Lately I’ve found myself intrigued by the resurgence, in popularity as well as success, of White-Blue Control in Modern. So I looked at some data with the goal to find out: What is the most common version of the deck? How do lists differ? And are there some differences more successful than others?
White-Blue Control is the ideal study subject in that regard. It’s reasonably popular, and many people agree on a lot of parameters. So it’s not completely like the Wild West out there, which, if it were, would make analysis difficult and probably useless. At the same time, it’s not as if every single list is a carbon copy of the one accepted truth either. There’s still a lot of relevant variation.
My search for a sufficiently large and comparable sample led me back to Grand Prix Bilbao. Not much in the way of metagame developments happened in the month since the event. It also remains the single largest Modern tournament of the year so far. Most importantly, I got access to almost all of Bilbao’s lists and was thus able to compare 65 white-blue main decks with another.
The Common White-Blue Control
For each card that appeared in any of Bilbao’s 65 White-Blue Control main decks, I worked out the median number of copies. If, let’s say, 14 of these players ran no Logic Knot, 26 of them ran one copy of Logic Knot, and 25 ran two copies, then the result for Logic Knot would be one.
And since this is exactly what happened in Bilbao, you can now find “1 x Logic Knot” below. The rest of the following list was created by the same process.
A Mean White-Blue
4 Celestial Colonnade 4 Field of Ruin 4 Flooded Strand 2 Glacial Fortress 2 Hallowed Fountain 6 Island 2 Plains 2 Snapcaster Mage 1 Vendilion Clique 3 Jace, the Mind Sculptor 2 Teferi, Hero of Dominaria 3 Cryptic Command 1 Detention Sphere 1 Hieroglyphic Illumination 1 Mana Leak 4 Path to Exile 1 Logic Knot 1 Negate 1 Spell Snare 1 Search for Azcanta/Azcanta, the Sunken Ruin 1 Supreme Verdict 4 Terminus 1 Timely Reinforcements 4 Opt
Some interesting stats: Path to Exile was the only card that appeared as a four-of in 100% of decks. The second Teferi was way more common than the third Jace—53 decks compared to 36—and the average planeswalker total was a little closer to four than to five. It also surprised me how widespread the use of Hieroglyphic Illumination has become. 43 lists included at least one copy, and 17 of them included even more.
Meanwhile, you may notice that the above adds up to 56 cards, or more specifically to 24 lands and 32 spells. Know that almost all decks contained 25 lands. There just was no majority that could agree on any single 25th. Ghost Quarter came closest, as an inclusion in 47.7% of lists.
The same applies to the missing spells. Closest to a majority—indeed as close as technically possible—came the second copy of Search for Azcanta, which appeared in 32 of 65 decks. Similarly, the second copy of Negate found its way into 27 of them. 25 players used some number of Absorb, and let’s not forget about the 25 who tied two Logic Knots.
So if you want to bring the above up to 60—maybe to include in a gauntlet against which you test your actual deck—then I suggest adding a Search for Azcanta, a Ghost Quarter, some counterspells, and likely one more removal spell. 40 players did agree to run a fifth creature removal for 1 white mana, but they were split, unevenly, between the choice of Oust and Condemn.
I’ll focus on differences such as these—and the difference they make—throughout the rest of the article. After all, the above is quite literally the most basic common denominator. It doesn’t have to be the lowest, at least not in a judgmental sense. It even could be pretty close to the optimal build. But further data might reveal some options for improvement.
Terminus or Bust?
43 players had four copies of the miracle in their main decks and another five had three. 15 players used zero. The average member of each group earned this many match points throughout the tournament:
With 12.46 points, the average Terminus player didn’t do much worse than the overall average of 13.49 points. But at 17.2 points, the players without Terminus outperformed the average noticeably. Frankly, this is the opposite of what I expected. I thought Terminus was so popular because it was superior to other options.
Then, further study revealed that the other options for mass removal weren’t in fact superior either. The average main deck contained 4.52 such effects, split between Terminus, Supreme Verdict, Settle the Wreckage, and the odd Wrath of God or Day of Judgment (because of Meddling Mage, presumably). But decks with more Supreme Verdict didn’t outperform the average, and neither did decks with Settle the Wreckage.
No, at an actual advantage were the configurations that simply featured less mass removal:
Granted, eight players isn’t a great sample, but it is the best explanation I could find. Loading up on mass removal—a number of players even ran six such cards—doesn’t seem to yield the most promising results.
Decks that included either of these spells fared better than the more common White-Blue Control build with its bigger focus on mass removal. The three players who used two copies of Condemn even earned an average 15.8 points, while the two double-Oust players made an average of 31.5 points, going 10-5 and 11-4. Not much in the way of statistics, but surely an interesting data point.
The benefits of additional spot removal didn’t apply universally though. Decks with Blessed Alliance performed minimally worse than those without, and adding a second Detention Sphere lowered the point output as well.
For another permanent answer, I’d suggest Runed Halo:
More Power! And Toughness
Another way to improve performance was to invite more creatures:
Running exactly three creatures was by far the most popular choice. But people who went higher ended up remarkably higher in the standings too.
I must admit, I was enamored with the idea of main deck Rest in Peace instead of Snapcaster Mage for a while. Now I would no longer dare to take that route. The only two players who ran that configuration in Bilbao earned an average of 7.5 points, but it isn’t their performance that deters me. Rather it’s the much more convincing sample above. It suggests that the Mage remains a crucial element that one discounts at one’s own peril.
Kerfuffle of the Spark
Not a fully fledged war, but there definitely was disagreement between Bilbao’s White-Blue players when it came to planeswalker selection. It already began with the number. One tiny minority ran fewer than four planewalkers, another tiny minority went with six. Both outperformed the average, but that’s neither here nor there. Between the most common choices—either four or five planeswalkers in total—no discernible difference existed.
Neither was there a notable difference in points between decks with two Jace, the Mind Sculptor and three. Teferi, Hero of Dominaria, on the other hand, told a different story. Players with fewer than two Teferis in their main decks won almost a full match more on average than their colleagues.
Decks with Elspeth, Sun’s Champion outperformed the average, whereas Gideon, Ally of Zendikar underperformed. But the two cards were present in only two main decks and one main deck, respectively, so that’s not conclusive. In contrast, 16 players put their hopes on Gideon of the Trials and had them dashed somewhat more reliably. The inclusion cost one match point compared to the overall average.
In conclusion, it’s probably a good idea to wait until War of the Spark to try out something more fancy than the established Jace/Teferi team-up.
Eight players chose to put Serum Visions into the slot more commonly occupied by Opt. They earned half a point more per player than the average White-Blue pilot. Though we should probably disregard a combination of a difference this small and a sample size to match. The same applies to a number of other cards I checked and discarded as not worth mentioning. A lot more research than text went into this article.
On a related note, you may find it suspicious that in basically every given example the minority outperformed the average. At first glance, you might think that all it shows is how outliers affect smaller samples disproportionately. But that can’t be right because outliers exist in the realm of abnormally good as well as abnormally bad results, and both affect smaller samples disproportionately in much the same way.
No, the easy explanation is: I set out in search of promising deviations from the norm and that was exactly what I found. Rest assured that I encountered plenty of deviations along the way that held no promise at all. For example, people with Sphinx’s Revelation in their main decks made 8 points on average. Main-decking Relic of Progenitus didn’t work out well either. And my new favorite misguided enterprise: transforming Thing in the Ice on the back of Path to Exile, Settle the Wreckage, and countermagic.
Let’s end on a minor counterexample. Hieroglyphic Illumination appeared most often as a one-of, and this majority also had the best results, earning 14.28 match points per player. In contrast, decks with more copies and zero copies both came in below White-Blue’s overall average of 13.49 match points.
Until next time—
Oh, the article’s title made you expect a list to copy and paste? Well, here’s what I’d try out, based on the numbers I got. It’s a mash-up of all the aspects the successful decks had in common, more or less without regard for interconnections and inter-dependencies. So it is a little experimental. As always, I trust that the comments section will point out obvious flaws.
An Experimental White-Blue
4 Celestial Colonnade 4 Field of Ruin 4 Flooded Strand 1 Ghost Quarter 2 Glacial Fortress 2 Hallowed Fountain 6 Island 2 Plains 4 Snapcaster Mage 1 Vendilion Clique 3 Jace, the Mind Sculptor 1 Teferi, Hero of Dominaria 1 Ancestral Vision 3 Cryptic Command 1 Detention Sphere 1 Hieroglyphic Illumination 1 Mana Leak 1 Oust 4 Path to Exile 1 Logic Knot 2 Negate 1 Runed Halo 1 Spell Snare 1 Search for Azcanta/Azcanta, the Sunken Ruin 1 Settle the Wreckage 2 Supreme Verdict 1 Timely Reinforcements 4 Opt