Is it Possible to Exploit the Bots in Arena Draft?

It’s a new year, and it’s a brand new world too. In this article, I want to share what I learned about MTG Arena in the roughly three weeks since I finally caved and downloaded the program. What follows are my greatest exploits, so to say.

Wikipedia tells us: “An exploit is [among other things] a sequence of commands that takes advantage of a bug or vulnerability to cause unintended or unanticipated behavior to occur on computer software […]”

This definition doesn’t fit perfectly, because it’s impossible to say how unintended the results really are, though I definitely found something that I myself didn’t anticipate. It may be a bug, but I’ve been using it as a feature, to great effect.

Before I decided to get into the Arena, I spoke with a bunch of knowledgeable folks. They all assured me that it was super easy—not to mention essentially free—to build a Constructed-level collection on Arena. Well, it appears I spend too much of my time talking to people who are way too good at Magic. And I’m not even being sarcastic here.

The problem is, if you’re a reasonably incompetent player like me, the following story may sound familiar: You work through your daily quests, and once you’ve saved up enough gold, you enter a Booster Draft. You draft well, read the signals coming your way, and end up with a deck that looks better than what you ever got in an offline, I mean tabletop, setting. You begin to feel optimistic about recouping your investment. Finally, here’s the gravy train that takes you from Draft to Draft, you think, and prepare yourself to watch the free prize packs come rolling in.

Then you run into an overpowered Boros decks against which you stand no chance at all. In one of the next games, you’re mana flooded. Eventually, your winning streak comes to an end before it has really begun. Then you face another Boros deck. This one’s even stronger than the first.

It felt like everyone else was using a cheat code. It began to dawn on me that drafting with bots is a completely different animal than a Draft with actual human beings. I can imagine that an exceptionally strong player may be able to make up for it by playing exceptionally well. Me, I needed a trick…

What I found went beyond my wildest hopes. As of New Year’s Eve, my collection contains all commons to make a playset of Guilds of Ravnica, as well as 313 of 320 uncommons, 194 of 212 rares, and 41/60 mythic rares, excluding Impervious Greatwurm and its ilk. I also acquired so many surplus cards, copies beyond the 4-of limit, that “the vault” opened for me four times just within the last week. I now possess wild cards for 66 commons, 95 uncommons, 46 rares, and 22 mythics. What’s best, almost all of these riches came from Drafts that indeed paid for themselves.

First Things First

There are two Draft options on Arena. One costs 750 gems or 5,000 gold to enter and lets you play a series of single games until you either lose three or win seven. If you win six, you walk away with a profit of 100 gems. If you lose three games before that, you’re down at least 100 gems. You get a prize booster either way, even if you go 0-3. You receive a second booster when you reach seven victories, at which point you’re also up by 200 gems.

The other “traditional” option comes with a heftier price tag of 1,500 gems and lets you play best-of-three matches until you either lose two or win five. If you manage to win three of your matches, you get back your 1,500 gems, and you can turn this investment into as many as 2,100 gems. This tournament structure also pays out way more packs, between one and six, meaning your collection fills up faster, despite the fact that each single Draft takes longer.

It quickly became apparent to me that this second variant makes for a vastly superior deal. But the two sets of parameters don’t line up sufficiently well that it’s strictly obvious. For the purpose of this article, I went and calculated how likely it was for a player with a certain game win percentage not to lose gems through either one of these Drafts.

It’s quite the difference, isn’t it? 57.7% to win a game translates to a 61.5% chance to win a match, and that’s enough to have slightly favorable odds to break even. Meanwhile, in the single-game version, you need a solid 68% win-rate not to shed gems. At no point along the spectrum do single games look more appealing. Unless you’ve given up hope and just want to spend less to fail.

This has two implications. The only reason to go for the bad Draft option is because you can pay the entry in gold, which you can mine for free through quests, and because it awards gems. In fact, this seems to be the main way to acquire gems for someone starting out who doesn’t want to expend any money. It is tempting to spend gems on single-game Drafts when you can afford the 750 entry fee but can’t afford the 1,500 for the better Draft option. Unless you’re confident to win games with 68% certainty, don’t. Wait until you have 1,500 gems and do a Traditional Draft.

The other implication is that it speeds things up considerably to buy some gems from the store. I first made the mistake from the previous paragraph, and then I reached for my wallet. This allowed me to make a serious attempt at figuring out the Guilds of Ravnica Draft format as it exists on MTG Arena

How I Became a Conclave Guildmage

While drafting, my first impression was that the bots overvalue black and under-appreciate white. Early on, I experimented with Boros, to medium success. Then I went back to my earlier analysis of Guilds Draft.

Back at Grand Prix Warsaw, Selesnya lost badly to the black guilds, but beat the red guilds. Surprisingly, this even included Boros, if only barely. On the other hand, Selesnya performed the worst overall, although it was, next to Golgari, drafted the least. So it seemed wrong for Selesnya to be a good choice to draft on Arena. I was skeptical, but I had to try and see for myself.

Well, by this point on our journey, you’ve come by one or two headlines, so you know where the trip is headed. It seems that the bots on Arena read my articles too! Selesnya may be the worst in tabletop town, but there are clearly limits to how little a color combination can be drafted before it starts to excel.

I dipped my toes carefully into the green-white waters at first and was immediately successful. For the next few days I drafted Selesnya whenever the opportunity arose. Then I began forcing Selesnya and never looked back.

After about 15 straight Selesnya Drafts I knew I was on to something. Only then did I realize that I should probably keep a record to share my findings later. Following that, I kept notes and pressed on. The following decks went 5-0:

Another ten decks carried me to a 5-1 record. One deck went 4-2 and I piloted another three to a 3-2 record. Overall, I went 98-33 since I started taking notes. You can find all of my latest 27 decks, some of which look quite more impressive than the ones above, in the order I drafted them, if you click here!

What can I say? Regular readers will already know that I like my samples on the large side. It’s also remarkably quick and easy to get a Draft done on Arena. Magic Online was already faster than tabletop drafting, and Arena takes this to a whole new level. Thanks to this nifty invention, I have now cast enough Rosemane Centaurs on turn 3 to last me a lifetime.

What does all of this mean, other than that I’m a degenerate Draft junkie? Well, personally, I’m convinced that Selesnya is the optimal strategy in the current environment. Granted, while I do have a reasonably sized sample, what I don’t have is a control group. If you have come to a different conclusion, I’m very interested to read your comments!

The bottom line, by the way, wasn’t outrageous. These 27 Drafts left me with a meager plus of 400 gems. But I did win 129 prize packs in the process, basically completing my collection of Guilds of Ravnica. I’m also left with a series of serious questions and concerns on how these Draft bots make their decisions. Foremost on my mind, how often do the bots change their minds, or rather: How often are their minds reprogrammed?

I heard stories about Dominaria Drafts, predating my time on Arena, that allowed people to force mono-red all the time. Wizards of the Coast apparently considered that to be an issue and fixed it.

So there’s really no telling whether or not Selesnya remains the optimal strategy by the time you read this article. I hope it does, and I think it’s likely. After all, it isn’t long before Ravnica Allegiance is due, and developers should have their hands and plates full with the new set. Adjusting the old Draft environment this late into the set’s run seems like a waste of resources.

But it could happen, so take that as a warning. If it does, consider the above a discourse on the possibly problematic nature of drafting with bots. Another question I have: When, if ever, will Arena implement drafting with human beings? Limited is a huge part of Magic, but bot behavior is eminently and evidently exploitable.

Should a legitimate esport be about finding the best exploit? That is another of my questions, and it is an honest question too. We don’t have to take issue with this issue, because even a weird map has everyone competing on a level playing field. Again, leave your thoughts in the comments.

Notes on Best-of-One

Another question: Why is it that the best-of-one Draft matches count toward my Limited ranking, while the actual match matches don’t?

I was mostly busy in the Traditional Drafts, but I squeezed in enough of the others that I believe bot behavior to be consistent across the two. That is, I’d recommend Selesnya for single-game Drafts as well.

There is one crucial difference, though. When sideboarding with the decks above, I often brought in an Intrusive Packbeast, a Hitchclaw Recluse, or Vigorspore Wurm, or went up to up to two copies of Righteous Blow, Prey Upon, Collar the Culprit, or Crushing Canopy. In single-game matches such is impossible, so I’d advise a more liberal main-deck use of traditional sideboard cards. See below for a successful example:

Final Question

Now that I can build any possible deck on Arena, where should I invest my hard earned wild cards? Is the Standard metagame lopsided to a similar degree, maybe overrun with mono-colored aggro decks? Your input is much appreciated!


Scroll to Top