Hearthstone CN vs EU 2017 Top 8 decks, results, and analysis

Hearthstone China versus Europe Championship 2017 was played over the past two weeks, and what a tournament it was! Top-level players and lots of good games.

In this post, I will take a look at the decks and results of the playoffs stage (top 8), including class distribution, archetypes, archetype performance, and ban decisions.

Class distribution in top 8

The players were allowed to submit new decks for the top 8 playoffs, and they reflect the lessons learned during group stages as well as the latest changes in the overall meta.

The decks were initially published on Facebook, and by now they are also available on many community deck sites. I like the format they are published here, so I’ll use that as a reference.

Classes in top 8 from the most popular to the least popular:

  • Paladin: 8
  • Rogue: 7
  • Druid: 7
  • Mage: 5
  • Warrior: 4
  • Shaman: 4
  • Priest: 3
  • Hunter: 2
  • Warlock: Zero. Zippo. Nada. Zilch.

Archetype distribution by class and performance

The archetypes brought into the playoffs showed signs of stabilization. No class featured more than two different archetypes, but on the other hand, all classes except for Hunter had two different archetypes represented. And well, there were still no Warlocks, just like in previous top-level tournaments during Journey to Un’Goro era.

Paladin decks were more focused on the Midrange Murloc archetype with even the more aggressive Murloc Paladins that run six one-drops opting to go for some good late-game cards as well. There were no Control Paladins in the top eight, possibly as a reaction to the abysmal performance of the Chinese-style N’Zoth Paladin in the group stage.

Secret Mage gained popularity, overtaking Discover Burn Mage as the most popular choice. Mage heavily underperformed in the top-8 as a class though.

Both Jade Druid and Aggro Druid were popular options, and both archetypes did well. Druid seems to be in a good position right now with two completely different viable archetypes – an advantage in tournaments where decklists are not public!

Dragon Priest went missing, and Silence Priest saw top-level tournament play as the most popular Priest archetype. Xhope’s Control Priest was the single best-performing deck though, going 3-0 in the playoffs.

Paladin

  • 5 Midrange Paladins (6-5, banned 1 time)
  • 3 Murloc Paladins (2-5)

Rogue

  • 5 Quest Rogues (2-4, banned 3 times)
  • 2 Miracle Rogues (3-2)

Druid

  • 4 Jade Druids (5-3)
  • 3 Aggro Druids (3-1, banned 2 times)

Mage

  • 3 Secret Mages (0-2, banned 1 time)
  • 2 Discover Burn Mages (1-3, banned 1 time)

Warrior

  • 3 Taunt Warriors (5-2, banned 1 time)
  • 1 Pirate Warrior (1-3, banned 1 time)

Shaman

  • 3 Jade Elemental Shamans (2-2, banned 2 times)
  • 1 Evolve Shaman (banned 1 time)

Priest

  • 2 Silence Priests (2-2, banned 1 time)
  • 1 Control Priest (3-0)

Hunter

  • 2 Midrange Hunters (2-3)

The archetype sample sizes are too low to draw far-reaching conclusions. Mage fared poorly in this tournament in sharp contrast to Dreamhack Austin while Taunt Warrior, Priest, and Druid did better.

Bans

I reviewed all the deck bans in the top 8 in comparison with Vicious Syndicate data from Data Reaper Report #47. Yes, I already got feedback on Twitter while discussing these live that VS data is not all-encompassing, and that players had also made tech choices for their decks. Nonetheless, any deviations from expected winrates should have some rationale behind them, and VS data gives a good baseline for comparison.

Overall, of the 14 bans during playoffs, only 7 were optimal according to VS data! Of the 6 bans by European players, only 1 was optimal, whereas of the 8 bans by Chinese players, 6 were optimal.

I will study this in further detail in a separate article, but it seems to me that there is a lot to be improved on when it comes to banning decks.

While players may have a grasp of whether matchups are favored or unfavored, it is hard to say off-hand exactly how favored or unfavored they are, and this greatly affects banning decisions.

I have also heard arguments that there is no single correct ban, but this line of thinking is simply incorrect. Banning is a fairly simple game that can be studied via game theory. If we know the winrates of all the decks, solving the Nash equilibrium for bans can always be done: sometimes the Nash equilibrium might be a mixed-strategy equilibrium (i.e. player should ban deck A x% of the time and deck B y% of the time) but studying actual examples has made me think that, in fact, the Nash equilibrium for deck bans is most of the time one of pure strategies (i.e. player should always ban deck A).

Deck spotlights

The lineups certainly had their fair share of cookie-cutter builds: Murloc Paladins, Midrange Paladins, Quest Rogues, Taunt Warriors, Jade Druids, and so on.

Nonetheless, here and there some slightly more hidden gems were among all the standard stuff. Some of the non-standard things I disliked, such as Evolve Shaman (that we did not get to see in action though) and the Burn/Secret hybrid Mages that had turned the tempo deck that is Secret Mage into a strange amalgamation that attempted to also have a lot of value in it.

Here are some things I liked though:

StanCifka’s Fast Miracle Rogue

I have had a soft spot for non-Giant Miracle Rogue builds since the early days of Un’Goro. While I did not like StanCifka’s original list that he brought to the group stage, I find his playoff list much more refined.

Cutting Sap is a move I wholeheartedly agree with in this type of a deck, and StanCifka also found room to include Shaku, which is such a great card. He opted to run Southsea Deckhand over Leeroy, a move I am not completely sold on, but it seemed to work admirably.

Xhope’s Control Priest

Bring Control Priest, check. Go 3-0 with it against some of the best players in the world, check. Xhope’s Control Priest is so greedy that I wonder whether it could work on ladder, but it obviously worked in this tournament.

Bonus points for bringing Medivh + Free from Amber combo AND two Arcane Giants. How do you even fit all of that into one deck?

StanCifka’s Jade Elemental Shaman

I still don’t think this is the Jade Elemental Shaman we will all be playing in two months, but it is nice to see players innovate around the archetype.

StanCifka’s deck really reminds me of BoarControl’s Jade Elemental Shaman builds, both his first one from the very first days of Un’Goro with the Pirate package and his later build that cut the Pirates but included Hammer of Twilight.

StanCifka’s deck features some of those choices and in addition chooses to cut elemental synergy cards to the minimum and include Stonehill Defenders for additional value.

There are so many pieces to Elemental Shaman floating around with the Elemental cards, Jade cards, Pirate package, Hammer of Twilight, Stonehill Defender, Thing from Below, and Spirit Echo to name a few, and we’re still looking for the best package to build from these. I’m just happy Volcano is not part of the list, I’m growing more certain by the day that it is not part of non-Control Shaman.

Appendix: Matches

Raw data on matches, lineups, and bans.

Quarter-final 1: OmegaZero vs OMXhope

Lineups and expected win rates based on Vicious Syndicate’s data:

Midrange Murloc Paladin Jade Elemental Shaman Control Priest Jade Druid Taunt Warrior
Midrange Hunter 0.52 0.38 0.47 0.52 0.47
Midrange Murloc Paladin 0.50 0.49 0.64 0.64 0.65
Jade Druid 0.36 0.53 0.54 0.50 0.67
Quest Rogue 0.49 0.67 0.61 0.70 0.60
Secret Mage 0.51 0.48 0.58 0.57 0.51

 

Expected bans: OmegaZero was expected to ban Midrange Murloc Paladin and OMXhope was expected to ban Quest Rogue.

Actual bans: OmegaZero banned Midrange Murloc Paladin and OMXhope banned Secret Mage.

Midrange Murloc Paladin wins against Jade Druid.

Quest Rogue loses to Taunt Warrior.

Quest Rogue loses to Control Priest.

Quest Rogue loses to Jade Elemental Shaman

Quest Rogue wins against Jade Druid.

Jade Druid wins against Jade Druid.

Midrange Hunter loses to Jade Druid.

OmegaZero 3 – OMXhope 4.

Quarter-final 2: StanCifka vs Orange

Lineups and expected win rates based on Vicious Syndicate’s data:

Quest Rogue Taunt Warrior Secret Mage Jade Druid Murloc Paladin
Miracle Rogue 0.39* 0.51 0.42 0.59 0.50
Aggro Druid 0.70 0.29 0.49 0.55 0.57
Jade Elemental Shaman 0.33 0.41 0.52 0.47 0.56
Midrange Murloc Paladin 0.51 0.65 0.49 0.64 0.43
Silence Priest 0.44 0.51 0.42 0.71 0.44

* Stancifka’s Miracle Rogue is a faster variant without Arcane Giants, so it is a bit closer to Tempo Rogue in nature. Therefore, the Quest Rogue matchup is likely better than indicated by this data.

Expected bans: Stancifka was expected to ban Quest Rogue and Orange was expected to ban Midrange Murloc Paladin.

Actual bans: Stancifka banned Quest Rogue and Orange banned Aggro Druid.

Miracle Rogue loses to Taunt Warrior.

Jade Elemental Shaman wins against Murloc Paladin.

Silence Priest wins against Murloc Paladin.

Midrange Murloc Paladin wins against Secret Mage.

Miracle Rogue wins against Murloc Paladin.

StanCifka 4 – Orange 1.

Quarter-final 3: Pavel vs Xixo

Lineups and expected win rates based on Vicious Syndicate’s data:

Miracle Rogue Discover Burn Mage Jade Elemental Shaman Midrange Murloc Paladin Aggro Druid
Quest Rogue 0.61 0.48 0.67 0.49 0.30
Discover Burn Mage 0.64 0.50 0.59 0.49 0.52
Taunt Warrior 0.49 0.42 0.59 0.35 0.71
Jade Druid 0.41 0.63 0.53 0.36 0.45
Midrange Murloc Paladin 0.47 0.51 0.49 0.50 0.38

Expected bans: Pavel was expected to ban Midrange Murloc Paladin and Xixo was expected to ban Quest Rogue.

Actual bans: Pavel banned Aggro Druid and Xixo banned Discover Burn Mage. With Pavel banning Aggro Druid, Xixo’s ban turned out to be nearly optimal, despite seeming strange at first sight.

Midrange Murloc Paladin wins against Discover Burn Mage.

Jade Druid wins against Miracle Rogue.

Quest Rogue wins against Discover Burn Mage.

Taunt Warrior loses to Miracle Rogue.

Taunt Warrior wins against Midrange Murloc Paladin.

Pavel 4 – Xixo 1.

Quarter-final 4: ahqDogggg vs Trunks

Lineups and expected win rates based on Vicious Syndicate’s data:

Murloc Paladin Evolve Shaman* Silence Priest Quest Rogue Secret Mage
Pirate Warrior 0.57 0.39 0.57 0.69 0.51
Midrange Hunter 0.50 0.38 0.45 0.61 0.46
Murloc Paladin 0.50 0.44 0.56 0.61 0.54
Quest Rogue 0.39 0.67 0.57 0.50 0.35
Aggro Druid 0.57 0.40 0.47 0.70 0.49

* There is not enough data on Evolve Shaman matchups. I have used Elemental Shaman data instead, it seems similar based on the few results available.

Expected bans: ahqDogggg was expected to ban Evolve Shaman and Trunks was expected to ban Pirate Warrior.

Actual bans: ahqDogggg banned Evolve Shaman and Trunks banned Pirate Warrior. Expected bans for once!

Murloc Paladin wins against Silence Priest.

Aggro Druid wins against Silence Priest.

Quest Rogue loses to Silence Priest.

Midrange Hunter loses to Murloc Paladin.

Midrange Hunter wins against Secret Mage.

Quest Rogue wins against Quest Rogue.

ahqDogggg 4 – Trunks 2.

Semi-final 1: OMXhope vs StanCifka

Lineups and expected win rates based on Vicious Syndicate’s data:

Miracle Rogue Aggro Druid Jade Elemental Shaman Midrange Murloc Paladin Silence Priest
Midrange Murloc Paladin 0.47 0.38 0.49 0.50 0.67
Jade Elemental Shaman 0.49 0.60 0.50 0.51 0.55
Control Priest 0.41 0.54 0.47 0.36 0.43
Jade Druid 0.41 0.45 0.53 0.36 0.29
Taunt Warrior 0.49 0.71 0.59 0.35 0.49

Expected bans: OMXhope was expected to ban Midrange Murloc Paladin and StanCifka was expected to ban Taunt Warrior.

Actual bans: OMXhope banned Silence Priest and StanCifka banned Jade Elemental Shaman. Neither ban seems to be optimal, but they are both second-best options.

Control Priest wins against Midrange Murloc Paladin.

Taunt Warrior loses to Midrange Murloc Paladin.

Midrange Murloc Paladin loses to Miracle Rogue.

Taunt Warrior wins against Aggro Druid.

Jade Druid wins against Jade Elemental Shaman.

Midrange Murloc Paladin loses to Aggro Druid.

Midrange Murloc Paladin wins against Jade Elemental Shaman.

OMXhope 4 – StanCifka 3. OMXhope advances to the Grand Final!

 

Semi-final 2: Pavel vs ahqDogggg

Lineups and expected win rates based on Vicious Syndicate’s data:

Pirate Warrior Midrange Hunter Murloc Paladin Quest Rogue Aggro Druid
Quest Rogue 0.31 0.39 0.39 0.50 0.30
Discover Burn Mage 0.54 0.49 0.42 0.52 0.52
Taunt Warrior 0.54 0.53 0.42 0.40 0.71
Jade Druid 0.50 0.48 0.36 0.30 0.45
Midrange Murloc Paladin 0.47 0.48 0.43 0.51 0.38

Expected bans: Pavel was expected to ban Murloc Paladin and ahqDogggg was expected to ban Taunt Warrior.

Actual bans: Pavel banned Quest Rogue and ahqDogggg banned Taunt Warrior.

Discover Burn Mage loses to Midrange Hunter.

Discover Burn Mage wins against Pirate Warrior.

Jade Druid loses to Aggro Druid.

Quest Rogue loses to Murloc Paladin.

Jade Druid wins against Pirate Warrior.

Midrange Murloc Paladin loses to Pirate Warrior.

Pavel 2 – ahqDogggg 4. Dogggg to the finals with his aggro lineup!

The Grand Final: OMXhope vs ahqDogggg

Lineups and expected win rates based on Vicious Syndicate’s data:

Pirate Warrior Midrange Hunter Murloc Paladin Quest Rogue Aggro Druid
Midrange Murloc Paladin 0.47 0.48 0.43 0.51 0.38
Jade Elemental Shaman 0.61 0.62 0.56 0.33 0.60
Control Priest 0.37 0.53 0.41 0.39 0.54
Jade Druid 0.50 0.48 0.36 0.30 0.45
Taunt Warrior 0.54 0.53 0.42 0.40 0.71

Expected bans: OMXhope was expected to ban Quest Rogue and ahqDogggg was expected to ban Jade Elemental Shaman.

Actual bans: OMXhope banned Quest Rogue and ahqDogggg banned Jade Elemental Shaman. Dogggg hit the expected ban in each of his matches!

Control Priest wins against Pirate Warrior.

Jade Druid wins against Murloc Paladin.

Taunt Warrior wins against Murloc Paladin.

Midrange Murloc Paladin wins against Midrange Hunter.

OMXhope 4 – ahqDogggg 0. A huge win for Xhope! Dogggg crumbled in the Grand Final, but on the other hand, Xhope played a solid tournament to take the win.

Leave a Reply

Your email address will not be published. Required fields are marked *

IMPORTANT! To be able to proceed, you need to solve the following simple math (so we know that you are a human) :-)

What is 12 + 11 ?
Please leave these two fields as-is: