greylag is an Epic-loving goose who prowls the BGG forums and, now, the alpha app. She loves the weird stuff in Epic: pulling things from discard piles, odd little card combinations and, always, mass death via zombies. And somehow, she always seems to end up theorising the games she plays…
I’ve had a lot of fun with the puzzle contests run on Tom’s blog, in which we’re given a challenge: play as many champions as you can in one turn, or see how much total offense/defense you can get. The numbers possible in those challenges are ridiculous, but they’d be impossible to achieve without the puzzle rule which says you can decide exactly what card you’d get on every draw. Wouldn’t it be nice to have that ability in normal Epic? Good news: you can. Well, sort of.
I’m going to give a name to the 30-card deck you start your game with: Deck 1. (I won’t be talking about Constructed here, because I don’t think what I have to say really applies to 60-card decks. If you think it does – cool. Say how in the comments.) Deck 1 is shuffled before you start the game, you draw 5 cards out of it, you can mulligan if you want. Your next 20-25 draws (depending on how many cards you mulliganed) are your Deck 1, and are completely random – all you know is which cards were shuffled to form Deck 1 in the first place, or perhaps not even that in Random 30.
But if the game goes on long enough, probably something interesting will happen. You’ll hit a card you’ve seen before. In most cases, this will be one of the cards you mulliganed; otherwise it’ll be the first card you recycled or which was banished, usually from play or discard. This card is the first card of Deck 2. Deck 2 is fundamentally different from Deck 1: Deck 2 is ordered. You probably don’t know the exact order, but you can potentially know a lot.
Let’s limit it to the (alpha) app for now. Looking at the game log, you can see a complete list of what’s happened. From that, you can pretty much see what was recycled and banished and when. In the case of recycles, you get to choose the order of the two cards. In the case of 2+ cards being banished, they are shuffled before being put on the bottom of the deck, but you still know which cards were in each group of banished cards.
More formally, Deck 2 is an ordered list of card groups. If all those groups are just 1 card in size, it’s an ordered list which you can have perfect information about. If the groups are larger, you still have a huge amount of information.
Let’s call the first group in Deck 2 the marker group. Once you see a (or the) card in the marker group, you know from now on pretty much what you will draw, almost every time you draw. If you play Mist Guide Herald (not in the app yet), you can make a good guess at which champions might be revealed out of the 5 cards. If you Surprise Attack, you can guess what’s coming up, and same with Fairy Trickster.
What does this mean for play? My suggestions are:
Memorise the marker group. This is usually but not the same as: remember which card(s) you mulliganed. Once you see a card in the marker group, you can start checking the game log to see what might be coming up next.
Memorise your opponent’s effective marker group. Interestingly, this will usually be different to the cards they are memorising for their own marker group, because you don’t know what they mulliganed. But since you don’t know their hand or their mulliganed cards at the start of the game, it doesn’t matter. All 25 of the cards in their deck (however many they mulliganed) are unknown, and so practically for your purposes you’re interested in the first time they will draw a card you know.
If you want a rough guess at when your marker cards will be coming up, use the figure of “25 turns, less mulligans, less 1 per recycle”. If a player spends their gold every turn, to keep up on cards they will need to draw a card on average once every turn. (This often works out as 2 natural draws and one draw 2 every 2 turns.) This means 25 turns to draw through the deck. 0-cost recycles (practically all the recycles) accelerate the arrival of Deck 2 because they generally replace a card in hand with a fresh draw. For your opponent’s marker cards, use the same rule of thumb but discount mulligans (see point 2 above).
You can card count if you want to and know exactly when both players’ marker cards (yours, and your opponents’ effective markers) will be coming up. This is very laborious in the current state of the app, though it may get easier if/when WWG update the game log. Basically, just make a note every time there is a draw, and have your card totals in mind. 26 draws will put your opponent’s first effective marker in their hand, and 26 less mulligans will put your first marker card in yours.
Expansions make this even more interesting. Mist Guide Herald brings you 4 cards closer to Deck 2 when you play it. Play it twice, and you’re 1/3 of the way there already. However, it also makes your Deck 2 more unpredictable, since you’re banishing 4-card groups which are internally random. Only the last card of each group can be predicted perfectly. Arcane Research and to some extent Fairy Trickster also accelerate Deck 1 into Deck 2.
There’s a special note to Arcane Research, which is that if you know how many cards remain in Deck 1 and you’re looking for a particular card – say, Flame Strike – you know exactly how many you need to banish to be sure of seeing that card. And if you’re already in Deck 2, Arcane Research is in theory a precision tool. For this reason, if Arcane Research is in my opening hand, and I’m not running an aggressive deck, I almost always want to mulligan it, to ensure I will draw it when it is most useful.
What you do with the information above is up to you. There is more to be written about the strategy of how you compose your Deck 2, and what you want in it. There are also interesting points around the fact that both players’ Deck 2s are built collaboratively. Both you and your opponent will be acting to compose your Deck 2, and these choices can be in tension with what is best for the current board state. (Do you banish their good card from discard now, so they can’t get it back, but know it will come earlier in their Deck 2? Or do you banish a number of other cards first, betting they can’t recur it from discard right away, so that they won’t get that Flame Strike until late in Deck 2?)
But I hope I’ve at least made you think about the possibilities of what it means when a game of Epic becomes semi-deterministic in terms of drawn cards. You get the possibility to re-enact the opening moves of the game, but this time with all your questions answered. Do they have Lash, for their Kong or for your Thought Plucker? Now you know, and can play accordingly.