to happen. In Limited, you have a lot more time. Still, that doesn’t mean you should keep everything.
Most of the wrong mulligan decisions in Limited happen on color issues. Take, for example, this hand in your average, RGw Naya draft, with six Mountains, one Jungle Shrine and one Naya Panorama for Red Mana, and 10 cards that require Red to play (so eight left in your deck), on the play against an unknown opponent:
I know a lot of people who would keep this hand, but in my opinion it’s the wrong decision. Sure, if you draw a Mountain, it’s a very good hand, but if you don’t it doesn’t actually do anything, since you can’t play a spell before turn 5 — and that’s if you draw more Forests and Plains. For every Red land you can draw, there is a Red card you can’t cast left in your deck — in many games when you draw that Red land it’ll just be too late. In this case, you have about 43.5% chance to draw a Red land in your first two draw steps — certainly not enough for me to keep it.
Sometime ago, there was a mulliganing exercise on MTG.com in which a hand of six lands and a Triskelavus was presented in a draft deck. The overwhelming majority of the posters said they’d keep it — I believe this is just wrong. You have some time, sure, but you don’t have infinite time, and Triskelavus is not Martial Coup that you’ll just win if you get to play. Of course, sometimes you’ll draw spells for the rest of the game and curve into the best end-gamer in your deck, then you’ll win, and sometimes you’ll draw two lands and a guy costing five and lose. Many people work under the “there are so many lands in this hand, there can’t be many more left in my deck” — still, there are plenty of lands left there, and every one you draw after that can also be counted as a Mulligan — the more lands you have in your hand, the less likely you are to draw one, but the worse it gets when you actually do.
One thing I’d also like to talk about, that I’ve seen a bit, and also experienced, and I believe is far more common than we can imagine is that sometimes people Mulligan because then they have something to blame for their loss. They need to win their match, then they have an okay hand, and they Mulligan it, and then they don’t even think about it properly and mulligan to five, and they are already envisioning how it’s going to be when they walk away and tell their friends “I couldn’t do anything, I just mulliganed to five both games” — it’s much easier than saying “I kept this six-card hand and lost because of it.” In one of them, it looks like it’s your fault; in the other one you just blame variance or something. Still, it might be your fault just the same — no one wants to be blamed for their own losses, but you can’t really let that get in the way of your Mulligan decisions.
I hope you’ve enjoyed this article — as I said, it’s hard to write about without being either too obvious or overcomplicated. If anything, just remember that you are in control of more than you think you are, and many matches are decided in that single decision. When you lose a game and you feel outdrawn, or too slow, or that you just didn’t draw enough disruption, just think of what you kept and see if you could have done anything about it.
Before I finish today, I’m going to leave you with a couple of scenarios that I believe are controversial and could be a good point of discussion. Next week I’ll get the opinions of a bunch of Pro Players, and then you can compare what you thought with what they have to say about them.
I’ll list the decks so you know what you are playing. This is not to mean I think this is the best version of the deck or anything — it’s just a reference so you can properly answer the question. All the examples are against unknown opponents unless specified.
1 – Playing this:
4 Spellstutter Sprite
4 Vendilion Clique
2 Sower of Temptation
3 Venser, Shaper Savant
4 Ancestral Vision
4 Mana Leak
4 Spell Snare
2 Thirst for Knowledge
2 Vedalken Shackles
3 Engineered Explosives
3 Umezawa’s Jitte
2 Chrome Mox
3 Riptide Laboratory
1 Breeding Pool
1 Steam Vents
1 Hallowed Fountain
1 Flooded Strand
On the play:
On the play:
2 – Playing this:
4 Mistbind Clique
4 Scion Of Oona
2 Sower Of Temptation
3 Spellstutter Sprite
1 Agony Warp
4 Broken Ambitions
4 Cryptic Command
1 Remove Soul
2 Jace Beleren
4 Secluded Glen
4 Sunken Ruins
4 Underground River
On the play:
Playing against the Mirror, on the play:
3 – Playing this:
4 Broken Ambitions
4 Cryptic Command
4 Esper Charm
2 Remove Soul
3 Volcanic Fallout
3 Ajani Vengeant
2 Cruel Ultimatum
3 Wrath Of God
2 Cascade Bluffs
2 Exotic Orchard
2 Flooded Grove
1 Mystic Gate
4 Reflecting Pool
3 Sunken Ruins
4 Vivid Creek
3 Vivid Marsh
3 Vivid Meadow
On the play:
On the draw:
On the play, against Five-Color Control (Nassif’s Version):
4 — You’re playing this:
2 Wild Nacatl
3 Wild Leotau
2 Resounding Thunder
On the play:
5 — You’re playing this:
2 Druid of the Anima
2 Corpse Connoisseur
2 Ridge Rannet
Yoke of the Damned
Obelisk of Jund
Gift of the Gargantuan
On the draw:
On the play:
That’s it! See you next week when I’ll get some company to talk about those hands.