Deck Guide: Fragment Demon Hunter by Ignatius
Hello everyone, I’m Ignatius, this is my 10th major contribution to this forum. (I most recently did a write-up on Token Druid, and in another lifetime, I have done previous write-ups on No-Trogg Shaman, Yogg Control Warrior, Yogg and Secret analysis data-grinds, and several others).
Today I would like to offer a few things I learned in climbing through legend ranks with Soul DH on both NA and Asia Servers this season.
I remember first seeing a decklist on Twitter, and a corresponding outrageous win rate. Sometimes an early meta breaker can do this. I saw a few others make similar posts, and I decided to take it for a spin.
The surprising thing was how my results with the deck in the early games were quite bad. The first 30-40 games or so I broke even, and I was scratching my head trying to figure out why the deck was so great. Usually when this happens, I get really excited, because it is probably indicative of needing to learn new things — MY FAVORITE PART OF HEARTHSTONE, being bad and getting better by learning.
I walked the deck through 4 different major iterations, and then 8 small tweaks to the final iteration (which dropped Magtheridon) — testing removing/adding the following,
EyeBeam, Magtheridon, Consume Magic, Polket, Wandmaker, Vulpera, Immolation Aura, Spectral Sight, Panthara, Shadoweaver, Battlefiend, Glaivebound, and Sneaky Delinquent. This was indicative of my second favorite part of the deck: how many cards can move in and out to bring about a new, fresh experience/analysis of the archetype.
It was through both grinding a ridiculous number of games of just this deck, alongside rotating cards and experiencing how the deck feels with changes, that I feel I am now piloting the deck to a reasonable level of mastery.
The Deck & Stats
In this entire season I had a score of 134 – 99, put it’s not just the most recent iteration.
On Asia Server in the last week, I held a 65% win rate over 87 games to climb from 6900 to 1200, including favorable win rates vs everything except for Demon Hunter. The list I settled on was 42-18 over this stretch (70%) and felt extremely strong and consistent when I played correctly. It was at this point that I was excited to do a write up and share both my enthusiasm and hopefully a few tips that will help others enjoy the deck as well.
Thoughts behind this list
The aspects of this list that I think are unique to others that have been successful are:
No Vulpera or Eyebeam, Yes Immolation and Wandmaker, and No Magtheridon. It may also be worth mentioning as I know it has been a point of discussion: Polkelt is outstanding. I should also concede that Glaivebound appears to be a really successful card in the deck, I just did not find out how it fit the greater whole as much as other cards in my own playing.
At this point I think we have enough stats where broader discussion is not essential, but as the deck still has a lot of iterations, it is worthwhile — at the very least — to explore the why’s of the stats that are available:
Vulp and Eyebeam: these two are documented over 17,000 top 1000 legend games as two of the worst cards (drawn winrate). I also felt that they were really poor in the first 50 games or so.
Magtheridon: just behind them is Magtheridon, but I think Magtheridon begs for a bit of discussion, because drawn winrate for cards like Mag (and Polkelt) can be a correlation/causation trap. That said, in my first 80 or so games with the deck, Mag just felt way more wrong than right. And, while Mag is strong against Druid, Mag is also a ridiculous liability against Priest (they steal it, and you can’t deal with it). I started to see way more Priest than Druid, and the swap felt great. By far my most successful list came as I dropped Magtheridon. Thought, I still wonder if I just use the card poorly because other players who are better than me insist it should stay.
Wandmaker and Panthara: these cards are just good. The 1-cost spell pool got a bit of a soft nerf, but even Double Jump and Felosophy found their moments, while Slice, Burn, and Consume were outstanding. Also, Wandmaker’s generation is an improvement over Vulpera almost always because your curve gets clunky with Vulp (1 turn later, and cards that cost more than 1).
Consume Magic: 1 copy of this card just seems great to me. When are you sad to spend 1 and draw 1 with this deck where you are almost always ahead or able to clear whatever is in front? Without Outcast, you can silence great targets vs Priest and Druid, late taunts as you’re racing a slower Rogue. This deck’s ability to draw means that you can actually find this card consistently in the matchups where it’s essential — I also hit it many times off the DJ from Wandmaker. If all this wasn’t enough, I found several games where I was really comfortable with a tempo Kayn Sunfury because I had Consume as a 1-mana backup to get through a taunt. Comfortable tempo Kayn is scary.
There are a few matchups where particularities in the mulligan make a world of difference. If you are just getting started, you can’t go too wrong keeping (in order):
Aldrachi Warblades (vs. aggro)
Shardshatter if you have Jailer/Shear in the right matchup
In slower matchups, if Skull or Spectral can be slotted into the left Outcast slot, keep
I’ll provide specific mulligan priorities per class below. PLEASE look at these, as the Winrates I had vs the 3 most common classes were largely because of mulligan choices (100% v mage, 86% v priest, and 75% v rogue)
Three Broad Tips for Improvement
One of the most not obvious things to me when I started playing this deck is that it is not always good to shuffle souls into your deck. I’ll deep dive this in several of the matchups. But generally, if your opponent is really aggressive (Hunter, Rogue): shuffle away, the passive Soul draw is likely to be fine. If your opponent is a bit slower but still fast (Warlock, Druid), I’d consider thinking of a line where you save the shuffle for when you need it. Most importantly, if your opponent is a lot slower (Priest) or is very unlikely to be threatening a lethal anytime soon (Mage), try really hard to save Soul shuffles if you can. More to follow.
Polkelt is an MVP in this deck when managed correctly. The easy first tip: realize that shuffling souls in after you Polkelt negates Polkelt. “Wait, why didn’t I draw my Skull!” (<– my first time realizing). The better tip: add up your turns, evaluate your opponent’s likely gameplan, and determine the best turn to play Polkelt. In a faster matchup, for instance, if you Polkelt on 4 and haven’t found Souls yet, you are likely setting your opponent up for an easy victory. In a slower matchup, if you play him 1 turn late and throw off the curve into Malicia and Skulls, the same result might be true.
Soulciologist is the highest drawn winrate card in the deck for a reason: if you plan it out in the right matchups, she is ridiculously uptempo or instantly presents a board that certain classes (like Rogue) just cannot deal with. And, with how much burst this deck can manage on turns 8 and 9, anything left behind from Malicia likely lead to lethal. All this said, it is a careful process to “manage” Malicia, similar to Polkelt. And, there is a tandem management hear, since Polkelt is an instant Malicia tutor. Again, don’t just shuffle Souls because you can, you might draw them and diminish the Malicia value for almost no gain.
I’m going to provide in-depth tips against the 3 most common classes I faced: Rogue, Mage, and Priest. This is in part because these are also the 3 where technique can make you significantly favored.
» Against Rogue
1. Immolation, 2. Blade Dance, 3. Jailer, 4. Aldrachi
ne huge advantage of this deck vs rogue is that it does not matter which rogue it is. You mostly play the same. I look HARD in the mulligan for cards that remove their early stealth minions — Immolation Aura and Blade Dance. If you have a Soul development (Jailer, Shear), then you can keep Shardshatter Mystic as well. The only card I will keep that is not these cards is Aldrachi Warblades.
Kill the early stealth minions. If you feel they are planning to play Grayheart, don’t worry about the value of your AOE, just get Spymistress off the board. It’s strange, but turn 2, 3, and 4 being Immolation, 1-damage Blade Dance, and then Shardshatter feels wrong, but you’ve probably won the game with that opening.
Get significant heal off of Aldrachi. A nice trick is to swing and then equip it, so you are removing something without using a charge, because you want to buff it up for big heals on later swings.
If you got out of turn 6 with 18+ health, you are probably on a winning path. Recognize as well that Rogue doesn’t answer Soulciologist efficiently. If you can plan to run her onto a board and go up-tempo, it’s another easy win.
Looking at my 4 losses over 16 games vs. rogue, in all 4 I did not find the early AOE. I recall in two of these games not digging hard enough for it (keeping Wandmaker / Panthara), and deeply regretting it when I lost.
» Against Mage
1. Sheer, Slice, or Chaos, 2. Jailer 3. Wandmaker 4. Panthara 5. Aldrachi
If I were to lose to a mage, something went unbelievably wrong in RNG (i.e. box), or I misplayed. Mage was 100% win for me when I got the hang of the deck, and each game felt like a blowout. However, there were some subtle things I did wrong vs mage in early games with the deck.
In the mulligan, prioritizing removal for Lab Partner is significant. The turn-1 partner into Cram Session is extremely powerful, but if you clear it with Shear, Chaos, or Twin Slice, you will be good to go. A good mage will realize the unfavorability of the matchup, so Tempo Chen, Firebrand, and early Apprentice should be anticipated.
Throughout the game, one thing you need to manage thoughtfully is when to put Souls into your deck. Just because you have a Jailer and 1 mana left, or you have a usable Soul Shear, it does not necessarily mean you should use it. Early games I lost were often because I shoved Souls in because I could, and it took awhile to finish the mage, and my synergies were exhausted in later turns from drawing Souls uselessly. Your health total is not a commonly valuable essential resource in this matchup, so the passive soul draw is REALLY bad.
In the late game, keep an eye out for face-freeze cards from the mage, and where you have an option to deal slightly more damage to the face before mage finds Frostbolt of Evocation/Cyclone, or worse, finds Deep Freeze, the better. Another late game tip is to hang onto a Blade Dance for the possible giants, which could be dumped out to try and race you to death.
» Against Priest
1. Sheer, Slice, or Chaos, 2. Wandmaker 3. Panthara 4. Keep Skull or Spectral in Outcast slot
Priest turned out to be one of my favorite matchups, because though it is favored, it is very easy to goof up and make it feel not favored. (I did not start feeling that it was significantly favored until I removed Mag, Eye Beams, and added Consume Magic).
If I have a removal for early Veilweaver in the mulligan, I will look hard for the 2-drops and try to prioritize getting a Skull onto the far-left side of my hand. I do NOT keep Jailer in mull vs. Priest.
There are two ways I could subtly screw up in this matchup: shoving Souls into my deck early, and not prioritizing every point of damage that comes from my face to the Priest’s face. It’s hard to explain, but one significant tip is that you don’t want to start “caring” about damaging the face until you are starting to snowball your damage past their healing, and until you have a line of sight to getting near the bottom of your deck. For this reason, you want to TRADE with your early minions (not hit face). I will often just let one minion sit on the board, so that the priest cannot use a single card to clear two minions. And most importantly, if I’m using Aldrachi, Marrow, Soulshard, Chaos, or Twin Slices to remove minions, I know I’m having to play inefficiently.
Another thing to mention is the value of baiting an early Apotheosis. Getting the Priest to 22-24 with minions, and letting Apotheosis come out so they go back to 28-30 is not bad if you haven’t used your weapon/face damage yet. There’s enough in your deck to obliterate them if you save it and send an avalanche over 4 turns in the late game.
I can’t say enough how bad it is to put Souls in early against Priest (like with mage, but moreover). Many victories came down to the bottom of my deck, and running out of Souls, or not having one available down the homestretch is devastating. Likewise, passively drawing them is almost always bad against Priest.
Honorable mentions for other classes
» Against Druid
I was happy with 50/50 vs. Druid, in part because I dropped Mag from my list (Mag performs best against Druid). Some key things that can really assist:
Don’t play unnecessary minions into Guardian Animals. Is it turn 5 and you have 1 mana left over for a Jailer on an empty board? Don’t play it. The Druid not being able to activate the 5/4 MONSTER is a huge advantage to you. AND, you can clear all of it with a good blade dance. You effectively neutralize Guardian Animals if done carefully.
Realize they have a lot of healing and taunt, and in the later turns, making the Druid feel like they might die is probably better than actually setting them up to die. If you over-play your damage and let the board go, they will heal and take tempo, almost guaranteeing your defeat.
» Against Warrior
I pretty much only saw Bomb Warrior, and it was hard to say whether it was favored. A few tips:
One thing that helps is tempoing out things like Shardshatter on empty boards. The Warriors feel a priority to trade it down which also involves them not smashing your face.
There is a huge gotcha vs. warrior, which is that the bombs reshuffle your deck when they go in. So if you do the traditional Polkelt expecting to smoke them with Skulls and Soulshard/Marrow, it’s probably not going to work. Have to find a different path to victory. I do not get excited about Polkelt in the mulligan or about drawing Polkelt early.
» Against Hunter
I would not be surprised if Soul DH is significantly favored vs. Face Hunter, I just didn’t see many of them. A few tips:
The mulligan priority changes vs Hunter, because if you can’t remove a few of their early drops, you’re likely to be fighting every turn to remove/heal and have a hard time recovering when needing to do both at the same time. Finding Shear, Slice, and Chaos are essential early.
Carefully play around Pressure plate. A lot of your minions are great to have back from Freezing Trap, and Explosive is mostly irrelevant because your minions have to trade. But, if you forget about Pressure and lose your Lapidary because you played Slice second, it will be unfortunate.
» Against Paladin
Paladin is probably the most unfavored matchup on paper. They heal and taunt at the same time. They do it for cheap. And they do it a lot. 🙂 However, you won’t squeak out a few wins without careful attention to what they do, and being ready to respond. A few tips:
Consume magic is going to be the MVP, and you want to make sure it hits either a Divine Shield taunt or a minion that receives the really dense paladin buffs. (If you end up using it on Goody Two Shield or an early minion that gets buffed, you will be in an unwinnable spot later). Watch out for Blessing of Authority, one of Soul DH’s biggest weaknesses is an extremely well-statted minion that sits on an empty board.
Blade Dance should be really good against Paladin, except that a lot of the time the minion you really want to hit with it efficiently also has Divine Shield (Devout Pupil and the 8/8 Guardian from Libram of Hope). It’s important to try to prepare to be able to bump the shield off and then Blade Dance — Immolation and Shardshatter can be great for this, or keeping a minion healthy for removing a shield.
It is one of two match-ups (Warrior included) where at a certain point you just have to blitz, and in holding back you might miss the small opportunity to achieve lethal. Find and embrace the moment — yes you might lose to Libram of Hope, but it’s often a you won’t win anyway if they have it.
- Fragment Demon Hunter #129 Legend – uberkaiserHS September 26, 2020
- Fragment Demon Hunter #216 Legend – Brownartist1 September 23, 2020
- Fragment Demon Hunter #13 Legend – MeatiHS September 21, 2020
- Fragment Demon Hunter #95 Legend – FoxmenHS September 19, 2020
- Fragment Demon Hunter #1 Legend – Rami94HS September 19, 2020
- Fragment Demon Hunter #64 Legend – KingVenomStream September 17, 2020
- Fragment Demon Hunter #96 Legend – VitorioRadke September 16, 2020
- Fragment Demon Hunter #162 Legend – VitorioRadke September 14, 2020
I have been blown away by what the HS creators have done since Descent of Dragons. 3 years ago, I took a sabbatical from the game for a few years, and there were clear signs of new direction when I decided to return about a year ago. I’m so glad I returned at the sight of those signs.
I see a few times daily a discussion about frustration over what I would contextualize, as just powerful cards. If there weren’t powerful cards, one of two results would be our game state: 1) the old decks would have just stayed and been out of balance, or 2) the game would be boring.
The creators have us in a place though, where every deck essentially breaks the game as we used to know it, but almost all the classes (sorry Shaman) are good, and multiple classes have several varied and viable archetypes. If all this wasn’t enough, the post-Scholomance meta appears to me to still be quite unsettled. WOW.
I say all this as preface to my perspective that this Soul DH is delightfully symptomatic of how good the state of the game is right now. It is dynamic, high-skill, extremely powerful, and has different techniques for each matchup that improve its capacity to succeed. It also just breaks the game sometimes (1-mana Marrowslicer, 2-mana Lapidary, 2-mana Twin Slices, Blade Dance for clear, and punch the face for 14 while playing a 5/5 … all of this in a calculated setup with Polkelt, are you kidding me!?) I just can’t get enough!
I hope some bit of information here can be food for thought or actual use during a game on ladder.
Thanks for the critiques, feedback, and discussion (looking forward to it). I’ll be planning to reply to every comment/question where input will be valuable.