Decks built around Command the Dreadhorde and the green explore package are among the most popular in current Standard. More than 10% of players at Grand Prix Kansas City decided to run this strategy. This makes for a decent sample to analyze. Lists have a lot in common: Wildgrowth Walker, Merfolk Branchwalker, Jadelight Ranger, a bunch of planeswalkers, and the eponymous sorcery itself. This makes direct comparisons possible.
But within the general framework a lot of variation exists and persists. Months of literal and figurative exploration haven’t yielded a conclusive consensus yet. This makes comparative analysis a promising endeavor.
The archetype checks the three most important requirements, so let’s get to it.
Command the Dreadhorde By the Numbers
How Many Does It Take for a Group to Become a Horde?
The differences already begin with the question of how many copies of Command the Dreadhorde to run. Throughout history, decks named after a key card have usually included four copies of said card: Necro, Survival, Tinker, High Tide, Slide, Company, Steel Leaf Stompy, Phoenix, and so on and on. At the same time, there have always been exceptions too, for example Goblin Bidding or Seasons Past Control.
In which category fall today’s green Dreadhorde decks? A majority of players went with four copies of Command the Dreadhorde. But the similarities to Patriarch’s Bidding and Seasons Past are neither superficial nor, admittedly, coincidental.
The wisdom of crowds has failed. Players with the full playset in their main deck made an average of 13.7 match points. Those with three copies left the tournament with 16.75 points. That amounts to more than one additional match win per player.
It is tempting to overload on Command the Dreadhorde in the face of Thought Erasure running wild. But the data is pretty clear. You should have three Commands in your main deck.
How Many Colors Are Too Many?
I was surprised to learn just how many players stretched their mana base to incorporate all five colors. I guess it can’t be that much of a stretch. If you splash Teferi, Time Raveler and Sorin, Vengeful Bloodlord off of Interplanar Beacon anyway, you might as well add the self-styled Dragon-God-Pharaoh himself at little extra cost.
But is it worth it?
The answer is a resounding no. The average Five-Color Dreadhorde player finished on about half as many points as the average Sultai or Four-Color player. Don’t go there.
Whether it’s better to stay Sultai or to include white, is less clear. With 17.2 match points compared to 15.9, the results favor Four-Color Dreadhorde, albeit at a rate that isn’t quite conclusive.
Are These the Druids You are Looking for?
The big advantage of three colors is that it allows the use of four Llanowar Elves. Not even all Sultai players went that way. But a 2.4-point lead of one group over the other suggests that Llanowar Elves are better than no Llanowar Elves when your land base can support them.
Note that this proved true even at Grand Prix Kansas City, an event where more than 18% of decks featured Goblin Chainwhirler.
The grouping here looks a little odd. Know though that lists with one copy and four copies were very rare. What stood out is the performance of main decks with exactly two copies of Paradise Druid.
Value or Won’t You?
The trio of Wildgrowth Walker, Merfolk Branchwalker, and Jadelight Ranger proved an almost universal inclusion. Indeed, 85 of 86 lists contained the maximum of 12 such creatures. Two players even explored further options. A single Ixalli’s Diviner ended up on 31 points; a playset of Seekers’ Squire ended up on three match points. Unfortunately neither tells us much of anything.
Decks differed more noticeably with regard to the bigger value creatures. Most notably, 31% of players had Hydroid Krasis in their lineup. The card had once been the reason for midrange players to dip into blue in the first place. Recently, there’s been a bit of a Hydroid Crisis thanks to Narset, Parter of Veils.
Hydroid Krasis remains quite successful in spite of everything. The average Dreadhorde player with four copies in their main deck won more than one additional match compared to the average Dreadhorde player with fewer than four copies.
Another spicy data point concerns Massacre Girl. You might have 99 problems with opposing creatures, but just one girl makes none.
Ten players with two copies of Massacre Girl in their main deck don’t really allow any inference of causality. These decks were so much more successful than their less girly counterparts, however, so you should seriously consider the second copy.
The 14 players with Hostage Taker in their main deck, most of them running two copies, earned 2.3 match points more on average than those without. The 37 players with Trostani Discordant in their main deck, almost always a one-of, earned 1.3 match points more on average than those without.
How Many Super Cool Friends Do You Invite to the Party, and Who?
A Magic duel represents the battle between two planeswalkers, or three, or four, or twenty. The war has given us many options, but we should not go overboard.
In case you’re wondering how those more successful Dreadhorde players made room for all the cards mentioned above … Here’s the answer. The eight players who ran seven or eight planeswalkers averaged 20.25 points. The 19 players who ran nine or ten planeswalkers averaged 15.6 points. The 14 players who ran 11 or 12 planeswalkers averaged 18 points. The differences between these numbers don’t show a clear trend, but the numbers for the people who went beyond 12 planeswalkers do. The 13th planeswalker already seems to be one too many.
Tamiyo, Collector of Tales was number one on the invite list. The average main deck contained 3.8 copies. Every party can benefit from someone who’s got stories to share, but at some point it becomes repetitive. The minority on three copies was vastly more successful than the majority on four, earning upward of five extra match points per player.
Teferi, Time Raveler was next in line, showing up as a four-of in 52 main decks and as a three-of in another seven. His presence or absence barely made a difference for points though. The same was true for Sorin, Vengeful Bloodlord. The 26 main decks with two copies of Sorin didn’t do appreciably better or worse than the 26 main decks with one copy or the 32 main decks with none.
Almost everyone had some number of Vraska, Golgari Queen …
But, unlike in other card games, a deck of Magic appears to need only one or two queens. Nissa, Who Shakes the World experienced similarly diminishing returns:
But the reclusive planeswalker remained a much rarer party guest overall. This makes the data on the differences between the Nissa versions almost as shaky as the world under the influence of Nissa herself. At 17.4 points per player and 13.5 respectively, the difference between the 30 versions with and the 56 versions without Nissa looks rather solid though.
30 players also included Teferi, Hero of Dominaria in their main deck. The ten players with two copies earned an average of 16.5 match points, the 20 players with a single copy ended up on 15.8, while the 56 players without had to contend themselves with 14.2 points. That’s too small of a gap based on too small samples to derive meaningful conclusions.
The same goes for all the rest of planeswalkers and, in fact, cards. So we’ve reached the natural end of this particular exercise.
Bonus Section: Bant or Simic Ramp?
Ramp players won 64.4% of their matches at Grand Prix Kansas City as well as 57.9% of their matches after the cut at Grand Prix Taipei. They only won 40% of their matches at Mythic Championship III, but that tournament had a very select and very short attendance list and thus not as much play to it. I’m inclined to trust in the results of 821 GP matches rather than 55 MC matches.
At the Mythic Championship, Simic Ramp went 6-4, whereas Bant Ramp went 16-29. But as with the overall performance, the Grand Prix results also differed in the specifics. Here are the numbers for all 18 Ramp players at GP Kansas City and all 39 Ramp players in Day 2 of GP Taipei:
30.5 and 22.2 points mean that the average Bant Ramp pilot won almost three matches more than the average Simic player. Note that you can’t compare these points with the points for the Dreadhorde decks above. The inclusion of Taipei’s top finishers inflates totals across the board.
The data set for Kansas City alone, while smaller, looks similar: ten members of the Simic guild earned an average of 17.4 points, whereas eight players from the Bant shard made 26.9 points. So yes, Ramp with white fared much better in the recent GPs than Ramp without white, and both did better than Dreadhorde decks.
Detailed analysis of different card choices for Ramp remains impossible, because I only got access to the full deck lists of 25 Ramp players. How many does it take for a group to become a horde? I still don’t know, but 25 is not enough.