Game design and skill ceilings

(This is a follow-up post to Eternal’s skill ceiling.)

I don’t think that a high skill ceiling is necessarily a feature of a great game.  Many strategy games have a high skill ceiling without being popular, such as obscure Chess variants (e.g. bughouse) and games like Go.

In terms of commercial success, I think that the most important thing is for players to love a game.  Players love Magic The Gathering for different reasons.  The game’s lead designer Mark Rosewater has labeled the main demographics as Spikes, Timmies, and Johnnies.  Spikes are interested in winning.  Timmies are interested in throwing down big creatures and winning in a splashy way.  Johnnies are interested in being creative and winning in a creative way, e.g. winning with jank.  It’s only the Spike demographic that cares about skill ceilings.

As far as skill ceilings go, there are different aspects to it:

  1. Building out an eSport around a game to drive interest in the game (so that it will make more money).  The most skilled players in a game may avoid becoming a tournament pro if their skill doesn’t necessarily translate to tournament success (and therefore being able to make a living as a pro).  I’m not sure if a tournament scene is necessary from a financial standpoint.  Some companies like Nintendo don’t even care- Nintendo actively antagonizes the grassroots Smash Brothers Melee competitive scene and seems disinterested in re-releasing the game.
  2. A player may leave a game once they hit the skill ceiling; a game may become boring after you have mastered it.  Personally, I stopped playing Arena Hearthstone once I largely figured it out and got bored of it.

Games that get old

Games get old once a player doesn’t enjoy the journey of learning that comes with mastering a strategy game.  This can happen for different reasons:

  1. The player has a misconception of a game’s skill ceiling.  For example, people used to that think that you cannot beat the house at the casino games Blackjack and Roulette.  Beating the house at Blackjack is possible via card counting.  For Roulette, it’s possible to predict the motion of the ball.  Perhaps I am missing some complexity of Eternal that hasn’t been discovered yet, much like these casino games.Among lesser-skilled players, some players say that Combrei mirrors involve little skill.  It is one of the most skill-intensive matchups in the game and is the matchup that I have punted the most.  So to some degree this perception problem might exist in Eternal for lesser-skilled players.  I haven’t really interacted with all of the Eternal player base so I’m not sure if people actually see Eternal as a game with a low skill ceiling.

    On the surface, it seems like Eternal has a lower skill ceiling than Magic the Gathering because there is less apparent complexity.  Eternal has fewer priority windows than MtG.  You cannot lightning bolt (Torch) a creature while an enchantment spell (weapon attachment) is on the stack.  I don’t really think that this reduces Eternal’s depth at all.  The lower number of priority windows make the game different but does not make the game less skill intensive.  Still, perceptions matter.

  2. The player has hit the game’s skill ceiling.  Speaking for myself, I feel like this is the case with ladder in Eternal.  There’s probably about 20+ players who have hit this skill ceiling in Eternal.

So if you look at the existing and potential player base, #1 would be more of an issue than #2.  Only a tiny portion of Eternal’s existing player base has hit the skill ceiling (roughly the top fifth of the ladder or more).  That is a very tiny portion of Eternal’s player base.  It is likely that the bigger issue is the portion of the player base with some issues with not seeing the depth in the gameplay.

Apparent agency

A card with options like Celestial Omen (a tutor card) makes it very obvious to the player that there are different lines of play.  With Siraf, the lines are less obvious.  Players may not realize that playing Siraf with 11 power is a line of play that may be superior to playing her earlier (this matters a lot in the Combrei mirror).  Some players see fewer lines of play than more advanced players.

To some degree, this is a minor side effect of good game design.  To aid the new player experience, you don’t want them to be overwhelmed with too much complexity and too many different lines of play.  Apparent complexity is what matters to new players.  If they don’t see too much complexity, they will not be overwhelmed with the game.  Mark Rosewater has an excellent article on lenticular design, which is about having new players see less complexity while having experienced players see a lot of strategic depth.

When Eternal leaves open beta, there will likely be a lot of people making content on the game’s strategy.  This will help with the apparent depth problem, as skilled players will explain the game’s depth to newbies.  Also, having visible ladder rankings will show a player that there is more to master in regards to the game.

Perceived depth and perceived skill ceiling

In an interview with Neon Blonde, Finkel stated that Eternal looked a lot like Hearthstone and therefore he thought that it wouldn’t be very good.  This is an issue that might affect Eternal’s commercial success.  The way people talk about the game and its apparent complexity will decide whether or not people even give it a shot.  And then once you start playing Eternal, you have to slog through the tutorial (which may be a turnoff to experienced CCG players because the tutorial doesn’t show a lot of apparent depth).

Some possible improvements are:

  1. Make the MMR rating less sensitive to short-term fluctuations.
  2. Display MMR in some obscure part of the interface (or on an external website).  That way, people can objectively track skill and write about it.
  3. Statistics on peak ranking (with the first half of the season excluded) would also be helpful.  One of the issues with the current system is that it discourages people from playing once they are very high up on the ladder.  Peak ranking would eliminate that problem.

One of the issues that Eternal may face is that people will say that it is halfway between MtG and Hearthstone in terms of complexity.  While this may not always be true (e.g. Hearthstone had stuff like Grim Patron), a lot of people will simplify it down to that.  I suspect that Eternal and MtG have similar levels of depth, though I play very little Eternal draft (I don’t see a lot of depth in Eternal draft) and I’ve never played MtG constructed.

Eternal draft

I perceive Eternal draft as overly complex while having shallow strategic depth.  I don’t think I’m bad at Eternal constructed (ended up #5 last season) so I don’t think I’m missing something about draft.  At the same time, there are players who think that Eternal draft has a lot of depth and really enjoy it a lot more than constructed.  They may perceive the skill ceiling in draft to be much higher than it actually is.  Regardless, I don’t know if it’s an issue.

I personally think that Eternal draft has some issues that aren’t fun.  There’s a lot of influence screw because the fixing isn’t as good as constructed (even if you run additional power, which is what I do).  It’s also heavily luck-based when it comes to drawing bombs and removal.  The Stranger deck will get randomly hated by opponents who just happen to draft strangers for fixing.

I also find that certain archetypes like Icebreaker and Hatchery Raider are almost always a trap, so it doesn’t make sense to go all-in on those archetypes.  Whereas with MtG, there are some sets where you can go all-in on a draft archetype like Spider Spawning, Burning Vengeance, multi-color Shrines in Eternal Masters, etc. etc.  I find that draft sets like that have more depth because it takes time for players to discover those crazy archetypes.


Complexity and depth

Some people equate complexity with strategic depth, but that’s a mistake.  To me, a great game is easy to teach but takes a lifetime to master.  Dominion and Chess are games with fairly simple rules, yet both games have incredible depth to them.  Games with complex rules (e.g. Magic the Gathering) have a higher barrier to entry, which stops a lot of people from getting into the game in the first place.  Contrast this with Hearthstone.  Hearthstone’s designers really think a lot about the “new player experience“, which makes Hearthstone a lot easier to pick up than Magic.  Eternal is somewhere in-between Hearthstone and Magic the Gathering.

Personally, I find that Eternal has some unnecessary complexity without benefiting from that complexity.  Hearthstone shows that it is possible to have complexity within its simplistic shell.  The Grim Patron deck supposedly has deep gameplay (probably deeper than Eternal though I’ve never played Grim Patron); my cursory glance at it suggests that this is true.  Although Hearthstone is moving away from one-turn-kills and completely nerfed Grim Patron, it does show what’s possible.

As far as Eternal goes, it has some unnecessary barriers for new players.  Endurance was hard for me to grok coming from MtG- it is like vigilance, but different than vigilance (you can Execute an Endurance unit after it blocks), and has this completely unrelated thing where it grants immunity to stun.  I suspect the reason why they have stun immunity in this game is to prevent a strategy from being unstoppable and allowing for a way for the metagame to balance itself.  Instead, the developers could get rid of the stun immunity and simply not have recurring stun (e.g. Eye of Winter, and to a much lesser degree Ice Sprite).  Recurring stun is inherently problematic because if it ever happens to be good, it will result in a non-interactive prison strategy (like Icaria blue and TJP control).  Non-repeatable stun seems fine.  That stun two units card (Flash Freeze) will get better with an Endurance nerf, but it won’t be a 4-of in Constructed because that card isn’t good in multiples.  Crystallize would become incredibly good without stun immunity, so it would need some type of costing adjustment.


There are still a bunch of little ‘grokability’ issues that hurt the new player experience in Eternal.  This includes the unnecessary complexity in the resource base.  I would prefer a system more like Hearthstone (you get 1 power per turn) with 1 difference: the game will randomly grant you an influence that you need at the beginning of each turn.  In Hearthstone, there is no complexity surrounding the sequencing of your resource base.  That’s fine because I don’t think anybody really loves the intricacy of micromanaging their resource base.

Adding strategic depth to Eternal?


Ideally, the way to go is to have simple cards with lots of gameplay depth.  Examples of such cards in Eternal:

  1. Tutors like Celestial Omen and Rise to the Challenge.  Often, what I tutor for is not clearcut.  I often need to think about whether I should Rise for Auric runehammer, Daisho, or Icaria.
  2. Siraf.  Do I wait for 11 power to play her, or do I play her earlier?
  3. Sweepers.  Do I wait a turn before playing Lightning Storm?
  4. Shadowlands Guide + Ticking Grenadin.  It opens up the possibilities for bluffs and janky attacks from the Jito Queen player.

Harsh Rule, while not an original design (it’s basically Wrath of God from MtG), is a great example of a simple card with deep gameplay.  Development could push these cards a little more.  Currently, the Shadowlands Guide + Ticking Grenadin combo no longer sees any play (8-silence Combrei does that).  That being said, development did a great job with the current metagame.

Ladder algorithm

The current Masters rankings are a little too sensitive to short-term win/lose streaks.

Bad randomness

It’s probably too late but the resource system in Eternal adds unnecessary randomness to the game.  It creates a lot of non-games that aren’t really all the fun.  Granted, a small portion of players do enjoy griefer strategies and seeing their opponents playing greedy resource bases get punished by influence screw.  However, I would assume that a lot of players get frustrated by non-games.  This is probably my greatest disappointment with Eternal’s game design.

There is randomness in Eternal’s resource system in that you can get massively power screwed or power flooded.  I don’t think that this randomness is as interesting as Hearthstone’s randomness.  Having a Piloted Shredder drop out a Doomsayer is hilarious.  While sometimes it is the opposite of hilarious if you’re on the wrong end of it, the highs and lows of that randomness are more interesting than power screw/flood (where nothing interesting happens).  That being said, it strikes me that Hearthstone has excessive randomness.  I feel like the players who enjoy randomness are fine with a low frequency of very high-impact randomness.

Good randomness

Singleton formats like EDH (the casual MtG format) highlight good randomness.  It creates very unique games since you aren’t always repeatedly drawing the same cards and playing the same combos.  Eternal moves slightly in that direction with the 75-card deck size rather than 60.

Hearthstone has simplified deckbuilding with its 30-card deck.  That might actually be better from a new player perspective than Eternal’s 75 card deck.  I don’t know if putting together an Eternal deck is daunting for new players.

What I would do if I were Direwolf Digital

I wouldn’t devote much resources into raising Eternal’s skill ceiling, which likely only affects a fraction of one percent of Eternal players.  Because the beta population consists of a lot of CCG vets, it is likely that a smaller percentage of the population will hit the skill ceiling once the game leaves beta.

I would focus a lot more on the new player experience.  There are a lot more casual gamers than hardcore gamers who like massively complex games.  To help casual players, there are little interface issues to work on- the stun duration and animation is not intuitive.  The tutorial should be skippable.  There’s a bunch of stuff like that.

While it would suck, I would seriously consider changing the resource base to get rid of power screw and flood.  Yes it will take a lot of development effort to re-balance everything and to increase the influence requirements on a lot of cards.  But I honestly feel like power screw, power flood, and influence screw are bad game design.  There are MtG variants like Stack that don’t have that issue.  I’ve played it and it’s a lot of fun.  It can be done.

For the draft format, I would specifically make drafting about draft.  It doesn’t need to be like Hearthstone Arena where there are ten million cards in the draft pool.  It’s fine to curate the card pool for a better player experience, and to do things like having build-around-me cards appear very early on in the draft.  Designing specifically for draft would lower apparent complexity and add a lot of depth to gameplay.  Phantom drafts in MTGO for unpowered and powered cube is an example of an environment designed specifically for draft, without weird game economics getting in the way (you cannot keep the cards so there’s no point in rare-drafting).

Lastly, I would focus on Eternal’s greatest strength: the archetypes are fun.  The Clockroaches archetype is an amazing design.  The game has some really sweet synergies and combos.  I would keep doing more of that.


3 thoughts on “Game design and skill ceilings

  1. I disagree about not raising the skill ceiling. What I find extremely frustrating is that so often, games are decided by A) did you have the card in your deck and B) did you draw it? For instance, try to set the world record on standing up. Odds are, you’ll fail, because it’s so easy to do. In my opinion, there are ways to improve the new player experience (simple commons/uncommons) while still adding a lot of skill-ceiling depth to top-end play (make rares and legendaries extensively complex, so that the correct answer isn’t always either A: obvious or B: limited to one of two separate things).

    The reason that chess has lasted as long as it has is that even though it is a mechanically simple game, there are a massive number of lines of play to take. From what memory serves about M:tG, there are similarly, far more lines of play to take than in Eternal. For instance, Ojutai’s Command has 12 different things you can do with it (if you choose to play it at all). Emrakul, the Promised End is exceptionally complex. Planeswalkers like Liliana provide complexity.

    Currently, Eternal completely lacks cards that do a number of different things. It lacks ways to really increase the lines of play. And while making a casual kiddie fun house is nice, is that what a bunch of M:tG pro tour veterans are really after?


    1. Thanks a lot for the comments!

      I think that MtG has the kind of complexity that isn’t simple to explain. Planeswalkers have their own set of rules that have to be explained to new players. Whereas games like Chess and Dominion are much simpler to learn and “understand” from a new player perspective.

      Dominion is much simpler than Magic the Gathering rules-wise. There is a lower number of cards and you don’t need a rules judge. I haven’t played it online in a while, but there isn’t a lot of confusion over Dominion’s rules like you have with MTGO (where you’re not sure if something is a MTGO bug or some rules interaction that you don’t understand). Yet Dominion is extremely deep and has a higher skill ceiling than MtG in my opinion. Dominion’s popularity online is unfortunately limited by bad UI (or at least the reputation of a bad UI… I haven’t tried anything other than which is dead).

      2- Casual kiddie fun houses like Hearthstone and Candy Crack Saga pays the bills many times over. It’s up to DWD whether or not they want to aim for mainstream success.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s