---

As I start a lot of my articles, I begin today with a very typical story seen in the competitive annals of Magic:

A player uncertain on what deck to pilot at the next tournament looks online for some lists. Naturally, he immediately goes to recent, highly publicized tournament top eight decks. The winning deck—tautologically—looks like a winner, so he opts for that in his next tournament. He goes 2-2 and wonders what the heck went wrong.

So what did go wrong? Obviously, this conversation could go a number of ways. But being that this is a statistical study, my big problem is that he looked at a very narrow set of results. There is a wide variety of explanations why that deck won on that particular day. Maybe he hit all the right matchups. Perhaps his topdecks were flowing as they would for Craig Jones. It is possible he is just a really amazing player who can make up for the shortcomings of his list. Or he actually has the best deck out there. The problem is that with such a small sample size, it is impossible to tell which mechanism (or which combination of mechanisms) is at work here.

What if there was a way to cut through the misinformation? Well, there is—you just need to look at large samples, large enough where you need a computer to do the calculations for you. Each of the problems will still be present, but the larger sample size minimizes their impact. That allows us to see which decks are performing well, not who is the best player, or who is getting lucky, or who sees all the right matchups.

That's where I come in. I've run the data and tested their significance, and I am here to present to you the results. Let's get to them:

4 Color Seas: 60.7%
Bant: 41.3%**
Boros Bushwhacker: 52.8%
Cascade: 42.7%
Emeria’s Pledge: 53.3%
Esper Control: 38.8%**
Goblins: 48.2%
Grixis Control: 50.3%
Jund: 54.9%***
Knightfall: 52.8%
Naya: 54.1%
Naya Lightsaber: 52.0%
Nissa’s Monument: 46.8%
Red Deck Wins: 52.2%
Spreading Seas: 52.3%
Standard Dredge: 37.8%**
Turbo Fog: 37.6%***
U/W Control: 53.8%
UWR Control: 55.0%*
Vampires: 45.5%*
Valakut Ramp: 59.4%*
White Weenie: 45.6% 

*Statistic is significant at 90%.
**Statistic is significant at 95%.
***Statistic is significant at 99%.
Decks are tested against a 50.0% hypothesis.
To be eligible, a deck must have at least 50 observed matches.

If it hasn't become abundantly clear by now, the deck I was referring to in my story is Vampires. Vampires won the Star City $5000 tournament a few weekends ago. Undoubtedly, this will cause a surge in its popularity. However, the basis for this is entirely unfounded. $5K win be dammed, Vampires’ win percentage actually declined since I posted last week's results, and it is now below 50.0% with 90% confidence. So while I appreciate looking at top eight decks for guidance, you absolutely have to delve deeper if you want a true understanding of the metagame.

In other news, the number of statistically significant results doubled from four to eight over the week. (Increases are to be expected—the more data there is, the smaller the margins of error become, which in turn makes rejecting null hypotheses more likely.) Valakut Ramp is a part of this group, making its debut with a 59.4%.

4 Color Seas: 60.7%
(Low: 47.6%; High: 73.8%)

Still no reason to trust this statistic—4 Color Seas has gotten incredibly lucky in its pairings, meeting Jund an inordinate number of times. With Jund a good matchup, Seas’ win percentage is substantially higher than it should be. As such, I can’t recommend it for tournament play.

Bant: 41.3%**
(Low: 33.7%; High: 48.9%)

Hey, at least it’s decent in Extended. Bant’s matchups aren’t terrible in general, but its number don’t do especially well anywhere either. A 14-22 record against Jund holds Bant back as well. 

Boros Bushwhacker: 52.8%
(Low: 47.6%; High: 58.0%)

Still a solid alternative choice to other creature based decks like Jund and Knightfall, even if its matchups against those two decks aren’t spectacular (42-52 and 10-18, respectively).

Cascade: 42.7%
(Low: 33.1%; High: 52.4%)

Although not “bad” at a statistically significant level, Cascade’s numbers have dwindled to virtually nothing in the past few weeks, likely as players have discovered it doesn’t have much game. Jund has all the cascade spells you need; running any more or for a different purpose other than express beatdowns is not a good idea. 

Emeria’s Pledge: 53.3%
(Low: 45.3%; High: 61.2%)

Faring much better than its Extended Emeria-centric counterpart, which will clock in at a lowly 40.7% in a future article.

Esper Control: 38.8%**
(Low: 28.2%; High: 49.5%)

Some combinations of colors are good at playing control. Evidently, this isn’t one of them. Pass.

Goblins: 48.2%
(Low: 39.0%; High: 57.5%)

Goblins suffer from the presence of Jund in Standard, as the quick onslaught of little red men does not stand up to Jund’s larger creatures. In a previous article of mine, I issued a challenge to express this problem in haiku form. Our winner comes from reader CB [begin soft background drumming]: 

Small and fast red men
When faced by the Jund menace
Are quickly outclassed

Grixis Control: 50.3%
(Low: 45.2%; High: 55.3%)

May get better if foolish people continue to flood tournaments with Vampires decklists—its 8-16 record against the Edward Cullens of the world just barely falls outside of the 90% confidence range.

Jund: 54.9%***
(Low: 52.6%; High: 57.3%)

What more is there to say about this deck? Perhaps I should stop using the over/under 50% as the metric to testing and start pitting everything against Jund. In other words, should you even bother playing deck X instead of Jund? There are very few decks that can withstand that test, which means you should consider selling out and joining the gigantic group of Jund players.

Knightfall: 52.8%
(Low: 48.1%; High: 57.5%)

Knightfall is the third most played deck in Standard (behind Jund and Red Deck Wins, see the list below), yet it receives substantially less internet chatter than even decks ranking below it. Go figure.

Naya: 54.1%
(Low: 42.7%; High: 65.4%)

For what it’s worth, Naya racked up the fifth best observed win percentage.

Naya Lightsaber: 52.0%
(Low: 45.1%; High: 58.8%)

Trust me when I say it gives me absolutely no pleasure to inform Michael J. Flores that Naya Lightsaber has a lower win percentage than Naya does. That said, the statistician in me feels that it is only fair to point out that this claim cannot be made at a statistically significant level, which is a shame.

Nissa’s Monument: 46.8%
(Low: 40.0%; High: 53.7%)

Sorry, elf fans—this one still isn’t coming out well for you.

Red Deck Wins: 52.2%
(Low: 47.7%; High: 56.7%)

Won the TCGplayer $5K over the weekend. I’m guessing its player successfully avoided the UWR matchup (6-19) and hit a lot of Grixis Control (24-7) instead.

Spread ‘Em: 52.3%
(Low: 41.8%; High: 62.9%)

Despite being designed to beat Jund, Spread ‘Em doesn’t necessarily do that (13-11), but it does hold its own against the rest of the field (32-30).

Standard Dredge: 37.8%**
(Low: 26.4%; High: 49.2%)

If Hedron Crabs or filling your graveyard with creatures really turns you on, I’d recommend switching over to Extended so you can play Dredge of Living End instead. No, seriously—this is a horrendously bad deck.

Turbo Fog: 37.6%***
(Low: 30.4%; High: 44.8%)

The only remaining question is why people are still playing Turbo Fog. For the last time, it doesn’t work! A 37.6% observed statistic is enough to make even the casual player cringe. 

U/W Control: 53.8%
(Low: 42.7%; High: 64.9%)

Still greatly overshadowed by the red splash. Unfortunately, there isn’t (and won’t be, since this is the end of the season) enough data to figure out which one is better.

UWR Control: 55.0%*
(Low: 49.9%; High: 60.1%)

Just narrowly missed out on 95% statistical significance. There is little doubt in my mind that UWR Control is the premier control deck of the format, yet Grixis Control is slightly more popular.

Vampires: 45.5%*
(Low: 40.3%; High: 50.6%)

Vampires still aren’t cutting it against Jund (30.7%***), and that is a one-way ticket to the “bad” column. What’s telling about the SCG $5K Open isn’t the fact that Vampires took first—it’s that there were fifteen other guys with the same deck in that room, and the next best finish was 50^th. And if you want to go down a bit further, the third best Vampire finish was 98^th. I can only cringe in pain at that result.

Valakut Ramp: 59.4%*
(Low: 49.7%; High: 69.2%)
[Debut!]

How Valakuts are still available for under $3 is a bit of a miracle: Mountain-based combo-ish decks are taking over both Standard and Extended. 

White Weenie: 45.6%
(Low: 36.0%; High: 55.3%)

In a previous article, I mentioned that White Weenie (without any support from other colors) was last successful in 2000 with Kai Budde winning Pro Tour Chicago. Someone wisely corrected me, noting that Budde splashed green for the Wax half of Wax/Wane. Consequently, I have absolutely no idea when the last time a purely white aggro deck was any good.

 

As a bonus, I wanted to include the share of the metagame that each of these decks have racked up over the course of the last few weeks. I recommend you take all of it with a grain of salt—metagames shift fairly frequently even if illogically, as noted in the Vampires anecdote—but it should nevertheless give you a feel for what Standard is like before the rotation.

Here is the list, in order of the most popular:

Jund: 25.06%
Red Deck Wins: 6.92%
Knightfall: 6.35%
Grixis Control: 5.52%
UWR Control: 5.45%
Vampires: 5.33%
Boros Bushwhacker: 5.20%
Nissa’s Monument: 3.01%
Naya Lightsaber: 3.00%
Turbo Fog: 2.73%
Bant: 2.45%
Emeria’s Pledge: 2.23%
Goblins: 1.65%
White Weenie: 1.51%
Valakut Ramp: 1.48%
Spreading Seas: 1.26%
Esper Control: 1.25%
U/W Control: 1.15%
Naya: 1.09%
Standard Dredge: 1.09%
4 Color Seas: 0.82%
Others: 13.94%

As you can see, no one deck other than Jund takes up a large percentage of the metagame. So the next time you talk about “tuning your deck to beat UWR Control” or some such nonsense, you need to realize the odds of actually playing that deck in a tournament are fairly low. In fact, the likelihood that you will play against some rogue deck is more than twice as great, Jund again excepted. In that light, the concept of tuning seems pretty ridiculous.

Just some food for thought. I hope you have a better idea of where Standard is before the switch over to Worldwake in a few short days.

William Spaniel
williamspaniel@gmail.com