The intricacies of Design Space
Sumadin takes a look at one of the most complex phenomenons in card balance: Design space, and what it has meant in the past and present.
It should be no surprise to readers of my articles that I like to discuss Hearthstone balance. Predicting which cards end up nerfed or sent on the standard rotation is something I do annually with my serries “Predicting the Standard”.
But when discussing problematic cards, few things are as hard to describe as design space. When a card is a problem, not because it is overpowered but because it limits the potential of other cards too much.
While the topic is abstract, game director Ben Brode has written a great piece about it on reddit and you should totally check it out.
Today I would like to share my own understanding about design space. One of the reasons I believe it can be so hard to get to know about design space is that players are never met with most of the cases of limited design space. Ordinary players don’t get to see what a neutral dreadsteed would do in a world with unnerfed Warsong commander. Nor will most of them understand why the old Master of Disguise lead to Animated Armor being a mage-exclusive card. To be fair, neither do I entirely. And these examples are just the ones Blizzard was kind enough to tell us about.
In this article I will take a look at past cases of design space problems and try to give my view on what design space problems lead to the current problems we see.
Design space in the past: Starving Buzzard
Most cases of design space problems arrives when there are too many conditions on balance in the card pool. Too many cases of cards that should be problematic but due to some condition aren’t, because, while that condition might be upheld for a time, it is very easy for new releases to break it. One card exemplifies this better than anything else.
When talking about design space, no card in the history of Hearthstone tells us more than Starving Buzzard.
Starving Buzzard at the release of Hearthstone was absolutely nonsense. A 2/1 for 2 mana with the same contentious draw effect (it was even 2/2 in the beta). It is probably the most broken card that has ever made it to the release of Hearthstone, which still only makes it the second-most imbalanced bird to be released in a TCG. Birds, always be wary of the birds.
Buzzard's design was so strong, something else had to give. That had to be the cards it worked with, and sure enough at the release of Hearthstone, beasts were really lackluster. They were mostly vanilla minions with poor stat distribution at the higher mana-cost with the sole exception of Savannah Highmane. Unleash the Hounds also gave very solid use to the Buzzard.
All of this made for a mostly balanced setup. Hunters didn't really fit too much in the meta at release. Their beast support was broken, but their beasts sucked. The variety made for some hefty inconsistency. All Blizzard needed to maintain this was to remember how powerful Buzzard was. Remember that it was only balanced on the notion that beasts sucked and that they should not release too many powerful beasts, especially at the lower manacosts and then...
Yeah, file that under “stuff that didn't happen”. In the very first expansion of Hearthstone, two cards broke monumentally with this condition. Webspinner, a beast so powerful it would have seen use by every class in the game, had it been neutral, and Haunted Creeper, a beast so powerful it DID see use by every other class in the game. Hunter instantly rose to the top. There was also the whole farce with Undertaker, but even seven months after its nerf, Hunter was still top tier in the meta. Playing beasts was no longer a tempo loss. The condition of Starving Buzzard was no longer a liability. The design space had been breached.
This is probably the most powerful instance of Design space there ever was, and it is also crystal clear what the problem was. The buzzard’s design was not sustainable. Most of the time, things are not as clear.
Design space in the present: The Caverns Below
The next example I want to talk about is the deck that recently saw a nerf in a patch, Quest rogue.
Quest rogue as a deck was not broken in terms of numbers, but it falls into the very elusive deck category that is “Anti-control.” It is easily the most hated deck archetype, even though we don’t see it a lot. I think the first real example of this kind of deck was priest in the beta when mind control was at 8 mana. Priests were still getting squashed left and right, by aggro, ramp druids and 4 attack minions. But people still had a fierce hatred of mind control, because you would lose every time things went too slow. People don’t want to be forced to play aggro decks. Aggro decks are hated by a lot of the community already and the only that gathers more hate is decks that forces you to play them.
Quest Rogue was never going to have much of a fanfare
Quest rogue, as such, was never going to have much of a fanfare. This is a case of designing for a fun experience rather than the designing for balance, something we also saw with Patron Warrior.
Looking at the meta game immediately after the change, hitting the quest requirements has worked. But from the context of design space it would have been more relevant to look for past cards that have been repeat offenders and would likely cause problems again with future releases.
So which cards in rogue fulfils this? Well if you ask pros, a lot of them point to Preparation at this point. I think this is a trap. It is true the card is powerful, and was powerful in quest rogue, but it does not limit design, not in the way we typically see with cards such as Starving buzzard.
It is a matter of design reach. Buzzard's design reached out to all beasts, Hunters own and neutral. This meant neutral beasts basically had to be completely unplayable for ordinary mortal classes who didn’t have godtier beast support. It is balanceable design, but it isn’t futureproof. It puts too many constraints on the neutral pool of beasts. Like how would we ever see the Un’Goro expansion with this condition?
Preparation reaches out to all spells rogue has, but all of this is still confined within a single class. We don’t have neutral spells. It is an isolated reach, unlikely to cause problems since Rogue is intended to be balanced as a single package. If some spell is too cheap at the discount level, one can just bring the original cost up.
No the card I think is the biggest design prohibiter in rogue is Shadowstep.
Shadowstep reaches out to minions, which means every neutral minion has to have its balance checked up against this. Most of the time this is not a problem, but back when Miracle Rogue was at its peak this was the primary win condition together with the then 4 mana Leeroy Jenkins or Arcane golem. These cards were also seeing play in Face Hunter, so their strength was not limited to Miracle Rogue. But none abused their power as much as Miracle Rogue did, and soon enough these cards would be nerfed.
With the quest, rogue once again got access to cheap but high attack charge minions. And Shadowstep was a core component in getting the quest done. Once again the charge minions, or rather, the source of the high attack charge minions was nerfed.
Shadowstep as a spell is ruthlessly efficient with certain targets. Costing zero mana and duplicating the immediate impact of a minion for a two mana discount is very powerful. Think of Novice Engineer. Combined with that minion, shadowstep essentially becomes a 0 mana, draw a card. Such a spell could NEVER be printed out on its own. This wasn’t a problem before of course because playing novice engineer has historically been a liability, and rogue has better ways of cycling cards. But with the quest, all of this now contributed to victory. A victory that could be secured as soon as turn four.
Nerfing Preperation would solve way too much.
Nerfing preperation would of course still have solved a lot of the problem cases with this quest. The problem with that is that it would solve way too much. It would ruin the basis on which every spell released in rogue so far has been balanced. Most of the spells in rogue would be unusable, in standard and in wild.
I believe the better change would have been to give Shadowstep a manacost of 1, or send it off to wild. It is not a spell that is ever going to be part of “honest” Hearthstone, because it forgoes card advantage too much and can’t be used for ordinary tempo either. Every time the spell sees serious use, it has been with a problematic combo of sorts. And at 0 mana, it is just way too good for those.
Letting bounce effects for rogue only be “bad” cards like Gadgetzan Ferryman or Pandaren Brewmaster would open up design space for powerful cards that work with them: like The Caverns Below. But it would probably also allow for a time where we see charge minions printed again. It is still a mechanic that has to respect taunt, which in the age of burn mages should be appreciated. Blizzard themselves might have been burned too much by this mechanic though. Shadowstep might also just appear too much as balanceable design to be hit.
But that is the nature of Design Space. It can be tricky. It is not always immediately clear what cards put up too many design conditions to be balanced, what brings too much of an edge and what has too much potential.
Design Space in the future: Alexstrasza
Burn Mage is, in more than one way, the hottest deck after the release of Un’Goro. As is the case with every deck that rises to the top, pros are voicing their opinion on any problem card. For any sort of Mage, this has been Ice Block for the longest time. This can also be seen in the recent AMA with Ben Brode.
Now, Ice Block is powerful. No doubt about that. Ice block most certainly prohibits design space to an extent. But I find it still to be the least of the problem cards in mage. And Alexstrasza is a much bigger issue that deserves more attention.
I believe Alexstrasza is by far the biggest problem card in the mage decks, made even worse by her being neutral and having a past in decks like Reno Warlock or Charge warrior. Lets just for starters sit down and look at the raw numbers of Alexstrasza. She leaves an 8/8 body and deals up to 15 points of damage to the enemy.
‘But that is only at full health’
Ignore that. Also for now ignore that she has defensive utility on top of all of this. Just look at the absurd fact where we accept that a neutral legendary deals 50% more damage for 9 mana, than a class spell can be allowed to deal at 10. While placing an 8/8 body on the board.
This shouldn’t even be a design space problem, just a straight up overpowered card. But people seem so intent on calling out Ice Block as the problem. When I don’t do that, it is because I don’t believe tempo should be the balancing factor for burn decks. It is wrong to try to keep burn mage a deck, that you just have to kill before they will inevitably kill you. We will just end up eventually with the same problem as Quest Rogue, an anti-control deck forcing people to play aggressive decks.
Instead, the limiting factor for burn mages should be card advantage. How much damage they can fit in compared to how much healing you have available. Healing is supposed to be a powerful tool. Not in design, just in application. This is because you don’t need to heal up every point of damage from a burn deck, only enough so that they can’t deal that final point of damage that finishes the game.
If we look at the current control mage, one can count the amount of damage that comes from burn spells, 38 in total. This should be easy to handle. First of all the mage is never getting away with putting all of that damage into the enemy hero, and second that would only demand 9 points of healing to stop (+ a few for the pings from Hero power). The basic set alone contains more healing than that and Warrior and Priest should have this one in the bag with their Hero power.
That is, until we add Alexstrasza to the equation and the total dmg becomes 53. And for this equation we have also ignored any dmg the that might be dealt with the minions of the mage. Now it is suddenly much more clear how this is a tier one deck.
It gets even worse, when we look at how Alexstrasza interacts with some of the other intended counters to Burn damage. Lets take the priest Quest reward, Amara, Warden of Hope. Now currently most deathrattles are bad, but it seems the Frozen Throne might try to fix that. How does leave Quest Priest as a burn mage counter?
Probably just as bad, because the absurd design of Alexstrasza makes her more powerful, the more you do to stop her. Raise your health to 40 and she will just do 10 more damage, and the end result is the same. This is a different side of design space, one that restricts effectiveness, making counters too weak, instead of being too strong with synergy. Thanks to Alexstrasza all future cards that increase max health are useless as far as counteracting Burn decks.
This is really not how it should be. Raising the max health of the hero to 40 should just be game over in a burn matchup, the max health is above all the spells in the deck combined. But instead this neutral evergreen card stops any idea of using max health as the answer. She also acts as a soft counter for mass heal cards like Forbidden Healing, leaving only the somewhat class exclusive mechanic, armor as the only way. At this point it just worth the question: Do we really want to see a ton of neutral armor cards just because Alexstrasza can’t figure out how to respect healthtotals?
For the record, I still believe that Ice Block could see some adjustment, without being sent to the Hall of Fame. Mind you, this can be done without adding another 4 paragraphs which is what I see on a lot of suggestions the pros deliver. My suggested change would be that regardless of the amount of damage taken in the last hit, the mage will always only have 1 health afterwards. It makes eye for an eye a guaranteed kill and adds potential for other counters on the mages turn, as well as making any fatigue damage lethal.
But Ice Block is not the card that worries me. I can think of tons of cards to provide a counter to that, not that I would need to when 10 healing should be sufficient to stop burn mage as a whole. But to stop Alexstrasza takes a level of power that I find straight up ludicrous. There will come a deck again after mage that can hold out until turn 9 and then use her together with some reach to end the game in one hit. That was the case with Charge Warrior, which got Charge nerfed, Reno Warlock which got Power Overwhelming on the card rotation and now Burn mage, where everyone cries that Ice Block is the issue.
This is her design space. Stall decks cannot exist. That is why she worries me. Just as she worried me back in 2015. And if Blizzard want to put an end to their own design barricade against max health and healing in general, then Alexstrasza should be the first card sent to the Hall of Fame for the 2018 rotation. It should probably have happened in 2017.