Demystifying the mystery of winning decks

I’ve found myself asking a question: How many decks that are similar to vault tour winning decks exist? And what does it mean for the game? I’m here to share with you my exploration of that question, and my answer to it. As always, I’m open to discussion on the subject, and would love to hear your opinion.

Investigation

The first step for me was to go look at vault tour winning decks, and see how unique they are, as in, can I find similar decks in the master vault?

Not long ago we had our first ever double vault tour champion, George Keagle, championing “Gasoline” Maximiliano, Dungeon Keeper both times. Although it could be argued he played an entirely different deck the first time, since it was before the Library Access errata. I started with this one, as it has been proven amazing twice.

Gasoline

So what makes this deck tick? It’s probably not the Bigtwig or the Mushroom Man. What stands out to me is two Control the Weak, Restringuntus, Nepenthe Seed and Skippy Timehog. During the first run of this deck, Library Access was certainly key, but with the errata it doesn’t quite do that much work anymore, so I’m not putting it up there. It also has an Expected Aember of 20, and an Aember Control score of 6.5, which is not very high, but higher than the minimum we’ve seen at vault tours.

Putting those four cards in the search parameters actually really surprises me. There are only three decks, including Gasoline, that have that combination of cards. Neither of the two other decks has comparable Aember Control in terms of score, but they seem pretty similar to me. This actually makes me petty happy, the Algorithm manages to diverge the decks, and doesn’t spit out similar decks with a few cards difference.

But let’s try and find a deck that would play similarly, without having exactly the same cards. Obviously the main things stopping us from finding a similar deck are the rares. So let’s drop the Skippy, which is undoubtedly a very powerful card, and replace it with the uncommon Scrambler Storm. It’s not a Skippy, but it similarly restricts your opponent. This produces an unimpressive seven decks.

It’s becoming evident that Gasoline is quite a unique deck. Even removing Nepenthe Seed, which is forcing an entire house choice, we only end up with 122 decks.

Let’s move on to a deck which I heard described as not very unique, the winner of the Krakow vault tour is Litmus, the Stone Keeper of Joy.

Keeper of Joy

What stands out to me here is the double Speed Sigil and Timetraveller. The Dis here is not terribly inspiring, it just has some Aember Control. I found 96 such decks, and 21 of which have Dis.

It is fairly evident by now that it’s rather impossible to find decks very similar to winning decks. This makes me quite happy, as it encourages everyone to discover their own winning decks, rather than copy the proven decks, which is what happens in most card games.

Scarcity of winning decks

We have not had a lot of top level tournaments, and therefore we have not had a lot of top tournament winning decks. If you compare the top decks among themselves, you’ll find they have very little in common. It’s not that there are three Gasoline decks out there, and all three of them won a vault tour. There are many different decks that play in their own way, and have won vault tours.

People have a weird way of looking at new evidence as if it was always the case. The Snappy Pariah (winner of Gencon) won a vault tour? Well, of course, look how amazing it is! But what if it didn’t win the vault tour? What if Viking Ballinger (top 16 at Gencon) won the vault tour? Well obviously, look how amazing that deck is!

If you consider the top 16 of every vault tour as decks that could potentially have won a vault tour. Maybe if they were played by a better player, maybe if luck went a little more their way. Then there are roughly 128 decks that could have won vault tours. If you search for decks similar to any of those, then you’d find at least a thousand.

The Metagame

The Meta of most card games is defined by the most consistent decks, the ones that perform the best. When a deck can be copied any number of times, what you get is a largely consistent field, with small variations and some counters and the occasional rogue player (Someone that brews their own unique deck). In keyforge however, you can’t imitate a winning deck exactly, so what you get is a large variety of decks that imitate the winning strategy, rather than the exact cards.

If a deck with a lot of burst and steal wins, then the field the next tournament will have more of those. If the winning deck has multiple Control the Weak, then expect to face a bunch of those the next day. But the nature of the game makes it so even those winning strategies don’t swamp the field, because players have the decks they discovered to be great, and they want to prove them great. In keyforge, the largest portion of the field is rogue players.

The metagame had no real reason to change over the past year of the keyforge’s lifespan, as only one new set has been introduced, and that set has only recently begun to make it to top 16 on a more consistent basis. And yet, I found that the top decks in different top tournaments all look quite unique.

There are definitely more strategies that are going to come up from new sets, but for now, there aren’t that many. You can play control, usually with Dis, preventing your opponent from doing things. You can burst Aember while stealing to forge keys as fast as possible. Or you can play a combo deck, that wins spontaneously out of nowhere. We’ve yet to see significant number of decks that win by building an insurmountable board, but there have been a few.

Pay to win

I myself piloted quite a unique deck at Krakow, and if I had won, people would undoubtedly be looking for decks similar to it. And that would have driven up the prices of decks like it. But since it didn’t win, you could probably find a similar deck for a reasonable price.

When Bahamut “Alp Larissa” Heifetz won the vault tour (without winning a single game in the finals) prices of similarly looking decks skyrocketed. You would have to pay a hefty sum if you wanted to pilot a deck similar to it. But has a deck similar to it win a vault tour since? No, it has not.

So rather than pay to win, I’d argue that it’s paying to imitate. It’s paying in order to skip the discovery aspect of the game. That’s similar to paying in order to skip the grind in an MMO-RPG and jump straight to level 60. I’m not here to tell you how to have fun, and if you have fun finding a deck similar to a winner and playing that, all the power to you.

But if your objective is to find a deck that can win, it is of my opinion you should discover it. And I’m not saying not to buy one on the secondary market, just that if you do, find one that is unique. One that is you. One that speaks to you personally. And probably one someone else overlooked because they didn’t connect with it, so the price is right.

Potential

So how many decks are out there that can win a vault tour? Is it in the hundreds? The thousands? The tens of thousands? I don’t know, but I feel that the number is large enough that anyone that has ever attended a high level tournament could own one, or even three.

And this number is only going to grow as new sets are being introduced. With FFG’s evergreen approach to never cycling out decks, and provided there isn’t too much power creep, we’re likely to see an ever expanding pool of top level decks.

Blindness

When I was looking for an Archon deck for Krakow we had a very lively discussion on how to do so on the Sanctumonious discord. The conclusion was a certain range of Aember Control, Expected Aember, overall AERC and SAS in order for a deck to qualify as a vault tour deck.

Flare

This gave me a sort of blindness when looking at decks that didn’t meet those requirements. After the vault tour I looked over my collection and found Flare the Conspicuously Philanthropic. Which I have completely overlooked when opening it, due to said blindness not letting me look past the fact it has very little Aember Control.

Besides being a blast to play, I’ve also been doing quite well with it. And I think my blindness is still partially there, preventing me from considering it for high level tournament play. So I have decided to test it more vigorously, and see how that goes.

Contact and afterward

So it is my belief that the winning decks are put on a pedestal because they have won, but there are many more decks out there that can do similarly well piloted by skilled players.

As always, you can follow me on twitter for updates. And join us at the Sanctumonious discord server if you’d like to chat with me, or join an awesome community of keyforge players.

 

 

3 thoughts on “Demystifying the mystery of winning decks

  1. Great post! Appreciate the research you did in the number of similar decks compared to the two-time VT winner. And perfect analogy for buying a top deck being like skipping the grind in an MMO. I personally prefer to discover and play with decks that work for me, rather than go online and find a LANS deck for 10x the normal deck price (this was pre-errata).

    Liked by 1 person

  2. The conclusion was a certain range of Aember Control, Expected Aember, overall AERC and SAS in order for a deck to qualify as a vault tour deck.

    And what were the parameters for this that you ended up at?

    Like

    1. Something like A>10 E>20 AERC>60 and SAS>80

      But the AERC has changed since.

      If you join our discord, we have a sticky in the vault tour channel with vault tour deck stats made by coffeesaga.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s