The Danger of Cool Things is an article written by Chad Ellis a while ago for the dojo. It describes how players are biased towards “cool” things. Players will make a “cool” play even if it is slightly worse than the alternative. In addition, people will choose “cool” decks like combo even when they are not good for a metagame. People can even overrate how a deck like Time Sieve is doing in testing, because every individual win is so “cool,” it seems like it’s winning more than it really is.
I think there is another side to the psychology of making bad decisions: scary things. Essentially, people hate not playing Magic. You’re at a Magic tournament for a reason, and it most likely isn’t for the chicks or the money. However, if your goal at a tournament is purely to win, it is important to realize that sometimes the correct decision is to choose an option that sometimes doesn’t let you play Magic.
When you make a decision between two options, you are making an expected value calculation in your head. Expected value multiplies the weights (how much you care about something) by the probabilities of something happening. In Magic, the weight is simply the likelihood of winning. For example, if you have a 30% chance of winning with one line and another line involves flipping a coin, but gives you an 80% chance of winning if you do hit, the coinflip line is 4/3 as good.
Unfortunately, our brains are not perfect at doing expected value calculations in our head. They tend to bias away from “scary” options, or options that sometimes give you a 0% or low chance of winning. Most people would rather have a guaranteed game in which they are behind (option 1 in the example above) then a 50% chance of being way ahead and a 50% chance of just being dead (option 2) even though expected value-wise, option 1 is the wrong decision.
A good example of making a line of play based on fear came in GP Paris. Luis Scott-Vargas was playing late in the tournament and was crushing. He had a Mirran Crusader, Serum Raker, and one more creature out and had just ripped a Steel Hellkite for his turn. His opponent had nothing out, so Luis decided that he might as well play around Phyrexian Rebirth and not play the Hellkite since he figured he couldn’t lose to anything else. On the following turn, Luis’s opponent cast a Wurmcoil Engine that he had seen in a previous game, and eventually lost a drawn out game in which his opponent had to rip a couple more times to win.
If we look at the play, it makes sense at first. If Luis played the Hellkite and his opponent played Rebirth, he almost certainly loses. Luis would then have nothing while his opponent would have a 4/4 token. On the other hand, even a bomb as good as Wurmcoil looked beatable without his opponent ripping more. Luis could just play the Hellkite next turn, and probably win barring artifact removal. However, what this analysis doesn’t account for is the probability of Luis’s opponent having each spell. Luis had already seen a Wurmcoil Engine in a previous game, but had no reason to think his opponent had a Rebirth.
If we say that Luis would have had about a 20% chance of winning after a Rebirth, and about a 70% chance of winning post Wurmcoil (his opponent still had to get lucky to win), and then say that it was about five times more likely that he had Wurmcoil in hand than Rebirth (card he had already seen vs. rare he hadn’t), the math becomes interesting: 5 *.2 = 1 whereas 1 * .7 = .7. This means that Luis should have played around the Wurmcoil and not the Rebirth even though the Rebirth would have been a much bigger blowout. Even Luis is capable of falling victim to fear. He was so scared of completely getting blown out by Rebirth that he didn’t consider that he could lose to a much more likely card.
Another place where fear effects people’s decisions is mulliganing. People LOVE four-land three-spell hands. And why wouldn’t they? You know you’re hitting your first four land drops, AND you have three spells so you can’t get flooded to boot. Well, I don’t know what they taught you in Magic school, but four-land hands aren’t as good as you think. For most decks, a four land hand is flooded. Hands with four or five lands can usually play Magic in any given game, but you are usually one or two spells behind your opponent, and assuming you draw a reasonable ratio of lands and spells for the rest of the game, the extra lands in your hand will most likely be meaningless. Thus, keeping a four- or five-land hand is often wrong.
On the other side of the spectrum, there are one-land hands. Everybody knows those are terrible. If you miss on your first two draws, the game is just over. Why would you want to keep one of those? Well, first of all, you aren’t necessarily dead if you miss your first land drop, just behind. Fortunately, your hand is mono spells, so it is relatively easy to make up for lost time. Second of all, if you do hit, you are probably in great shape. You should be able to outspell your opponent very easily and win the vast majority of games that you hit the land.
Math time! Let’s we say that the average four-land hand simply gives you a 40% chance of winning since it’s mildly flooded, and the average five-land hand gives you about a 30% chance of winning, since you are down so many spells. Then let’s say that the average one-lander on the draw gives you about an 70% chance of winning if you hit (I know this seems high, but one-landers are just very very good when they hit), and about a 10% chance of winning when they miss (if you hit the next turn you might still be in o.k. shape, but you are pretty far behind if you miss). With a 24 land deck, your odds of hitting a land in the first two turns are: 1-((36/60)*(35/59))=.644ish that means your odds of winning with a one-lander are about: .7(the odds of winning if you do hit) *.644 (the odds of hitting) + .366 (the odds of missing) *.1 (the odds of winning if you miss) = .487.
That means, assuming you believe these estimations, one-land hands are even better than four-land hands on the draw. Obviously, it is very hand-dependent and these numbers were simply my guesses, but I do think it exemplifies the way fear can affect people. Keep more one-land hands and fewer safe, land-heavy hands.
Just as playing a “cool” deck can be tempting, playing a deck that sometimes doesn’t play Magic can be scary. In Extended a year or so back, Dredge had an absurd winning percentage. The deck wasn’t that popular, but still won more PTQs than any other deck, even Counterbalance. However, even as it put up great result after great result, it still never became that popular. The deck was frustrating to play, because almost every game you lost was due to facing a hate card and not having an answer and just being dead regardless of what you did. Playing through hate was possible, but it was often pretty miserable and certain hate cards like Leyline of the Void were just unbeatable if you didn’t have an immediate Chain of Vapor.
However, with the amount of hate people were playing, it is easy to see why Dredge was dominant. It won about 80% of its game ones, and then could fight through hate about 20% of the time. If you think about it, those numbers make Dredge’s winning percentage gross. People were averaging about four Dredge hate cards in their sideboard during that season meaning they have one about 65% of the time in post board games if they mulligan aggressively. Thus, they were 80% for game one and .65 (odds of opponent finding hate)*.2 (odds of winning if they have hate) + .35 (odds of opp not finding hate)*.8 (odds of winning against no hate)=41% for games two and three. That means an overall likelihood of winning a match is about 60% (.2 *.41*.41 = .033 (odds of winning games two and three) + .8 *.41 = .328 (odds of winning game one and two) + .8 * .59 *.41 = .236 (odds of winning games one and three)) =59.7%. Dredge was clearly the deck to play at the time, but people refused to play it for fear of losing miserable games to hate.
Emotions play a huge role in the decisions people make. However, it is important to analyze these decisions rationally and make sure your emotions aren’t effecting your decisions when they shouldn’t be. Realize that the expected value of a decision is not just based on how good or bad something is for you, but also the likelihood of it happening. Grow a pair and keep that one-lander, play Dredge when it’s the nut, and run Steel Hellkite into Phyrexian Rebirth. Man up and move up.
Bonus list: A bunch of people have asked me for an update on Standard Elves. I ended up not playing Elves because I thought the metagame would be too full of sweepers, but I feel like I did have a relatively interesting take:
This build is much more based on the [card ezuri, renegade leader]Ezuri[/card] + Copperhorn Scout combo. With 4 Lead the Stampede and 4 Zenith, it is very easy to find both Ezuri and Scout and enough mana to use it very quickly.
Also, I don’t think I’m going to do a report on Paris, but I did play what I think is a relatively tight list of Valakut, so I figured I’d throw that in too:
