Placeholder

Theoretically, there should always be a best deck in Hearthstone. In a perfect world, there very much is one that will have the highest win-rate based on the rest of the decks that make up any given meta. Things have been and will continue to shift in the wake of the Caverns Below nerf, at least until Knights of the Frozen Throne releases. We don't live in a perfect world so to find out what the meta should actually look like we're going to have to make a few assumptions.

Reddit user Aaron_Lecon is back with another Nash Equilibrium using Vicious Syndicate's data.

What's a Nash Equilibrium you ask? In short, it's a way to solve a noncooperative game assuming every player knows each other's equilibrium strategies. Aaron explains it well:

 

In this state, there are two types of deck : non-viable decks (also called bad decks), which have sub-50% win rate, and that nobody plays. The other decks are called viable, and they do see a certain amount of play and all have exactly 50% win-rate. Thus, there is no way of 'countering the meta' since switching deck gives you 50% win rate or worse. Since no one has any incentive to switch from one of the viable decks, the meta stagnates here for ever.

 

As we mentioned earlier, there are also a few assumptions that must be made, even if they aren't accurate or plausible.

  • Vicious Syndicate's data reaper table (as it was yesterday) is accurate.

  • If a deck has a positive (over 50%) win rate, more people will play it. If a deck has a negative (less than 50%) win rate, fewer people will play it.

  • The number of people who start or stop playing a deck is proportional to the difference between its win rate and 50% (so for example, if two decks have 54% and 51% win rates, then for every person who starts playing the second deck, there are 4 people who start playing the first one).

Aaron previously pegged Midrange Paladin as the deck to beat in the Journey to Un'Goro meta.

Given this, here is what the perfect meta would look like right now. In short, Paladin, Warrior, and Mage should be the best classes with Jade Druid keeping them in check periodically.

  • Control Paladin - 33%

  • Pirate Warrior - 21%

  • Midrange Paladin - 19%

  • Freeze Mage - 17%

  • Jade Druid - 3%

  • Murloc Paladin - 2%

  • Taunt Warrior - 2%

  • Silence Priest - 2%

  • Midrange Shaman - 1%

Note that none of this actually means anything. We don't live in a perfect world so the meta won't settle exactly like this.

Secret Mage, Miracle Rogue, Token Shaman, Token Druid, and Burn Mage may very well deserve a spot in this meta as well, but Aaron explains that the Syndicate numbers have them narrowly missing the field. They're within the realm of error as well.

If you need a guide for any of these archetypes, make sure to check out the Top Decks section on our front page for the latest deck lists and guides.