When I last looked at Draft data, I focused on the shape of successful Draft decks. The obvious follow-up question was: How did these decks end up on top?

Specifically, I found that Golgari decks went 3-0 more often than decks of any other guild. But was this because of Golgari’s intrinsic strength or because Golgari was horribly underdrafted? Either way, the commonly held belief that Golgari was weak should probably become less common and drafting Golgari should become more common. More data can provide a better understanding than just this simplistic view.

In this case, more data means all deck lists from the first Booster Draft at Grand Prix Warsaw. Whereas previously I only looked at the 3-0 decks from the top 29 Draft tables, now I got access to all 369 lists from all 46 pods. This is almost one booster box case full of deck lists. Thanks to the folks who brought just such a box to Liverpool for me!

The 3-0 Decks

Golgari still accounted for most 3-0s among all 46 Draft pods. 13 players whose decks were predominantly black and green went undefeated. The Boros Legion went undefeated 12 times.

Dimir was the second most successful archetype at the top 29 Draft tables, but it fell behind when I looked at all Draft tables. Only nine Dimir agents went 3-0. The Izzet League earned seven trophies, and this included what could be considered a freak accident in an epic experiment and thus a total flavor win: At a table with nine drafters, two Izzet mages went 3-0!

Selesnya, finally, continued to disappoint. At 46 tables, only three decks based in green and white were piloted to a perfect record.

Points Per Drafter Per Guild Per Table

I calculated how many match points players earned relative to how many people drafted their guild. We expect drafters who didn’t have to share their cards with anyone at the table to get more points than those who did. And except for some outliers resulting from too small of a sample, this is exactly what the data shows.

For reference: the average player will get 4.5 match points out of a Draft’s three rounds. This is our baseline. If players of a guild earned more points than that, the guild did above average, or possibly, it’s an indication that it was underdrafted.

Note that throughout the article I will use “X players drafted a guild” and variations thereof to mean: “X players ended up with a deck clearly based in this guild.” Obviously, someone may have started drafting Boros before settling on something else and may have taken some cards that Boros drafters were then missing. While relevant to a degree, there’s no way to account for all variables.

A Guild of One

  • The 17 drafters who were alone at their table in drafting Boros earned six match points on average. Six of them got the maximum nine points—that is, they went 3-0. Being the only Boros drafter at a table was the best position to find oneself in, but the next best thing came close…
  • Being the only Golgari drafter at a table resulted in an average of 5.96 points, and eight of twenty-four people won three matches from such a starting point.
  • Dimir was so popular that only ten drafters competed with no other Dimir players at their table. They made 4.3 points, which is the lowest average for drafters who were alone in their guild. Stranger still, none of them went 3-0.
  • Izzet drafters were alone at 13 tables. They averaged 4.38 points and two of them went 3-0. Inexplicably, Dimir and Izzet were more successful when two people drafted the guild. It may be that the packs at some tables were uncharacteristically weak on blue, but most likely it’s just an effect of the low sample size. Indeed, a table was more than twice as likely to feature two drafters in Dimir than to have one, and the situation for Izzet didn’t look much different.
  • There’s no problem with a low sample size when it comes to Selesnya. 24 tables had a single Selesnya drafter. These 24 players earned an average 4.42 points, and none of them went undefeated.

The difference in match points is quite telling. It indicates that Boros was underdrafted at 17 tables at Grand Prix Warsaw, but Golgari was drafted even less. 24 tables had only one Golgari drafter. At least another eight tables had no one focused on black and green at all! Even Selesnya was picked more often, with way worse results.

That Rhizome Lurcher still managed to be one of the most common creatures in the 3-0 decks is astonishing. Or maybe the reverse is true: maybe the achievement was predicated solely on how underdrafted Golgari was? Let’s look at further numbers to find out!

Dynamic Draft Duos

  • 30 Selesnya drafters sat at one of 15 tables where one other player drafted Selesnya. They made an average of 4.47 points. Three of them went 3-0.
  • 44 drafters shared Izzet with one other player at their table. They earned 4.52 points on average. Four of them went 3-0.
  • 50 drafters shared Dimir with one other player at their table. They averaged 4.52 points. Seven of them went 3-0.
  • One could find two Golgari drafters at just eight tables. Each of these 16 players earned 4.63 points, and three of them even went undefeated.
  • 40 drafters shared Boros with one other player at their table. They averaged 4.86 points, and four of them went 3-0.

The naturally larger player base in this tier explain why all match points move closer to the overall average of 4.5. Boros continued to do exceptionally well, while the blue guilds again didn’t do appreciably better than average. Unfortunately, the sample size for Golgari is the lowest here, but all data suggests that its success can’t be attributed to being underdrafted alone.

Three’s Bad Company

  • At one table, three people drafted Selesnya. One went 2-1, two went 1-2.
  • Eight tables had three Izzet drafters. These 24 players averaged 3.63 points. One of them went 3-0.
  • Nine tables had three Dimir drafters. These 27 players averaged 3.67 points. Two of them went 3-0.
  • Six tables had three Boros drafters. These 18 players averaged 3.83 points. Two of them went 3-0.
  • Two tables featured three Golgari drafters. Strangely, at each table one of them went 3-0 and one went 2-1, while only their third went 1-2 or 0-3. This makes for an average of 5.5 match points, which is way higher than what we can consider realistic.

Discounting the Golgari outlier, this shows that none of the five guilds are so overpowered as to support three drafters. One should definitely strive to avoid ending up in a guild that’s drafted by three people at a table. While Golgari and Boros left the best impression at Warsaw and while Selesnya seemed somewhat sad, overall the format’s color balance looked fine. It was mostly the players who introduced an imbalance!

It’s another good sign that the blue guilds didn’t turn out to be as strong in practice as many experts had theorized them to be. Even after months of drafting, there’s still new insight—or possibly metagame developments. For instance, Martin Jůza noted that everyone used to be stoked for Boros initially, but that the fire had cooled down in favor of blue cards since then. This sounds exactly like what happened in Warsaw, where Boros was drafted way less than the blue guilds and was vastly more successful.

Finally, the format is great because you can go undefeated—admittedly 2-0-1— with a deck like the following. If you want to see an example of something that I didn’t sort into any of the five guilds, look no further!

Dimir/Golgari Splash Izzet/Selesnya

Jan van der Vegt

Jan van der Vegt is definitely my kind of drafter. If you get a chance to watch him play Limited, do!

Who’s Beating Whom?

Finally, an analysis of Guilds of Ravnica Booster Draft wouldn’t be complete without a breakdown of how each guild matched up against every other guild.

The data was even self-correcting in one regard because of the very nature of Draft pairings. That is, players who had to share the cards of their guild with more drafters also played more mirror matches on average, so their results have lower representation here. Meanwhile, the theoretically best decks of each guild didn’t face any mirror matches at all.

Then again, none of the following should be taken too seriously, because I didn’t actually control for unfair matchups resulting from over- or under-drafting. For example, Julien Berteaux, who was the only Boros drafter at his table, beat up three blue drafters in a row, of which there were five at his table. Is it any wonder that he went 6-0 in games against them?

So yeah, seems like Golgari is officially the best guild now. All hail Vraska of House Golgari, the Unseen, Queen of the Undercity, First of Her Name, Protector of the Swarm, Seeker of Relics, Schemer of Gorgons!