How unique is unique?

By CountBlah, in KeyForge

Some of the spoilers have collected over a 100+ decklists now, and I've been looking at them to get a grasp on the random unique nature of each deck. Each deck has 12 cards from 3 distinct houses, so I looked at how much overlap exists between decks with the same house. Specifically how much duplication can be expected in those 12 cards, based on the current set list.

  • I found that most decks seem to share 2-4 cards when they both contain the same house. In rare cases the numbers were as low as 0 or or higher. So normally if you compared a decklist that shared only 1 house, 2-4 cards or 6-12% of the decks matched.
  • Looking at the few examples where all 3 houses matched, the numbers were averaging out to the lower end and I generally found 6-8 matching cards across the entire decklists. Even in these cases, only 16-22% of the decks matched so they were very distinct from each other.
  • What percentage of 2 decklists need to match up in order to be considered similar? I expect it would come down to specific cards, but expecting anything north of 33% looks like it would require 10s of thousands of decks.

I'm not sure if this unique nature will kill out any collecting aspect or increase the perceived value of any specific deck because you have no real chance of finding a similar deck.

On the other hand, embrace your decks because you won't likely see any others like them.

Given the premise of the game, I think it would be disappointing to stumble across two decks that were very similar, though presumably it’s inevitable at some point (and it depends on where you draw the line for what constitutes “too similar”).

I think it would be a bummer to open two decks that shared the same three houses and then also had a lot of cards in common, but then you also have an interesting exercise in giving both decks a workout and seeing how those similarities and differences play out on the table, so it could still be cool from a certain angle.

Edited by timhodge

Honestly if I found 2 decks that were very similar I would treasure them.

When everything is different, similarity is a novelty.

Besides, even if 20 of the cards are the same, you still have 16 different ones. and those 16 can alter how you play and what you can do. Especially since they will likely be the rares which are more swingy.

Even 1 card difference will make a deck unique. Maybe not in playstyle but unique in composition and deckname.

1 hour ago, dirtmuncher said:

Even 1 card difference will make a deck unique. Maybe not in playstyle but unique in composition and deckname.

If you end up getting two decks that close to each other, I think you go out doors and either win the lottery or get struck by lightening. ?

So, a quick mental exercise on this...

First off, the houses matter. If two decks share two houses but the third isn't the same, they'll still play differently. Sure, they may play more similarly than two decks with completely different houses, but they'll still play differently.

Next, we can look at the individual cards. There's over 300 cards in the set, so that's somewhere around 45 unique cards per house. Let's assume those 45-ish cards can be broken down into three "archetypes" per set, and that each deck gets one "archetype" (with a handful of cards from the entire house) for each house.

What does that mean for combinations? Well, first off, that's 35 unique combinations of houses. Next, if each house has one of three broad archetypes picked at random, that means there's 27 different combinations of archetypes for each of the 35 different house combinations, for a total of 945 different "broad strokes" decks. The odds of pulling two different decks that are very similar (same three houses with several repeats in each house) are small enough that you'll probably never see a pair of very similar decks unless you go to a lot of public events for the game (or buy a lot of decks).