Rhythmik Study: Taking a Mulligan on Current Theory

In a game like Magic, variance is king. In a game with infinite possibilities between card draws, deck configurations, and game states, no two games will ever be exactly the same. Even over the course of a game, certain plays become more advantageous than others and one judgment call can win or lose a game, even if you don’t realize it. Questions like “Should I cast Path to Exile when my opponent is at five lands? He may be holding a Broodmate Dragon” or “Should I kill his Vengevine and open myself up to card disadvantage from recursion?” What is the right play this turn may be horrible next turn, and vice-versa. All of these questions are answered by judgment calls. Since you have no access to the information your opponent is holding, or what he is going to draw next turn, you have to make decisions based on the information available to you.

Lately, I’ve been talking with players about advanced game theory, and it’s becoming alarmingly prominent that players are beginning to shoehorn game theory into absolute “Yes or No” answers – especially concerning mulligans. With a game that has such a high level of variance, making judgments based on one factor as opposed to the entire picture can easily cost you a game, a match, or money in a tournament. During Grand Prix D.C., I was having a discussion with a player who also made Day 2 about one land hands, telling him that not every one-land hand his unkeepable. He simply dismissed my argument, saying that he threw away every one-lander. In my opinion, this method of thinking is absolutely wrong. There are too many factors to evaluate to throw away a hand that can easily win a game simply because of one factor that may look bad on the surface.

Over the last two weeks, I conducted a poll among the better players in my area, as well as a couple of my friends on Magic Online concerning mulliganning one-land hands.

Question One: In Standard, do you always Mulligan 1-land hands?

60% Yes
40% No

The general consensus was that you should always or almost always mulligan one land hands. While this survey was meant to be simply Yes or No, everyone that said they wouldn’t also claimed that there were very specific exceptions to throwing the hand back. I find it somewhat alarming that sixty percent of players I surveyed answered that they would always mulligan a one-lander.

Question Two: Would you mulligan this hand? (Given information: 1.) You are on the play. 2.) You are in Game One. 3.) You are playing Vengevine Naya (decklist here).

90% Yes
10% No

I definitely agree with these results. In a format with as much one-cost removal as the current Standard format, a hand such as this makes you dependent on the Birds of Paradise, and if your opponent realizes this, you can be in a bad position when they remove the Birds, especially if you miss your second land drop. Even though the hand can be relatively explosive if you hit the untapped land on Turn 2, you have a much better chance for a winning hand by just taking a six.

Question Three
: Would you still mulligan this hand in a sideboarded game, if the information derived was as follows: 1.) Your opponent is playing Mythic. 2.) Your opponent runs the Sovereigns of Lost Alara Package. 3.) You are on the Play.

10% Yes
90% No

This is the figure I am happiest with. This is the situation I was in during Round 13 of Grand Prix: DC, and is the hand that brought up the conversation about one-land hands. On my Turn 1, I played the Birds and passed, and on my opponent’s Turn 2, he played a land and passed. On my Turn 2, I drew my card and passed the turn, to which my opponent responded with a bit of surprise.

“Ouch, a one-lander.” My opponent remarked.

“It’s a ridiculous one-lander,” I replied.

On my opponent’s Turn 2, he played a Lotus Cobra which I immediately bolted, and passed the turn. On my Turn 3, I drew an Arid Mesa, and cast my Cunning Sparkmage, and my opponent knew exactly why I kept the hand. About seven turns later, my opponent and I were filling out a Match Result slip, and I was three points richer.

A hand like this contains all the tools you need to beat Mythic Conscription, with the distinct possibility of a Turn 2 Cunning Sparkmage, which is Mythic’s Silver Bullet. This hand, though a one-lander, should require almost no thought to keep. The ability to survive as many as three turns without a second land-drop and the possibility of disrupting the opponent beyond that point if he keeps a hand with few lands and many mana dorks is well worth the reward reaped when the second land drop is hit. The riskiness of this hand is even further mitigated by the fact that the opponent’s only early removal spell is Path to Exile, and would accomplish what the Birds of Paradise does anyway putting the opponent at a card disadvantage.

Factors To Keep In Mind:

1.) Know the Format. In order to accurately evaluate the risk vs. reward of keeping a hand, knowing the format is very vital. In Game 1, an opponent could have any number of cards in his opener, and knowing the format as well as each big deck’s popularity can really help in a decision to keep. Always assume your opponent is playing the deck that is your worst matchup. Does your current hand auto-lose to your worst matchup’s average hand? You should probably throw it back.

2.) Know Your Opponent. In sideboarded games, you should know exactly what your opponent is playing, and have a rough idea of what his win condition is. Does your hand have any of their silver bullets? Is your hand soft to any of their early cards? For example, if your opponent is playing a deck with Red, do you lose to a Turn 1 Lightning Bolt? If so, you should probably throw the hand back. Remember, for decks like Mythic and Naya, not all hands lose to a Turn 1 Lightning Bolt, and some may even warrant trying to bait out the Turn 1 Bolt, which leads me to my next point.

3.) Know Your Deck. This is arguably the most important. What are you trying to accomplish with this hand? How can you either win or prevent your opponent from winning with the cards in this hand? Would you be more likely to accomplish this with a random assortment of N-1 cards? Can you play everything in your current hand, if you were to draw one more land that produces colorless mana (or the color with the highest number of sources in your deck) in the next two turns? Can you play at least half of your hand (and approximately half of the spells in your deck) if you do not draw a land in the next two turns?

This One’s a Keeper

While every player does assess mulligans in different ways, due in part to their play style, their deck, and their personality, I have found this method to work the best for me. The key is to look at each hand objectively, using all information obtained from the current match about your opponent to aid in your evaluation. One thing I do not recommend is trying to read your opponent as he assesses his mulligans. Any player can easily bluff a bad hand, and “Jedi Mind Trick” you into keeping a marginal hand.

The best way to decide which hands to keep, just like anything else you could ever want to know about your deck is through playtesting. When testing, keep different hands to see how they play out against each matchup. The best way to figure out whether or not to mulligan is to keep bad hands during testing, since playing with bad hands shows you first-hand why the hand in question is bad.

Develop your own style of mulliganning, and learn which hands are good against each matchup and why. With enough practice, you may find yourself winning more games, more matches, and more prizes. Until next time, take a mulligan on your old strategy, and I’ll see you in Amsterdam!

Remember, as always you can follow me on Facebook and Twitter. I’ll be happy to answer any questions you may have.

~Jeph Foster (Rhythmik on MTGO)

  1. Great reminder that Magic (and especially mulliganning) is about risks vs. rewards. Everyone called me dumb for keeping one-land hands with SDTop in Kamigawa Block Gifts, but I did it all the time (and found success), even keeping Miren-Top hands if they had a Sak-elder and Kodama’s reach or something.

  2. 90% not mulliganing that one-land hand in the sideboarded game shows that the first poll result wasn’t accurate. I think the result should be read more not that they automatically mulligan, but that 60% believe they should often mulligan a 1 land hand.

    Like the 40% giving specific cases for not mulliganing, the 60% probably has that same set of cases, but used ‘unless’ instead of ‘because’.

    I liked the two other results though, and the explanation that came with them. It’s very true that some hands are keepable against certain decks and unkeepable against an unknown field. Sometimes a great hand is even not keepable if the opponent has the deck to beat exactly that hand.

    One last thing: I wouldn’t suggest assuming your opponent is playing your worst matchup. What if that matchup is only 5% of the field? Then you’re needlessly mulliganing hands that are fine/good/great against 95% of the field, just for the offchance you could have had incredible bad luck of facing your worst matchup!

  3. Very interesting article, I have recently realized that I should mulligan more, though not usually because of one land hands. The 4-5 land hands are the ones I end up keeping and wishing I hadn’t, especially if 1 or more of those lands are fetches. I have had those hands turn into mana flood games much more than I expected.

    Also, I agree with the last point Zage made, unless your worst matchup is really prominent, or your other bad matchups are close to your worst matchup.

  4. Nice article Jeph, I definitely agree to most of your points, and this kind of thinking has definitely allowed me to exponentially increase my game after returning from a long break.

    The thing about Magic that I enjoy, and sometimes hate, is that in certain situations, making abstract plays and/or awkward calls will sometimes put you ahead in the game. Looking into plays and hands subjectively is also a great lesson that I would encourage many players to learn early, it makes mulling a lot easier (doing so effectively is pretty difficult, especially with the factor of variance,) and also gives a player the baseline for becoming a skilled mage who looks at situations on a per-turn, per-player, per-deck basis, which allows them to effectively decide the values of their cards and plays for the entire span of the match.

    Great article, hopefully I’ll catch ya at the next PTQ buddy!

  5. I love mulligans. When are you going to bring me some live-play videos?! I want to hear the voice of you!

  6. Love, love, love this article, Jeph. You continue to bedazzle me!!!

  7. It is a bit shocking that people play on automatic so often. I will always gauge the individual hand, my opponent’s deck (if known), if it is a game with or without SB and so forth before deciding on a mulligan. I’ve even kept no mana hands on the draw against certain decks like Reanimator and/or Ichorid if the hate is there (Crypt, Macabre, Leyline).

    All in all a very nice article that I found extremely informative.

    Keep ‘em comin’.

  8. Never thought about practicing with bad hands and trying to figure out whats good and bad about them. Normally I just go with the odds. High mana costing hand I’ll throw back with one land. If its low casting cost I’ll debate about keeping it.