vS Data Reaper Report #6

A weekly Hearthstone Meta Report based on data from over 33,000 games now with a LIVE (beta) Meta tracker!

Data Reaper Report Logo

vS Data Reaper Report #6

 Welcome to the sixth edition of the Data Reaper Report!

We’re happy to inform the community that Data Reaper Live (beta) now has over 740 contributors. Over the past week, we’ve compiled 33,000 games. This is getting us closer to another threshold which will allow us to further expand our analysis. If you have not done so already, please sign up with your Track-O-Bot information here:

sign up - data projects

Take note that next week’s article, the Data Reaper Report #7, will be published on Saturday, July 2nd, instead of Thursday.

If you’d like to help support our initiative to provide high-quality data-driven content to the community for free, you can join our team by supporting us on Patreon:

patreon ad wide

On to the numbers! We begin with our deck frequency charts. These are the numbers from June 15 to June 21. The first chart shows all ranks, the second can be switched between different rank groups. We’ve recorded about 4.5k games at legend, 10k games at rank 5-1, 10k games at rank 6-10 and 5k games at ranks 11-15.

Overall Games

By Rank Games

Note that we’ve made some improvement on the way we classify and identify the Druid archetypes. “Yogg Druid” will now refer to every spell heavy Druid that utilizes Yogg and could have different win conditions (much like Miracle Rogue), while the old school Ramp Druid which strictly utilizes big minions and taunt walls will be called “Ramp Druid”. Druid is certainly a tricky class to classify since the different archetypes have great similarity in their structure, and it’s been undergoing many changes for the past month, but we feel that we’ve further honed the accuracy of our identification and tagging.

Next is a graph displaying the popularity of classes during the last six weeks: since the Data Reaper Project launched.

Class Frequency by WeeksClass Freq. by Weeks DRR 6 Small

Interesting notes we take from the data:

  1. Tempo Mage is the star of the week. Mage has now surpassed Druid to become the 5th most common class on ladder on the back of recent resurgence in the archetype. Its numbers are even higher at legend ranks, so it is certainly receiving more respect across all skill levels.
  2. Dragon Warrior has finally been noticed and is soaring in popularity. It has eclipsed its mid-ranged cousin, Tempo Warrior, to be the 3rd most played Warrior archetype. The numbers of Dragon Warriors also increase at higher levels of play.
  3. Shaman is continuing its trend of becoming more and more aggressive every week. We’re now at a point where Mid-Range Shaman is almost a rare occurrence at legend ranks. Don’t be greedy with your mulligan against Shaman, search for your answers. Thrall wants your face.
  4. The overall number of Warriors is becoming obscenely large, especially at legend ranks. We believe there is a tendency in many ladder players to “run back to Warrior” when they try to grind the legend ladder, especially during the final week. It occurred towards the end of May, and this pattern is showing signs of repeating in June.

Match Win Rates Header

Now let’s continue to our updated matchup charts. The numbers are taken from the past five weeks. We’re aware of the community’s desire to split these numbers into skill groups, and we fully plan on doing so. It simply requires us to have a larger database, and we’re getting there with your help and support. Also, we’ve changed the alphabetic ordering of decks to be based on class first, to make it easier for you to find the matchup that you’re looking for.

Matchup Winrates Data Reaper Report


Power Rankings Header

Finally, we present the updated “vS Power Rankings” table for this week. The numbers we present are the expected win rates of each archetype based on their matchups against the field, factoring in the frequency of all potential opponents on ladder at different rank groups over the past week.

We can see that the top 3 decks remain just about the same: Dragon Warrior, Aggro Shaman and Zoo Warlock. However, Mid-Range Hunter has increased its score, and is flirting with Tier 1. It’s also relatively stronger at legend ranks, which we believe is the result of the higher numbers of Warriors. Another interesting deck to note is Yogg Druid. With our increasing database, we felt comfortable to add the archetype (as well as others) to these rankings as it’s gathered enough games. We can see that the deck is not only fun, but is also inherently strong against the field. Is this under the radar archetype about to trend upwards as we’ve recently seen with Dragon Warrior?

Also, we’ve followed and read a lot of feedback and discussion regarding the vS Power Rankings. Many mentioned the “skill cap” of decks and how these numbers as well as the matchup win rates could be influenced by people not piloting certain decks correctly at lower ranks. While we do believe that some decks will have significantly better scores at higher skill groups (such as Freeze Mage), not all of them can. When one deck wins more, another deck has to lose more, so we wouldn’t expect a deck like Control Warrior to suddenly jump into Tier 1 when it’s being played at legend ranks (hint: it won’t, it’s just inherently disadvantaged against the current field).

Of course, we repeat that there are many variables that can alter a deck’s performance from the expected win rate, such as the individual’s proficiency with the deck, understanding of certain matchups, as well as card choices. This can push the actual win rate significantly higher than the expected win rate, and this is why certain individuals such as Fibonacci perform so consistently on ladder with a specific archetype.

The important thing to take away from the rankings is the relative strength of the decks, when piloted by the average player in the field on ladder. It is an objective measure, with a transparent formula, albeit with some assumptions. A particular player who pilots deck X and achieves a win rate far higher (or lower) than the expected rate, can conclude that they play the deck better (or worse) than the average player on ladder, provided that they have a sufficiently large sample size.

Data Reaper Report - Warrior

Once again, we see Tempo Warrior being included in a spring championship winner’s lineup. First, it was Thijs who won the EU championship, and last weekend Handsomeguy has done the same at the APAC championships. Will this pattern continue at the America’s championship? Tempo Warrior appears to be the optimal Warrior deck for this tournament’s format, which is a best-of-7 conquest, using five decks and one ban. However, most players will never play in such a format, since open tournaments are usually best of 5’s. Either way, with Handsomeguy’s victory, we’re confident that the archetype will maintain consistent presence on ladder. His list is vanilla and straight forward, which makes it suitable for ladder play.

Dragon Warrior finally got a decent showing in a major tournament at the hands of Edward Elric. The deck feels like Tempo Warrior with a stronger mid-game. It was only a matter of time until this deck succeeded in tournament play and we suspect this trend to continue in the future. While Tempo Warrior is still the more established archetype, Dragon Warrior is beginning to eclipse it in numbers, as more people are beginning to realize the deck is legitimately strong and has good matchups across the field. Edward’s list should serve you well on ladder.

Lately, Control Warrior builds have been straying away from relying on the Golden Monkey to provide you with the tools to end the game, switching to a more consistent win condition: N’Zoth. Heisnotaxel’s list is quite optimal, but flexible enough to include different deathrattle minions and tech cards, such as Barron Geddon, depending on the Metagame. The archetype seems to be falling out of favor slightly, but all Control Warrior diehard fans know that it will always remain relevant in the Metagame to some degree. Piloting the deck well can still produce success on ladder.

Finally, C’Thun Warrior feels stronger in the current Metagame than a standard Control Warrior because of a stronger mid-game as well as blow-out potential of drawing C’Thun. Pinpingho piloted a slightly different variant of C’Thun Warrior at the APAC championship. The inclusion of Emperor Thaurissan enables playing Brann and Doomcaller, putting two copies of C’Thun back into your deck, which allows you to dominate other control decks. Against decks such as Control Priest and Control Warrior, it is not out of the question to Shield Slam your own C’Thun in order to guarantee that it dies in order to pull off this game winning combo, as stealing C’Thun with Sylvanas/Shield Slam or Entomb is an available counter play that can lead to your defeat.

Data Reaper Report - Shaman

Pinpingho brought a unique version of Mid-Range Shaman to the APAC championship. It was not too successful, though. His deck has a higher end curve, leaning more towards a value-centric play style, utilizing Ancestral Spirit on giant minions such as Earth Elemental while still possessing burst damage to finish the game. Another interesting choice in the Mid-Range Shaman archetype was Foot’s inclusion of Nexus Champion Saraad, instead of a second Thunder Bluff Valiant.

Meanwhile, Xixo’s Aggro Shaman variant continues to produce impressive results. RDU used the deck to win Dreamhack summer 2016 while Handsomeguy used it to win the APAC championship. It’s proving to be one of the most powerful decks in the game, based on its ladder and tournaments achievements. Pick it up while you can.

Data Reaper Report - WarlockZoo Warlock remains the dominant warlock archetype while Reno Warlock maintains a respectable presence in the Metagame. Taking a look at the APAC Spring Championship, Zoo and Reno (with one of the Reno Warlocks containing a C’Thun package) were played evenly.

At Dreamhack Summer, which had a very different format of nine rounds of Swiss and Last Hero Standing, five of the top 8 finishers had Zoo in their line-ups, while one player had a Reno N’Zoth variant (Hoej).

Interestingly, of all the Zoo builds, not one Leeroy was to be found. While Dreamhack is open to players from around the world, it mostly had players from the European region. It seems like EU prefers the Doomguard variant of Zoo, while NA as well as APAC hold a preference for Leeroy builds.

Data Reaper Report - Hunter

Mid-Range Hunter is continuing to look stronger and stronger with every passing week. With the continuing significant presence of Warrior in the Metagame, specifically C’Thun and Control Warrior, Mid-Range Hunter is in a very favorable position going forward.

RDU won Dreamhack Summer utilizing the archetype in his line-up. Much like Thijs’ European Championship list, his deck features Desert Camel, with the combination of injured Kvaldir. This package is strong against control decks, but less recommended if you’re facing a lot of aggressive decks on ladder that have scary one drops such as Zoo, Tempo Mage and Aggro Shaman.

At the APAC Championship, 6 out of the 8 participants had Mid Range Hunter in their line-ups, which made it the most popular archetype in the tournament. Interestingly though, the two players who did not include it, ended up making the finals!

This week we are featuring three of the most successful ladder lists we have experienced strong results with, all of them hitting top 10 legend on the NA server.

Data Reaper Report - Mage

Mage is a class on the rise with some interesting developments over the past week. RDU was victorious at Dreamhack Summer, utilizing his own brew of Tempo Mage which tops off at two Flamestrikes, Ragnaros the Firelord, and Yogg-Saron. The deck performed very well at the event, picking up wins solely off snowballing early leads. The Flamestrikes allow you to starve Zoo Warlock, an otherwise poor matchup, to the late game.

Meanwhile, our Mage expert, LBYS, ended up taking Freeze Mage to rank #5 legend after a large win streak. Freeze Mage has many flexible slots, and has seen significant card changes since the release of the expansion, something you never would have seen before. Being able to change builds to counter the expected Metagame on ladder and tournaments is very healthy for the game as it can reward you more for preparation.

Data Reaper Report - Druid

Druid has seen a slight decrease in play on ladder this week with the resurgence in Tempo Mage, which seems to be the flavor of the week. However, this doesn’t mean that Druid hasn’t found success on ladder. DacianWolf hit top 10 legend with his version of C’Thun Druid. His build has more anti-aggro options than SilentStorm’s build, which has been featured in previous weeks. DacianWolf includes a second Living Roots, Feral Rage, Mulch, and Ancient of War.

At the APAC Spring Championships, four players brought Druid, of which three were C’Thun variants and one was a Yogg Token variant. Handsomeguy, the winner of the tournament, brought a C’Thun Druid with some variations compared to other lists; most notable is a second Nourish for more ramping and draw options, two Klaxxi Amber-Weavers and a Beckoner of Evil. Similarly to DacianWolf’s, Raven Idols were cut, which might be an ongoing pattern.

At Dreamhack Summer, Fr0zen got second place using a Yogg Druid variant, created by Muzzy, which doesn’t focus on token synergy as much as J4CKIECHAN’s list featured previously. This build includes two Ancients of Wars, which allow the deck to better deal with aggressive decks, and an Emperor Thaurissan, to allow cheaper combos as well as getting the bigger minions out earlier.

Although Beast Druid has died down in popularity since its initial appearance, the deck has also found recent success on ladder with Thijs climbing from 1200 to top 20 on EU while playing it.

Data Reaper Report - Rogue

The two finalists of both the Asia Pacific and European Championships had Miracle Rogue in their lineups. With the recent tournament success of the archetype, it leaves us pondering whether Rogue can become just as dominant on ladder. Miracle Rogue is seeing more fine tunings, with most players opting for Thijs’ variant. The two innovations of the week come from Superjj at Dreamhack Summer, and Jako1910 at the APAC Championships.

Jako chose to remove Fan of Knives from his list, replacing them with Harrison Jones and Ragnaros, the Firelord. With the tournament consisting of only 8 players, it is much more possible to tweak your decks in a certain way based off the other players’ tendencies. Against Control decks, Fan of Knives is very often a worse Arcane Intellect, so if Jako anticipated a slower Metagame, his line of thinking makes sense.

SuperJJ also showed us a unique take on the archetype, removing 1 Conceal, 1 Cold Blood and 1 Sap to add Shadowstep, a second Earthen Ring Farseer and a Shiv. As we’ve discussed in previous weeks, the second copy of Conceal is arguably a ‘win more’ card. SuperJJ took that reasoning another step forward with the removal of the second Cold Blood as well. Shadowstep offers more flexibility than Cold Blood, while not reducing the amount of damage your deck can do with Leeroy. It provides you with the ability to have swingy turns in the early game against aggressive decks with Earthern Ring Farseer and Si7 Agent, allowing you to activate their battlecries twice while also “healing” them after a favorable trade.

Data Reaper Report - Priest

At the APAC Spring Championship as well as Dreamhack Summer, we got to see a lot of competent, competitive decks win games. We also saw some Priest play.

Jokes aside, N’Zoth Control has remained the most common Priest archetype over the last few weeks, but we wouldn’t be surprised to see C’Thun Priest make some appearances in the future. SuperJJ brought his version of C’Thun Priest to Dreamhack Summer and had some solid success with it, getting to the top 8 of the stacked tournament. SuperJJ seems to have a lot of faith in the deck, so it’s certainly worth monitoring in the comings weeks and seeing whether it ends up developing into the successful archetype Priest enthusiasts have been searching for.

At the APAC Spring Championship, we saw just one Priest build being brought to the tournament: Heisnotaxel’s N’Zoth Control Priest. This list is similar to JAB’s list that we have been featuring in recent weeks, in which Northshire Cleric is cut among other cards, in order to maintain value in the mid-game through active consistency.

With the N’Zoth and C’Thun variants, Priest Post-WoToG remains a class built around finishing games in fatigue. While this strategy can be successful in the appropriate Metagame, other control archetypes can seemingly do it a lot better, which makes Priest somewhat redundant.

Data Reaper Report - Paladin

Paladin had one bright spot this week: at Dreamhack Summer in Sweden, Ersee won a beautiful series 3-0 against Orange with a modern version of Secret Paladin.  His list looks similar to Jambre’s recent top 10 legend build, adding a Flame Juggler for the early game, replacing one Defender of Argus from the contested 4-mana slot, and using Ragnaros as a late game threat instead of Avenging Wrath.  Though his Secret Paladin had an impressive record in the tournament, Ersee did not make top 16. It will be interesting to see whether this is a one off occasion, or whether this deck can gain further traction on ladder in the coming weeks. Meanwhile, Control Paladin lists continue to stagnate, leading Paladin to secure its place at the bottom of the class rankings.

Data Reaper Report - Meta Breaker

This week we recommend Mid-Range Hunter. It is a deck that is very powerful against the most dominant class out there, Warrior. It generally struggles when the opponent is the aggressor and causes it to play off its optimal curve, which happens often against Zoo and Aggro Shaman, but even those matches are winnable. When Mid-Range Hunter is allowed to play its curve of minions, it’s extremely hard to stop for any deck

In addition, we have reasons to believe that towards the end of the season, the number of Warriors will only increase further, especially at legend ranks, which might give Mid-Range Hunter an even bigger inherent advantage against the field. Whether that will be the case though, remains to be seen.






Here are all the people that participated in bringing you this edition of the vS Data Reaper Report:

wwlosheavyjSchismrayc591demigodShooplbysVandoomHC015TzachiFenom lookitzjoeEndofDaysNaramo