Old Gods Card Reviews: Arena Edition, Part 1

Introduction

League of Explorers Post-Mortem

About four months ago, I did a review of League of Explorers cards in Arena. While it may be difficult for me to own up to my mistakes, I wanted to take a look back at those reviews before reviewing the Old Gods cards. In doing so, I hope to make these reviews better by examining what I didn’t do well enough in those reviews. Every Hearthstone commentator will whiff on cards every now and then, but it’s still important to look at those whiffs and learn from them.

First, the big trends:

I undervalued the Discover keyword. Like most people, I was excited by the possibilities of Discover, but I also felt that having to choose between 3 random cards would mean that you often would only have the illusion of choice. I called the balance between getting to choose the right card for the immediate situation and the risk of getting poor Discover choices “basically a wash.” I will admit now that I was completely wrong: Discover is not only super fun, it’s also really strong. The ability to pick a situational card that is relevant in the precise situation you’re in is well worth the occasional dud of a choice.
I said Warlock would get worse in LoE. While I was spot on with most of my class evaluations, I did miss the boat on Warlock. Part of that was undervaluing the Discover on dark-peddler (see the point above), but part of it, too, was not recognizing that an excellent class common more than offsets a terrible class common. It’s easy to see a card like curse-of-rafaam and think that it will offset the good that Dark Peddler does, but that ignores the ways Arena drafts play out. In general, you’ll pick every Dark Peddler offered you, but at least half the time you’re offered Curse of Rafaam, you’ll also be offered a good to great card anyway.
I rushed some of the card reviews. Churning out 8,000 words in a weekend about a set of cards that just got announced turns out to not be the greatest idea. This time around, I’m not only taking more time (thank you, Blizzard, for the leisurely release schedule), I’m also organizing this differently. Instead of organizing by class cards and neutrals, I’m organizing my reviews by tier: terrible, bad, average, good, great, and high variance. In doing so, I hope to force myself into thinking more deliberately about which tier each card belongs in and why.

Okay, what about the cards I missed on?

dark-peddler: undervalued curve flexibility and Discover
ethereal-conjurer: undervalued aggressive stats and Discover
sacred-trial: overvalued destroy effect
unearthed-raptor: overvalued Battlecry effect
wobbling-runts: overvalued cuteness factor
arch-thief-rafaam: didn’t realize how slow it would be or how easily opponents could play around the Discovered spell
reno-jackson: underestimated the fact that a full heal can be game-winning at times, even in Arena

Still with me? Excellent! I know all this can make me seem like I have no idea what I’m doing, but I still think I got most of the LoE cards spot-on. By thinking about the ones that I didn’t, though, I hope that I—and perhaps you by extension, dear reader—can get better at card valuation. With that out of the way, onward into the depths of darkness we go!

Terrible Cards

This tier includes all the cards you never want to see come up in an Arena draft, either because their stats or terrible or their effects are too situational. Examples include master-swordsmith, angry-chicken, and stoneskin-gargoyle.

renounce-darkness — 2-mana Warlock Epic spell, Replace your Hero Power and Warlock class cards with another class’s. The cards cost (1) less.

Renounce Darkness is certainly a unique card, but it’s one of many build-around-me Epic cards, like feign-death and lock-and-load, that don’t really have any chance of working in Arena. For Renounce Darkness to be worth it, your deck would need to be heavy on Warlock cards, and you would need to be able to play it early enough to reap the cost-reduction rewards, and your deck’s neutrals would need to be heavy enough to survive losing life-tap as a Hero Power. That’s extraordinarily unlikely to happen. Yes, you could Life Tap your way to card advantage, and then play Renounce Darkness like a sir-finley-mrrgglton without Discover. But then you just get a bunch of random class cards, which will usually be worse than cards you deliberately chose to be in your deck.

shadow-word-horror — 4-mana Priest Rare spell, Destroy all minions with 2 or less Attack.

Priest spells in Arena tend to be either really good—power-word-shield, shadow-word-pain, shadow-word-death—or really bad (I’m looking at you, power-word-glory). Unfortunately, Shadow Word: Horror fits into the latter category. The problems with this cards are numerous: it costs too much, it is unlikely to hit more than one enemy minion at a time, and it kills your own minions.

You see, the thing about Arena is that the vast majority of decks don’t just faceroll: they trade in the early game in order to establish board control. As such, you’re likely to face a few 1-attack and 2-attack minions, but if you face several, you’ll likely have already worked to deal with them by the time this card becomes relevant. For instance, think about all the times you’ve played stampeding-kodo. Of all those times, how many did the Kodo have two or more targets to hit while you had no 1-attack or 2-attack minions on your board? Even if you can think of a few times, the 4-mana clear doesn’t really let you develop much alongside it until Turn 8 or later. Yes, this will occasionally pull out a spectacular clear against a flood Warlock, but it’s a really bad card pretty much any other time.

tentacles-for-arms — 5-mana 2/2 Warrior Epic weapon, Deathrattle: Return this to your hand.

Tentacles for Arms is a classic example of a cool card that’s terribly costed. 5 mana for 4 damage is pretty terrible, but the fact that the damage is split up at a time when most minions have at least 5 health makes it even more terrible. Yes, this is a decent card in topdeck situations (assuming you don’t topdeck better weapons), but it really only offsets your opponent’s Hero Power if your opponent is a Paladin or Shaman. This won’t be as terrible as bolster, but it’ll be close.

Bad Cards

This tier includes cards that are not good but have some redeeming qualities, such as situationally useful cards like kobold-geomancer or cards with acceptable stats like murloc-warleader.

the-boogeymonster — 8-mana 6/7 Neutral Legendary minion, Whenever this minion attacks and kills another minion, gain +2/+2.

Okay, so The Boogeymonster has boulderfist-ogre stats for 2 extra mana. The effect better be good, then, right? Well, no. The problem with The Boogeymonster is that its effect is really specific and somehow significantly worse than the effect on comparable cards like kvaldir-raider, which can grow while attacking face and can grow the turn it comes out. The main thing that keeps this from being terrible is that big cards with high health (even overcosted big cards) tend to survive at least a turn, so The Boogeymonster growing is within the realm of possibility. Still bad, but at least it’s something.

cult-apothecary — 5-mana 4/4 Neutral Common minion, Battlecry: For each enemy minion, restore 2 Health to your hero.

Cult Apothecary suffers from the same problem as nightblade: a weak body for an effect that just doesn’t do enough in Arena. While this card would be a better antique-healbot when you’re facing a flood of minions, that isn’t likely to happen in Arena. Much more often, you’ll be healing for 2 or 4 health, making this a pretty bad card. As with all healing minions, Rogues and Warlocks should be happier to see this than other classes, but if I’m playing as either of those classes, I still take “back from the junk heap” over Cult Apothecary.

eater-of-secrets — 4-mana 2/4 Neutral Rare minion, Battlecry: Destroy all enemy Secrets. Gain +1/+1 for each.

If Eater of Secrets eats even a single secret, it’s a good card: 4-mana 3/5, your opponent discards a card. But how often will that happen? Let’s say that 50% of your games are against secret classes (a reasonable estimate in Arena). Of those, let’s say that 40% have secrets actually in their decks, since all three classes only have one high-tier common secret, and freezing-trap is the only great secret. That means that Eater of Secrets will successfully trigger in only 20% of your games, making it a 4-mana 2/4 80% of the time. A terrible card with a high upside means we’ve got ourselves a straight up bad card.

forbidden-healing — 0-mana Paladin Epic spell, Spend all your mana. Restore twice that much Health.

A lot of regular Arena players probably wish this card was a common, because it is not a good card. Healing-only cards in Arena tend to suffer from the fact that they are dead cards in many games (see flash-heal, holy-light, and healing-touch). Yes, this can be used as a massive heal, but it’s not like tree-of-life is a good Arena card. The flexibility does help, and the fact that you can miraculously win a game you were going to lose keeps it out of the terrible tier, but just barely.

scaled-nightmare — 6-mana 2/8 Neutral Epic Dragon, At the start of your turn, double this minion’s Attack.

Much like validated-doomsayer, Scaled Nightmare is a bad card on an even board or one you’re behind on, since its initial stats are so poor, but it gets scary pretty quickly if your opponent can’t handle it. However, unlike Validated Doomsayer, this takes two turns instead of one to really become a good card (an 8/7 on your second turn after playing it). There’s high upside here, but the 2 turns it takes to become scary and the fact that this comes out on Turn 6 will likely make it too slow.

In the end, then, Scaled Nightmare will be a lot like gurubashi-berzerker: a poorly-statted minion with the potential to grow quickly after you play it. If you’ve played a lot of Arena, you know that Gurubashi isn’t a good card because it is often so easy to deal with (and that’s on Turn 5). Buff considerations are not non-existent, but they’re also not relevant enough to elevate a card like djinni-of-zephyrs, so they’re not really relevant enough to elevate Scaled Nightmare, either.

spawn-of-nzoth — 3-mana 2/2 Neutral Common minion, Deathrattle: Give your minions +1/+1.

Spawn of N’Zoth is an awkward card: its effect is potentially powerful, but it will rarely ever be meaningful in the early game. If Spawn can manage to hit 2 other minions with its Deathrattle, it will be good, but that’s unrealistic, especially if you just played a 3-mana 2/2 minion, which is a terrible stat-line. This is just going to end up being too weak of an effect in the early game and too poor of a body in the late game. This might have been an interesting effect on an anubisath-sentinel cost and body, but on a 3-mana 2/2, it’s a dud.

tentacle-of-nzoth — 1-mana 1/1 Neutral Common minion, Deathrattle: Deal 1 damage to all minions.

Tentacle of N’Zoth is basically a delayed whirlwind. Whirlwind itself is bad-to-average, but Tentacle loses the ability to trigger immediately and isn’t in a class that actually benefits from self-damage. Yes, it could allow you to deal 2 damage to a single minion by running it into a minion, but that will generally mean you’re already in a good position. Like explosive-sheep, this will be a bad Arena card.

validated-doomsayer — 5-mana 0/7 Neutral Epic minion, At the start of your turn, set this minion’s Attack to 7.

Validated Doomsayer is a polarizing card: played when you don’t have board control, it’s terrible, but played when you do have board control, it’s a 5-mana 7/7 that’s vulnerable to silence but resistant to aldor-peacekeeper and keeper-of-uldaman.

Let’s consider a middle ground, then: your opponent has a 4-drop on the board, say a 3/5 or a 4/5. In effect, then, you’re getting a 5-mana 7/3 or 7/4, or salty-dog. Salty Dog is a bad Arena card. Validated Doomsayer has a higher ceiling, but his floor is basically bolf-ramshield, so they even out to bad in the end.

Average Cards

This tier includes all the cards that you’re neither upset nor happy to see in an Arena draft. Think about cards like bloodfen-raptor or djinni-of-zephyrs.

doom — 10-mana Warlock Epic spell, Destroy all minions. Draw a card for each.

Well, this is certainly flashy. twisting-nether’s bigger brother, DOOM! is a decent Arena card. Unlike Nether, it can’t be played on Turn 8 or Turn 9. However, it does allow you to refuel without spending a ton of health, and Warlock’s health pool tends to be low in the late game. The inflexibility of DOOM! makes it slightly worse than Twisting Nether, but it’s still a reasonable Arena card.

faceless-shambler — 4-mana 1/1 Neutral Epic minion, Taunt. Battlecry: Copy a friendly minion’s Attack and Health.

As with most cards with variable stats, to evaluate Faceless Shambler, you have to ask yourself: how big does this need to be to be good? In general, 8 stat points (such as a 4/4) is enough to make it a 4-mana value (the weakness to silence is only a minor consideration in Arena). Go over 8 stat points and you have a good card, especially because of the Taunt.

Given that it needs to generally copy at least 8 stat points to be reasonable, Faceless Shambler is not going to be a Turn 4 play. However, played on Turn 5 or Turn 6 if one of your minions stuck around, or played on Turns 8, 9, or later alongside a big minion, this card can be good. Unfortunately, Turns 8, 9, or later are not the times when you often need a Taunt, and this card suffers from the same problem as many others in this expansion: it isn’t playable unless you have at least some semblance of a board. It can, however, help swing a close fight for the board, which makes it a decent card.

hallazeal-the-ascended — 5-mana 4/6 Shaman Legendary minion, Whenever your spells deal damage, restore that much Health to your hero.

Hallazeal reminds me of djinni-of-zephyrs: a reasonable stat-line with an effect that is generally negligible, but will occasionally win you the game. Hallazeal is best played with AoE, but that’s something Shaman historically struggles with in Arena. Even in a more likely scenario, such as Hallazeal plus crackle, you’re still only getting a slight boost over a vanilla 5-mana 4/6. You’re never thrilled to pick spiteful-smith in Mage, but you’re also never broken up about it, either.

herald-volazj — 6-mana 5/5 Priest Legendary minion, Battlecry: Summon a 1/1 copy of each of your other minions.

Let’s start with the obvious: Herald Volazj is a bad card played on an empty board. Okay, well what about with 1 minion on the board? At that point, Herald is basically a more expensive silver-hand-knight. 2 minions? Now we’re talking: 7/7 in stats split up over three bodies is a pretty good 6-mana play.

Occasionally, Herald will copy minions with meaningful abilities, but they will almost always be neutral minions, since the only particularly relevant Priest minions here are dark-cultist and northshire-cleric. The chances of doing that in Arena are fairly low, but that does give the card a very slight boost. Ultimately, I’m putting Herald in this tier based on the most likely scenario for Arena Priests: one minion on the board. It can easily be bad or good, depending on your deck and board, but it will usually end up being mediocre.

infested-tauren — 4-mana 2/3 Neutral Common minion, Taunt. Deathrattle: Summon a 2/2 Slime.

Infested Tauren is an interesting case, since its stats total up to chillwind-yeti levels, and sticky cards are generally preferable to ones with only a single body. The problem with Infested Tauren, though, is that both bodies are small, and they come at a time in the game when bigger bodies become more prevalent. That said, Infested Tauren will still give you 4 attack, a useful thing for a 4-drop, and it can’t really be traded into by lower drops easily. A smaller sludge-belcher this is not, but it’s still a decent Arena card.

polluted-hoarder — 4-mana 4/2 Neutral Common minion, Deathrattle: Draw a card.

Polluted Hoarder makes it into at least the good tier for card puns, but as an Arena card, it falls short of its uncorrupted version, loot-hoarder. With Loot Hoarder, you’re generally okay if your opponent pings it off, because they just spent 2 mana to handle your 2-mana card. However, there will be many times that your opponent can gain a huge tempo swing by killing your Polluted Hoarder with backstab, argent-lance, or arcane-shot. When your opponent can’t remove Polluted Hoarder that easily, it will be a good card, and it will certainly be better in topdeck situations than Loot Hoarder, but the fact that it’s pretty rough to play this on curve definitely makes it less desirable than its un-evil twin.

stand-against-darkness — 5-mana Paladin Common spell, Summon five 1/1 Silver Hand Recruits.

There is absolutely no way that Stand Against Darkness will be even half as good as muster-for-battle—the cost is high enough that the guys are easier to deal with and the stats are underwhelming—but it will be a functional Arena card. In general, splitting up stats into multiple bodies is good for Arena, and board clears are relatively rare, as well. fan-of-knives is punishing, but most classes will struggle to clear five Silver Hand Recruits, leaving you the ability to split up your damage however you want to.

Buff considerations are also important: you can pretty much guarantee a guy for blessing-of-kings when you play Stand Against Darkness. Paladins can struggle to deal damage when they’re locked out of the board, and Stand Against Darkness creates a board that your opponent pretty much needs AoE in order to clear. Finally, the fact that this costs 5 mana is huge, since it can be played alongside a 5-mana minion in the late game. I wouldn’t choose Stand Against Darkness highly, but I also wouldn’t be upset to have one in my deck.

Good Cards

These cards aren’t slouches, but they also aren’t the cream of the crop. ogre-brute, flame-juggler, and assassinate fit here.

corrupted-healbot — 5-mana 6/6 Neutral Rare minion, Deathrattle: Restore 8 Health to the enemy hero.

Corrupted Healbot joins a stable of solid Arena cards that trade a drawback for increased stats, such as ogre-brute and fel-reaver. Some will compare this to zombie-chow, but it’s actually much more comparable to pit-fighter. The 1 extra attack will not always matter, but the 8 healing drawback also won’t matter much of the time.

One of the reasons zombie-chow can be a dead draw in the late game is that it doesn’t do enough to offset the amount of healing: it would need to live and attack 3 times to make up its healing in damage. Corrupted Healbot is more okay, though, since even one attack on your opponent’s face makes this close to even. Played in the mid game, when the extra health probably won’t matter, since you’re still likely fighting for board control, this is a great card. Played in the late game, the drawback is pretty easy to recoup. Perhaps the only class I wouldn’t want to take this in is Hunter, since there are plenty of games where Hunters completely lose the board and win solely with steady-shot.

forbidden-flame — 0-mana Mage Epic spell, Spend all your mana. Deal that much damage to a minion.

Forbidden Flame is perhaps the most straightforward of the “Forbidden” cards introduced in Old Gods: you spend whatever mana you haven’t spent this turn and hopefully kill one of your opponent’s minions. The fact that this can’t go face may be a big deal in Constructed, but in Arena, it’s really not that big of a deal.

So, why do I have this card listed as good? As with the other “Forbidden” cards, Forbidden Flame sacrifices power for flexibility. That flexibility, though, will definitely make up for the decreased power of the card in this case. There are plenty of turns where you float a mana or your opponent plays a card with equal to or less health than its mana cost, making this at least equal in terms of tempo. A card like forgotten-torch is pretty solid, even if you don’t draw the roaring-torch, and Forbidden Flame is like Forgotten Torch on steroids. The only way I don’t want to take this card is if my deck is already full of more efficient and less flexible removal like frostbolt and flame-lance.

forbidden-shaping — 0-mana Priest Epic spell, Spend all your mana. Summon a random minion that costs that much.

Okay, so Forbidden Flame is pretty straightforward to evaluate, but Forbidden Shaping is much harder. An enterprising redditor put together stats on Forbidden Shaping, and it turns out that at most mana costs, you sacrifice 0.5-1 point in stats (e.g., 2-mana averages out to ~2/2.5). Those stats are underwhelming, but the flexibility of Forbidden Shaping is even better than Forbidden Flame, since it doesn’t require you to react to your opponent’s play. Missing a drop? Play Forbidden Shaping. Need a big minion? Wait until the late game and play Forbidden Shaping.

The one thing that holds Forbidden Shaping back from being great is the RNG factor. There will be times when you summon a really fantastic minion or exactly the one you need, but there will also be plenty of times when you summon a terrible minion. Not being able to plan your turn around the results will hurt. Also, good Arena players tend not to like RNG-heavy effects (see spellslinger), since they increase variance, and consistency is key in doing well in Arena. Still, Forbidden Shaping should be consistently good, even if it occasionally whiffs or hits big time.

giant-sandworm — 8-mana 8/8 Hunter Epic Beast, Whenever this minion kills another minion, it may attack again.

This card is an interesting case, since it’s something we haven’t seen in Hearthstone before. If you just attack face with Giant Sand Worm, it basically has the mana-to-stats ratio of something like war-golem, a functional but not particularly good card for Hunters. However, there are two other common scenarios that are much more appealing. First, you can attack into a minion and then attack face, basically making this an 8/8 with Windfury. Or, you can attack a bunch of smaller things until this dies, making almost like an icehowl + v-07-tr-0n combo (minus Charge). The flexibility here is huge, and in a mode where most classes struggle to find enough hard removal, Giant Sand Worm will be a good card. The fact that it’s in Hunter, a class that doesn’t usually want to play the late game, is the only thing holding it back.

hogger-doom-of-elwynn — 7-mana 6/6 Neutral Legendary minion, Whenever this minion takes damage, summon a 2/2 Gnoll with Taunt.

Hogger, Doom of Elwynn will be a strong Arena card, especially in a faster deck, when you can control where it attacks. Yes, there will be times when your opponent can simply fireball Hogger, Doom of Elwynn or attack a boulderfist-ogre into it, but in most Arena games, it will trigger twice, giving you 10/10 in stats. The Taunt keyword is also very meaningful here: as we saw with obsidian-destroyer, tokens with Taunt can be very tricky to get around. Hogger, Doom of Elwynn cannot be played as a desperation Taunt like Obsidian Destroyer, but the spawned tokens also can’t be pinged like the 1/1 Scarabs can. Hogger 2.0 could easily be less powerful if Mages take back over the Arena throne in the new meta, since Mages have lots of good answers for this card. Even then, though, most classes won’t have efficient answers, making it a value powerhouse that also helps slightly with survivability.

mukla-tyrant-of-the-vale — 6-mana 5/5 Neutral Legendary minion, Battlecry: Add 2 Bananas to your hand.

Mukla 2.0 is actually quite comparable to grand-crusader: the 5/5 body on Turn 6 is not bad, and the card draw makes it a good card. bananas are generally worth about half a card each, so this amounts to something like a single card draw. While you will occasionally get lots of spell synergies, those aren’t much of a consideration in Arena.

Another way to think about Mukla, Tyrant of the Vale is as a larger shattered-sun-cleric. Played on Turn 8, it’s basically a 7/7 with 2/2 of those stats split up however you want and with initiative. That’s a good—if unspectacular—card that I would be happy to see in most decks.

nzoths-first-mate — 1-mana 1/1 Warrior Common Pirate, Battlecry: Equip a 1/3 Rusty Hook.

Some have taken to calling N’zoth’s First Mate a mini-muster-for-battle, and that’s actually a much more valid comparison than stand-against-darkness/Muster. Perhaps the best comparison point to First Mate, though, is lights-justice. This card basically gives you an immediate ping, two pings the next turn, and another ping after that. In a class that struggles to find pings in Arena, this is a good thing. The only thing that holds this back is that many of Warrior’s power cards are weapons, and you’re somewhat likely to overwrite at least one charge of the Rusty Hook with another weapon. Value this card higher in a deck without many weapons and lower in a deck with a couple of fiery-war-axes.

ragnaros-lightlord — 8-mana 8/8 Paladin Legendary minion, At the end of your turn, restore 8 Health to a damaged friendly character.

lightwell is a terrible Arena card, but it turns out that if you make its ability better and put it on an average body, you end up with a solid card. Ragnaros, Lightlord will not be great, since the healing will often be irrelevant, and even when it is relevant, it might just accidentally heal your damaged 2/3. However, lesser-heal is one of the best late-game Hero Powers in Arena, and Ragnaros, Lightlord can function as a random Lesser Heal that you can still control to an extent. The fact that Ragnaros, Lightlord will often heal himself helps a lot, too.

In some ways, this is a very flexible card: it can vary depending on the situation between a bigger antique-healbot, a bigger stoneskin-gargoyle, and a big guy with a free Lesser Heal attached. Of course, being able to choose between these scenarios requires you to control the board such that you can control Ragnaros, Lightlord’s ability, but that’s not entirely unrealistic, especially in the late game when you likely won’t have a ton of damaged smaller minions.

undercity-huckster — 2-mana 2/2 Rogue Rare minion, Battlecry: Add a random class card to your hand (from your opponent’s class).

This card is basically like an upgraded loot-hoarder, a good card. Moreover, it has great synergy with unearthed-raptor. Why isn’t this in the great tier? Two reasons: first, as burgle has shown, getting a random class card from your opponent’s class isn’t quite as good as drawing a card you purposefully chose for your deck, since it can get you eye-for-an-eye and other trash cards. Second, the extra health doesn’t usually matter: in the early game, attack matters more than health, and you’re perfectly happy if your opponent spends 2 mana to Hero Power down your Loot Hoarder. The extra health only really matters in the late game, then, and it should roughly offset the fact that the Deathrattle is slightly worse than drawing a card.

xaril-poisoned-mind — 4-mana 3/2 Rogue Legendary minion, Battlecry and Deathrattle: Add a random Toxin card to your hand.

I love the idea of Xaril, only the second combination Battlecry and Deathrattle after toshley. Like Spare Parts, the Toxin cards that Xaril gives you are pretty variable, though: bloodthistle-toxin and fadeleaf-toxin aren’t very good, kingsblood-toxin is solid, and firebloom-toxin and briarthorn-toxin are great. The 1-mana cost of all the Toxins is pretty important, since it allows you to trigger Combos, though those aren’t incredibly common in Arena. Also, it’s pretty common to float a mana, so a card like kingsblood-toxin is pretty easy to play.

Given all these things, I can see an argument for putting Xaril in the great tier. That said, the chance to get weak Toxins and the fact that this card is best later in the game, when the Toxins aren’t as relevant, holds Xaril back.

Great Cards

Some cards you pick almost every time you see them, and those fit in this tier. muster-for-battle might break this tier system, but piloted-shredder, north-sea-kraken, and bomb-lobber certainly go here.

Nothing made it into this tier for this installation, but check back for Part 2 to see if there are any great Arena cards there.

Draft-Dependent Cards

Some cards depend heavily on the deck you’re drafting and can range from terrible all the way up to great with different decks. Build-around-me cards like bloodlust or brann-bronzebeard go in this tier.

infest — 3-mana Hunter Rare spell, Give your minions “Deathrattle: Add a random Beast to your hand.”

Much like soul-of-the-forest, the strength of Infest depends very heavily on your ability to have multiple minions on the board. At 1 mana cheaper, Infest is more flexible than Soul, and it combos incredibly well with unleash-the-hounds. In a deck with a couple of Unleashes, this is basically ball-of-spiders with Charge, which is a great card.

Infest is mediocre when you have 2 minions on the board (basically a more random arcane-intellect) and good when you have 3 minions on the board. Many Arena Hunter decks play extremely fast, and Infest is a good way to refuel without having to spend too much mana to do so. I expect this to be a solid choice in most Hunter decks, but it will really shine if you get a couple of haunted-creepers and Unleashes.

mark-of-yshaarj — 2-mana Druid Common spell, Give a minion +2/+2. If it’s a Beast, draw a card.

Mark of Y’Shaarj is pretty clearly comparable to mark-of-the-wild, trading guaranteed Taunt for conditional card draw. Obviously, card draw is better than Taunt, and it does help fix one of the main problems of Mark of the Wild and other smaller buffs, which is that they often aren’t worth a full card. Druid does have a ton of desirable class beasts—druid-of-the-saber, druid-of-the-flame, mounted-raptor, savage-combatant, and druid-of-the-claw are all good to great—as well as beast synergy support, especially wildwalker.

The strength of Mark of Y’Shaarj will depend somewhat heavily on your deck construction: if you have 4-6 beast minions, you can likely reliably trigger the draw condition. Fewer than that, though, and you will probably struggle to trigger the condition, making this card worse than Mark of the Wild. In a deck with even more beasts than 6 (especially early game beasts), Mark of Y’Shaarj should be good to great.

nzoth-the-corruptor — 10-mana 5/7 Neutral Legendary minion, Battlecry: Summon your Deathrattle minions that died this game.

Let me be clear: in the vast majority of Arena decks, N’Zoth is going to be either terrible or bad. However, occasionally you manage to draft a half dozen or more Deathrattle cards, and in these cases, it could be good or even great. For instance, I recently drafted a Rogue deck with 2 loot-hoarders, a haunted-creeper, a piloted-shredder, and sneeds-old-shredder. If N’Zoth had come up in that deck, it would’ve generated a massive and sticky board with one play. In other words, I would absolutely not pick N’Zoth early in a draft, but if you have a Deathrattle heavy deck already, you could do worse than N’Zoth as a finisher.

steward-of-darkshire — 3-mana 3/3 Paladin Rare minion, Whenever you summon a 1-Health minion, give it Divine Shield.

I really love what Blizzard has done with Steward of Darkshire, because it seems that they’ve managed to create a really interesting and powerful Paladin card without making it toxic for Arena. Let’s start with the “always there” effect of this card: it basically upgrades your hero power to an argent-squire, making it like a slightly-less-good silver-hand-regent (you generally would rather have the two separate bodies than the single body with Divine Shield). Silver Hand Regent is solidly in the good tier, and I would put Steward’s base level there, as well.

However, there are a decent number of commonly-played Arena cards that summon 1-health minions. Let’s take a look at them, excluding terrible cards like magma-rager and ones with Divine Shield already: abusive-sergeant, bluegill-warrior, clockwork-gnome, dragonling-mechanic, elven-archer, gnomish-experimenter, harvest-golem, haunted-creeper, imp-master, jeweled-scarab, leper-gnome, loot-hoarder, murloc-tidehunter, muster-for-battle, noble-sacrifice, novice-engineer, razorfen-hunter, southsea-deckhand, stand-against-darkness, twilight-drake, violet-teacher, wolfrider, worgen-infiltrator, young-priestess.

Not all of those cards are premium cards, but some of them are, especially Abusive, Harvest Golem, Haunted Creeper, Loot Hoarder, Muster, and Worgen. The strength of Steward will vary based on the number of these cards in your deck. Like hobgoblin, you generally can’t afford to choose a ton of these cards unless you’re playing Warlock, but even a single Muster for Battle will make Steward of Darkshire meaningfully better than it already is.

How much better? Well, that is yet to be seen, but the effect is tailor made for Paladins in Arena. Despite keeper-of-uldaman, Paladins still struggle with playing off the board in Arena, and one of the ways Paladins make it so that they can play on the board is Divine Shield. Moreover, Divine Shield is often at its best on minions with low health and higher attack, so these minions are precisely the ones you would want to target anyway. All that said, there just aren’t enough good-to-great 1-health minions for this card to take over the Arena like keeper-of-uldaman has. In that way, I actually think of this card as a lot like mysterious-challenger: your Arena opponent might occasionally pull off a spectacular combo, but most of the time it will just be a solid card that feels very fair and balanced (in Arena: don’t hate me for calling MC balanced, Constructed friends).

Closing

Thanks for reading, and check back soon for more Arena card reviews!