A while back, I surveyed 1000 players’ results from Zendikar limited games and tracked how frequently individual commons contributed to wins and losses. (You can find the complete archive here.) Well, I did the same thing for Magic 2010, and I am here today to generalize the results of the past set into lessons for Magic 2011. Is card advantage key? Does removal rule the world? Did the most broken Magic 2010 card make it into Magic 2011? The answers are inside…

 If you are unfamiliar with my methodology, I have put it into the appendix at the bottom of this article. You should read it before going through the rest of the article. If you have read it (or just finished), let’s go straight to what is important.

Removal is overrated…
If you think removal is the key to victory in every limited format, think again. Excommunicate went 35-35, or 50%. Pacifism fared only slightly better at 79-78, or 50.3%. Ice Cage was just as average, going 15-15, or 50%. Doom Blade could only muster a 74-69 record, or 51.7%. 

That said, Assassinate was the worst offender by far. Its 43-57 mark left it at only 43.0%, which is statistically significant to 95% confidence.

The bottom line: Don’t think that a deck full of removal is going to get you very far. Some removal spells—namely, Assassinate—might even hurt you.

…but burn spells aren’t.
See Consume Spirit (24-14, 63.2%), Earthquake (16-6, 72.7%*), Fireball (45-29, 60.8%), and Lightning Bolt (78-43, 64.5%). The key difference here seems to be dealing damage to players. If you can hit a player, you can topdeck wins from certain defeat on the next turn. You can do that with these spells; you can’t do that with Excommunicate, Pacifism, Ice Cage, Doom Blade, or Assassinate. Possible card advantage doesn’t seem like enough, with the creature-only Pyroclasm only hitting 27-23, or 54.0%. 

Corrupt—Magic 2011’s answer to Consume Spirit—should fare better than black’s regular allotment of removal spells, but I doubt it will be able to keep pace with Consume Spirit. Unless you are running an overwhelming number of Swamps as your mana base, Corrupt is going to deal less damage. And if it is dealing less damage, then you will not be able to take out as big of creatures or finish off your opponent at as high of life totals. Nonetheless, I still expect it to perform better than an average spell.

Overrun is ridiculously good—thankfully, it is gone.
Players cast Overrun 41 times in this study and won 35 of those games (most of them right then and there), good for an 85.4% finish. That 85.4% was the highest observed winning percentage among all cards with more than four appearances. The card is undeniably broken. It’s hard to even make a case why you should draft any bomb rare over it. Even something like Baneslayer Angel loses to Doom Blade. For Overrun to lose to a removal spell, your opponent pretty much needs to be able to destroy all of your creatures in response. Anything short of that, and they will probably be dead. 

Actually, the only compelling reason to draft Baneslayer Angel over Overrun is purely economic. If you see an Overrun, take it. You are playing green.

On another note, why was this card still an uncommon in Magic 2010? Wizards printed Serra Angel at uncommon for years before moving it to rare, supposedly because it had a rare feel to it. How does Overrun not feel like a rare? The name is epic and the ability is completely game shattering. A card that wins 85% of its games and only costs five mana should not be seen as often as it is at uncommon. Perhaps Wizards finally caught on to this, decided that they could not make it a rare for whatever reason, and then did not print it altogether. Regardless of the reason, however, I am glad to not be seeing Overrun in any Magic 2011 games.

Card advantage is important…
Conventional wisdom says that card advantage is extremely important in Core Set draft formats, as it is (supposedly) harder to outplay your opponent to gain card advantage in the (supposedly) more straightforward Core Set limited formats. While I have no basis of comparison to verify that claim, drawing cards certainly looks good: Divination went 39-14, good for 73.6%. Perhaps more impressive, it was one of only four cards to be statistically significant to 99% confidence. (The other three were Overrun, Warpath Ghoul, and Safe Passage.)

How that applies to Foresee and Preordain remains to be seen. It seems that Divination's power comes mostly from digging deeper into your deck and not so much from recovering from mana screw. Since Foresee can go through up to six cards (15% of your deck!), I'd imagine it will do well. Meanwhile, I can't imagine Preordain hurting a player, but I also don't see it burning up the charts like Divination. So we'll see on that one. 

…but it isn’t worth paying your life for.
If drawing two cards for three mana is really good, then surely Sign in Blood must be as well. After all, paying two life is not that big of a deal, especially if it trims one mana off of the casting cost. Not only will that save you some tempo, but the ability to play Sign in Blood while stuck on two lands is a luxury that Divination will not afford you. Oh, and you also get finish your opponent off when they are sitting at one or two life. That is what separated Lightning Bolt from Doom Blade, right?

That all sounds good in theory, but it ends up not being true in practice. Sign in Blood finished at 36-36, or 50%. Apparently drawing cards is important but not worth paying your life for. There seems to be a couple of things snagging Sign. First, the difference between two and three mana does not matter all that much. You rarely want to play Sign in Blood on the second turn or Divination on the third—you mostly cast these spells after you have emptied your hand of threats. This does not happen until the sixth or seventh turn or so, at which point there is very little difference between tapping two lands and three. Sign in Blood is better at fixing mana screw, yes, but you will not use it for that purpose all that often.

Finally, while you can Sign in Blood to finish off a vulnerable opponent, you will almost never encounter such a situation. If you didn’t cast Sign in Blood on turn two because you had nothing else to do, and you didn’t cast it on turn six or seven because you had not drawn it yet, then you will play it as soon as you top deck it. You just do not see players holding Sign in Blood for five or six turns waiting to target their opponent. In fact, you are almost always better served by throwing it into your graveyard to get two fresh cards off the top of your library. So if Sign in Blood is going to be a finisher, it will be that because you just drew it and your opponent was at two life or less. Otherwise, it is just drawing cards, and Divination is a lot better.

Safe Passage is a trap—for you.
The general consensus is that Fog effects are bad in limited. Indeed, there is usually exactly one good use for them—when your opponent commits to an all-in attack that you cannot survive, you Fog, and you can counterstrike for the win on your next turn. Anything short of that either leaves you facing impending death, down a card, or both. 

Safe Passage seemed to be an exception. Here, you could also counter important spells like Lightning Bolt, Pyroclasm, or Earthquake. You could also trap your opponent into attacking with a bunch of creatures, blocking them fiendishly, playing Safe Passage, and watching most the opposing army die while all of yours survive. Failing that, you could actually attack into your opponent, have them make some decent blocks in their favor, and then turn the tables with a Safe Passage. And, of course, it did everything that Fog could do.

Luis Scott-Vargas even said the following about Safe Passage:

Safe Passage is insane, and I didn’t give it the respect it deserved initially. It destroys Fireball, Earthquake, and Overrun, and is at the worst a solid 1 for 1, although usually better. I would play 2 or 3 in most decks without thinking about it, and would pick it over most commons that don’t fly or Pacify.

After all of that, how did Safe Passage do? It finished at 21-42, or 33.3%***.

33.3%***.

A list of white cards that performed better (though not necessarily at a statistically significant level):

Armored Ascension (57.4%)
Baneslayer Angel (68.2%)
Blinding Mage (54.3%)
Captain of the Watch (60.7%)
Divine Verdict (54.8%)
Elite Vanguard (41.3%)
Excommunicate (50.0%)
Griffin Sentinel (51.3%)
Guardian Seraph (46.2%)
Harm’s Way (47.5%)
Lightweilder Paladin (55.0%)
Pacifism (50.3%)
Palace Guard (45.6%)
Razorfoot Griffin (49.0%)
Rhox Pikemaster (49.2%)
Serra Angel (53.1%)
Siege Mastodon (57.1%)
Silvercoat Lion (49.4%)
Solemn Offering (68.0%)
Soul Warden (50.0%)
Stormfront Pegasus (47.6%)
Undead Slayer (53.6%)
Veteran Armorsmith (51.9%)
Veteran Swordsmith (44.6%)
White Knight (56.9%) 

So what happened here? Well, my intuition is that keeping three mana untapped at all times does not really help you win games. If you look over the list of white spells above, you will see that a vast majority of them are creatures. So if you are playing white, you are playing creatures; you are not drawing a card and passing the turn. Yet that is exactly what Safe Passage requires you to do. Yes, it could be potentially game breaking against an Earthquake or Overrun (or for Magic 2011 purposes, Destructive Force). Yes, it will do some wacky things if you ever get into complex board positions. Yes, it will buy you an extra turn and an extra top deck when you are on the verge of death. But will it help you win games more than it will hurt you? It certainly does not look like it.

That's all for now. I'm thinking about doing a running track of Magic 2011 cards, so maybe you will hear from me soon.

William Spaniel

Appendix: Methodology
I assume that skill, luck, and the quality of a player's deck determine who wins any particular confrontation. While undoubtedly skill matters, this study is focused on the luck and card quality factors. Players actually have a great deal of control over both of these, as a poorly-constructed deck will win less often than a well-constructed one. From this, we can conclude that some cards contribute to wins more frequently than others. If an average card ever reached play in a game, we would expect its controller to have only won that game around 50% of the time. But if a truly exceptional card reached play, we would expect its controller to have won upwards of 70% of the time.

Watching replays of Magic 2010 premiere sealed tournaments, I recorded the results of more than a thousand players. Every time a card hit play, I would record it as either a win or a loss, depending on what ultimately happened in that game. If the card reached play multiple times (perhaps because of a recursion effect, like Disentomb), it only counted once. But if a player cast multiples of a single card, I counted that card multiple times.

Such a large number of observations were necessary to remove the play skill bias that would have shown up in a smaller-n study. It also shrinks the margins of error, allowing for better hypothesis testing, which I ran at 90%, 95%, and 99% confidence.

For those of you unfamiliar with hypothesis testing, I go over the intuition behind it in this video. But here is a brief explanation of what everything means in words:

90% Confidence: When I say we can be 90% confident that a card positively contributes to victories, it means that there was only a 10% chance that the card has no impact and the data came back so eschewed based on pure luck. While the odds of being wrong here are only 1/10, we should be very skeptical of these results as statisticians. Generally, it is only a good idea to accept these results if we have a good theory behind them. For example, I would accept Lightning Bolt—a quality removal spell as being true—but I would cast doubt on whether Drudge Skeletons was actually affecting things.

95% Confidence: This is the gold standard of statistics. When a card meets 95% confidence, the likelihood the card is merely average but we got this extreme of data back is only 1/20. At this point, it is a good idea to start thinking of theories to justify the results if you do not have one already.

99% Confidence: While rare, a 99% confidence virtually guarantees that a given card has an impact on the match—there is only a 1/100 chance that this result is wrong. You should pay careful attention to these.

Just because a card does not show up a significant does not mean you should not care about the results. But you should not treat them as gospel, either. The best analogy I can draw is to that of a baseball team. It is possible that your star hitter goes through a minor slump at the beginning of the season and an average player goes on a torrid streak at the same time. That does not mean the average player is better than the star; it just means he was better during that period of observation. So don’t be surprised if the study ranks an average card lower than one you perceive as a top-pick.

Additionally, just because a card is sub-50% does not mean you should automatically stop playing it in all of your decks. Going back to the baseball analogy, a team cannot field nine players with batting averages all over .300. But it can maximize its performance by putting its best players in the lineup. So if you need to play a terrible card to have enough white cards to justify running Baneslayer Angel, go right ahead.