Alright, I’m going to try to set up another situation for you to judge and lead into another topic. Being a work in progress, let me know if the problem is poorly presented or doesn’t really apply because I’m going to need to use these eventually.

You’re playing something like Florian Michelac’s build of white weenies, and your opponent is playing something like Pierre Simoni’s Jund Cascade deck (both can be found here). Scenario:

Your opponent: 2 cards in hand (1 is definitely Jund Charm, unsure of the other), 1 red mana open (and all colors available to play jund charm next turn), no creatures in play

You: Knight of Meadowgrain in play, 2 white mana open, and Honor of the Pure, Harm’s Way, and Figure of Destiny in hand. Here’s what you know:

- If you swing with Knight of Meadowgrain alone and it gets through, you have a 1/2 chance of winning
- If you swing with Knight pumped by Honor of the Pure and it gets through, you have a 2/3 chance of winning
- If you can’t hit your opponent this turn, you have a 1/3 chance of winning

It’s your precombat main phase. What do you do, and more importantly, why?

This puzzle is really poorly thought out.

What I suspect you’re looking for in this case is calculating the odds that your opponent has a lighting bolt in hand, and using that information to adjust the odds of victory of a given plan. However, without knowing how many bolts are in his graveyard, and how many non-bolt spells we’ve cascaded to the bottom of the deck, we can’t actually make this call. Also, we don’t know our own life–if we’re below three, then it’s a safe bet he he doesn’t have a bolt because he’d have bolted us last turn!

Furthermore, the 1/2 chance of winning seems odd given the Harm’s Way in hand. Is Harm’s Waying the opponent to the face equally valuable as getting through with Meadowgrain? What are the odds of winning if we get through with an unpumped meadowgrain AND Harm’s Way damage to the opponents face?

In short, there is no good answer here. Any course of action is pure guesswork.

Obviously we must assume the opponent may have Lightning Bolt in hand. We don’t know the life totals, so we need to go by an abstract win percentage. Here are the scenarios:

Scenario One: Swing with the Knight alone. If he has LB, use Harm’s Way to redirect 2 damage back at him. Next turn, the opponent plays Jund Charm. 50% chance of victory.

Scenario Two: Swing with the Knight alone. If he doesn’t have LB, pass the turn. Next turn he plays Jund Charm. You redirect the damage at him with Harm’s Way. You win. 100% chance of victory.

Scenario Three: Play Honor of the Pure and swing. He has LB. You have a 33% chance of victory. Next turn you play Figure of Destiny and back it up with Harm’s Way mana.

Scenario Four: Play Honor of the Pure and swing. He doesn’t have LB. You connect. You have a 66% chance of victory.

Scenario Five: Pass the turn. Clearly suboptimal. You have a 33% chance of victory.

Scenarios One and Two rely on attacking without Honor of the Pure. You have a 50% chance to win if he has LB, and a 100% chance if he does not.

Scenarios Two and Three rely on attacking with Honor of the Pure. You have a 33% chance to win if he has LB, and a 66% chance if he does not.

Scenario Five relies on passing the turn. You have a 33% chance to win if he has LB, and a 33% chance if he does not.

Thus, it doesn’t matter if he has Lightning Bolt or not. The correct play is to attack without playing Honor of the Pure to give you the maximum chance to win.

[…] "The Theory and Design of Magic: The Gathering" Class Page All about teaching a class on Magic « You Make the Play: bolt edition […]

Reyemile: Yeah, it probably was poorly thought out, but perhaps even more poorly explained about what I was looking for. You were right on the money about the factors to consider here, which is what I was getting at. I think an inherent problem with using Magic examples versus classic Game Theory is that game theory can pull examples out of their butt that are perfectly contrived. I’m having a lot of difficulty coming up with contained examples in Magic, so let me know if you have any tips on how I can write these up better. Absolutely a work in progress. And I consider swinging with the knight better than harm’s way because of the lifelink, though I guess I didn’t really special case the 4 damage from both. I’ll work that situation out of the example. Thanks for being critical.

Salivanth: Yeah, absolutely agree with your analysis, except scenario 2 as described in my take 2 post. I try not to deal with life totals because I think they’re a game mechanic that is more a game mechanic that a good payoff metric, since there’s so much more to the game there. I guess you just have to trust me on those numbers?

Salivanth, you’ve thought out all the scenarios well–however, where are you getting the %100 chance of victory from?

Kevin: Fundamentally, the problem is that you’re trying to simplify a complex game. MtG has too many variables to summarize a game as a “50% chance of victory”

That being said, the situation you set up was an interesting challenge. You’re clearly trying to reach beginners, and that’s great. But a better question would be, “what information do you need to make the right play here?” That question will get readers THINKING about the information holes that inevitably crop up in any hypothetical magic situation, rather than encouraging them to gloss over them.