Going Deep – The Chapin AMA


Hi Friends! Last week we had the pleasure of doing an “ask me anything” on Reddit with Patrick Chapin, as you probably all know. Today’s article has a two-fold purpose. The first is to preserve some of the points in that article. Obviously the thread will stay on Reddit for a very long time, but it has already left the front page. Reddit is also not a perfect format for digesting the content from an AMA like this. It is useful to aggregate and parse the responses in a manner that is easier to comprehend. There was also a reasonable amount of repetition, which we can obviously skip.

My second purpose in this is to comment on some of the responses that were given – point out interesting comments, hypothesize on what specific hints might mean, etc. Although I have a lot of thoughts about all this, it is important that Chapin’s comments be the focus, as my personal wild speculation is much less interesting than the comments that actually come from Chapin. I will leave my own commentary to the end of each section, and will differentiate it in green. The questions, as written (with minor correction as needed) will be in black, and Chapin’s responses will be in red. I have broken up long responses into paragraphs, as it allows for easier reading. I will also say that this focuses on card design related responses, as these were the most definitive and concrete. If people would like me to include comments on topics such as future competitive formats etc. for the sake of posterity, let me know, though most of these responses are variants of “Soon (+3*5)” and “You’ll see”.

1. Metagame Balance and Diversity

1.1 Most players seem to agree that Primal feels weak overall in terms of competitive viability and faction identity. It is mostly relegated to being a support faction. Is this something you agree with and are currently considering how to remedy? If so, can you provide some insight as to what changes may be in Primal’s future to boost its effectiveness?

This is two very different topics, competitive viability and faction identity. Regarding competitive viability, I would first note that all of the supported factions pairs are in a relatively tight band of success, at a macro-level, at the moment. It’s also worth noting that balancing the game for wildly disparate levels of skill and experience can sometimes lead to certain balance decisions seeming strange to those experiencing a different part of the metagame. Not every strategy is equally challenging to build and play effectively. I would definitely expect the balance of power to move around as new sets are released, and as the player base grows more familiar with the engine.

Primal’s faction identity is definitely a subject we take very seriously and continue to discuss daily (as well as the other factions). It has a great deal of range in its toolbox; the question is really just about how many power points are spent, and where. In terms of what to expect moving forward, I would look to see Primal’s strengths highlighted and explored in new ways, particularly its ability to overpower enemies with brute force.

There are two questions here that often get layered on top of one another. The first point relates to the success of Primal-based strategies on ladder. The line “all of the supported factions pairs are in a relatively tight band of success, at a macro-level, at the moment” stands out. Although I have no doubt that Chapin is telling us the truth, there are several levels of subjectivity embedded. 1) How do we define success? Are we talking about just win rates or also taking into consideration the amount they are being played? 2) How tight is this tight band? 3) How do we define “macro-level”? I’m big into statistics and numbers so while I would love to look at the actual numbers DWD has, I understand why that info is not shared. Still, I am very curious how they slice-and-dice the data. It is not that I think they are hiding anything nefarious, I am just curious about how they quantitatively assess metagame health.

Chapin also basically makes a “be patient” argument in here, which I buy. It is impossible to have a flat metagame, and even if you could I don’t think it is desirable. In addition, there should be an ebb-and-flow to deck rankings, which we have already seen.

I am very interested to see what Chapin refers to by Primal “overpower enemies with brute force”. There are several cards now that follow this philosophy already – Channel the Tempest and Rimescale Draconus are interesting examples. How to leverage this identity into playable cards that cost less than 5 is going to be an interesting problem to solve.

1.2 Can you articulate DWD’s vision of control decks? Some members of the community have argued that control is too weak (especially Primal-based control), or that XYZ deck is not real control. For example, does DWD see Armory as a control deck? What kinds of control do you see as healthy/unhealthy for a metagame? Are you happy/unhappy with current position of control? If you are unhappy, what do you think needs to change?

We’re committed to supporting a wide range of strategies, including control. Eternal is more fun when there are more ways to potentially interact with the enemy player’s cards. Spells are an important component of the game; but strategies with all spells lead to very few interaction points. Playing cards like units and relic weapons that can be attacked from a variety of angles involves a certain degree of risk and should have rewards to match. Some players prefer to play few, if any units; and while such play patterns can take a greater toll on the metagame, their existence can also deepen and enrich the experience.

We don’t want as many people in the queue playing Four-Faction Control as a more straight-forward two-faction strategy; however, it’s fun to exist. It’s also more fun when it’s a challenge to build the best list for any given week. One of the challenges of building control decks has been the evolution of the metagame, and it can sometimes take a very dedicated and disciplined control player to be able to keep up with the times.

Some Armory decks play more like a control deck, others more like a midrange deck. In general, we’re interested in supporting control strategies that reward their pilot’s careful planning, decision making, and problem-solving; more so than strategies that attempt to remove the other player’s ability to play the game, entirely.

This is a great answer. First, we get a fascinating take on one of the challenges with spells as a card type. There are seemingly infinite ways to interact with units. Every deck needs ways to interact with units, but there is no need to interact with spells. As a result there is just less “spell interaction” space that needs to be filled. This has a cascading impact on game design. You can push units if you push answers a corresponding amount, and these answers can come in the form of weapons, spells or other units. Units can interact with spells in a certain way (Aegis or Entomb for example), but it is asymmetric.

Units are also, in general, “more fun” than spells. You get to use them turn after turn, in attacks, blocks, or other applications. This is not the case with spells, which are generally one-and-done. Not only are units fun to use, but there is a lot of skill in managing your units properly. Obviously, you want to maximize the benefit you get from every resource, but Vanquishing a Titan or Torching a Champion of Chaos can only be so wrong. For a unit you must choose to risk it in combat every turn. The skill cap of effectively managing combat is very high, and it is fun when calculated risks pay off. It may just be my opinion, but I feel units do a wonderful job of maximizing the “fun and skill per minute”, which is a theme that comes up in some of the later answers.

All of this gets lost when there is a deck that is near unit-less. This is the case of many control decks and combo decks in the history of Magic the Gathering, and was the case for the old Excavate control/prison deck, as well as the Party Hour combo deck. Huge areas of decisions and interaction are all removed. Although it is certainly fun to interact with an opponent “on the stack” it is possible to create this experience through unit combat rather than just counterspell wars. Chapin doesn’t rule out future unit-less decks, but he gives the impression that they will be rare, and will rarely be top-tier when they do exist.

The specific comment about 4F Control was interesting. This was a popular deck in closed beta for those who are new. 4F Control was problematic since it could do practically anything. It was a TJP deck that splashed Shadow for Azindel’s Gift. It had a mix of efficient blockers, powerful sweepers and devastating finishers, making it extremely difficult to attack. Although I am happy the deck existed once upon a time, I’m also happy its no longer at top of the meta.

At several points in the AMA Chapin discusses how DWD likes control decks to reward good deckbuilding. I think one of the frustrating elements of playing control in Eternal is the meta-sensitivity of the control archetype. If you hit an unfavourable pocket of the metagame your deck can go from finely tuned weapon to pile of wet cardboard. According to DWD this is a feature rather than a bug. Although I find this challenge fascinating, I don’t quite know if it is fair that control must be continually rebuild while aggro and midrange tend to be more resistant to metagame shifts. I need to think about this more. I also find it fascinating that DWD feels they can engineer this feature into the game.

This entire answer is very important in my opinion. DWD does not intend to support draw-go or any other near-unit-less control decks to be a major part of the metagame. There are some players that have a fixation on this style of deck. I read this answer as meaning these players should not expect unit-less deck to be a metagame mainstay, and if that is the only thing you will be satisfied by, you might be disappointed. Chapin pointed to both Armory and 4F Control as both being controlling decks. If you demand a “more pure” form of control, you might be out of luck. There is nothing wrong with liking hard control decks, but you may need to widen your view if you want to continue enjoying Eternal.

1.3 I’ve read your Next Level Deckbuilding book. Over the course of the game, we’ve seen constant nerfs to red aggro (Stonescar), linear aggro (to the point that nerfing Rakano has become a meme), swarm aggro (Madness Queen, Jito nerfs), the outright removal of the functionality of the one fish deck that ever existed (unstable Feln). We’ve seen the powering down of Stonescar’s rock/jund type deck (Stonescar midrange), and even nerfs to our non-blue control decks (Witch, Staff, Stronghold Visage nerfs). We’ve seen the removal of Echo Excavate locks, and whatever competitive traditional combo decks have existed in the past (Crown-roaches, Party Hour, Reanimator) are gone. We have yet to have storm or lava spike type decks, and all decks and players seem to be pushed towards true midrange type decks (Felnscar, Big Combrei, TJP Midrange). Does Eternal ever intend to make every part of that sixteen-part metagame clock equally well-represented in this game (red aggro, linear aggro, swarm aggro, fish aggro, rock midrange, true midrange, non-blue control, aggro control, tap-out control, draw-go, lock, combo control, big spell, traditional combo, storm combo, lava spike), or is the intention of Eternal to be focused on unit combat, particularly in the midrange slice of the pie?

That’s a lot of premise. Let’s see. The 16 schools of strategy described in Next Level Deckbuilding are not intended to be a reflection of good game design tenants. I’m sure you’ve noticed, many of the strategies described therein are almost never present in Magic, any more (and with good reason, it would be a disaster). The last thing I would want to do is try to force them. What I can tell you is that we, roughly speaking, target somewhat of a balance between Aggro, Midrange, and Control+Combo. And of these strategies, we want to support a variety, a range of different spots on the spectrum. Some possible cards may seem to make things interesting looking at the micro; but actually cut off more depth than they add.

Ok, this question requires a little context for those unfamiliar with Patrick Chapin. He has written two books – Next Level Magic and Next Level Deckbuilding. I’ve read them both, and I recommend them to everyone who wants to improve at card games. The question refers to a “sixteen part metagame clock”, which is shown below. This breaks up the most common archetypes in the history of Magic, and gives you a rough idea of their “proximity” to one-another.

Screen Shot 2017-04-14 at 10.22.17 PM.png

The core of the question is really the following – “Does Eternal ever intend to make every part of that sixteen-part metagame clock equally well-represented in this game”? The answer seems to be a clear “no”, although Chapin’s reasoning is fascinating. “The 16 schools of strategy described in Next Level Deckbuilding are not intended to be a reflection of good game design tenants”. This is an important message to remember when using Chapin’s Magic writing as a lens to understand game balance. The incentives of a game designer and a deck designer are very different. One wants to create an environment that is fun, interactive and diverse using a whole milieu of cards and effects. The other wants to build the winning-est deck possible, and doesn’t care how many strategies are invalidated or how fun the games are. Although there is tons of content out there about card games, I would remind people that the approach that players and designers take to the game is very different. Chapin loved playing Jace the Mind Sculptor and building decks with it, but he is likely not in a rush to design the next JTMS.

Chapin slips in a comment here that might be over-looked. “…we, roughly speaking, target somewhat of a balance between Aggro, Midrange, and Control+Combo.” Neat! First, this implies that DWD sees the current meta as being close enough to a 1/3, 1/3, 1/3 mix that they do not feel the need to intervene. This, in turn, implies that some of the decks currently popular that some see as really big midrange decks, may be considered control in the walls of DWD. Second, the fact that control+combo are lumped together is interesting. Although there are important differences in the play patterns associated with each deck, the approach of the opponent has some important overlap. Finally, it obviously begs the question of what this balance means moving forward. How strict is this 1/3 rule? How often is combo going to be part of that 1/3 rather than control? What decks are considered “combo” versus “synergistic”? It also puts an added weight on the question of what DWD sees as control. Lots of interesting questions here, but unfortunately no solid answers for the time being.

1.4 Infinite combo decks. How does DWD plan to handle decks that run absurd combos that cause issues in MTGO due to the amount of interaction? Is it a case of avoiding printing cards that have degenerate interactions, and nerfing them if that does happen? Or do you plan to rely on players’ reluctance to play action-heavy decks due to turn timers and tediousness? Or is there some pipe dream in the works to enable auto-resolution of infinite combos?

Our design philosophy has a big focus on fun per minute and skill per minute. Sitting around watching one player go through mechanical operations does not involve much of either. As a result, we are cautious about the frequency and rate at which we make cards that lead to recursive game states. That said, we are also fans of combos, synergy, and generally just getting more of the sum of the parts, even if sometimes this leads to winning the game outright. We’re more focused on quality of the play pattern, replayability of the experience, and quantity and quality of interaction points, than on whether something is a non-traditional combo kill or whatever.

The first sentence here is very important, and seems to go to the core of Chapin’s approach to designing games. I actually really love how simple and compact it is, although the implications are profound. Any card/mechanic/deck can be measured against this standard. I imagine this will be a subject of much conversation moving forward when the community discusses decks and balance. I do feel this also addresses on of the advantages of the unfun games against aggro. You are on the draw against Queen, with a clunky hand and they have the perfect curve. At least the game is short, if brutal, as opposed to unfun games against Excavate-prison that may take a full 15 minutes to conclude. Neither experience is fun, but if you want to maximize fun per minute, you must also minimize the number of unfun minutes. Later on we discuss the question of “what makes a play-pattern fun” and unfortunately we did not get a full answer. This will be my first question to Chapin the next time we have a chance to talk with him.

The specific application of this standard to combo is fascinating. For example, a combo deck that is in the game currently is Vodacombo. This deck is very fun to play with and against in my opinion. Although the combos in the deck center around Vodakhan, there are multiple routes for “going off”, which makes the games feel varied and unique. Compare this to something like Splinter Twin in Modern or Freeze Mage in Hearthstone – there are a lot of repetitive games. From the opposing side, you have a ton of interaction points, and must try to either break up their combo or race your opponent down before they hit 7+ power. In addition, the combo is fairly fast all things considered. The person writing the question is referring to Magic the Gathering Online, which has a notoriously clumsy user interface. If you find the Vodacombo kill frustrating to execute in Eternal… you don’t know how good you have it. I appreciate that DWD is not going to push combo decks that require excessive operations, as they are often miserable to play with or against. There are lots of ways to execute infinite damage combos without such tedious mechanics.

1.5 Even more broadly then the previous questions, how does DWD see their role in terms of balancing the meta? I understand balancing for the different experience levels, formats, and play styles is very very hard, but it is clear that DWD has a different philosophy than HS or MTG development given your actions in the past. What is your standard for saying that something has to change?

Our biggest force in navigating is “Going where the fun’s at.” Balancing the meta is part of the equation; but players have a tendency to do that, anyway. It’s just that sometimes the meta gets balanced around too few of elements or elements that aren’t as much fun the 100th time. Additionally, sometimes it’s just fun to explore a new world with new puzzles to solve, new challenges. Our role is to use wisdom and experience in which tools to provide players for building their own experiences; using player feedback and analytics to help navigate possible changes.

I like the note about the meta being balanced around “too few of elements”. The Aether Revolt Standard for Magic the Gathering ended up be very well balanced in a certain fashion, it was just that this balance was around 2 decks. It is quite possible we need to refine our vocabulary, and better define what is meant by a “balanced/healthy metagame”.

1.6 What’s the balance philosophy in general? Should we expect semi frequent nerfs and occasional buffs, or should we expect rotations and card releases to do the majority of the balance work?

Balancing one card pool for a variety of game modes can be very tricky; but that’s already true. It’s an ongoing evolution; but the biggest thing is maximizing for fun. We’re generally not trying to change stuff, just to change it; but if changes will lead to more fun for the world as a whole, we’re into it. In general, we’re interested in promoting a diversity of strategies, games having a satisfying number of decision points, and increasing opportunities for counterplay more than nerfing strengths.

Some important notes; firstly, DWD does not rebalance the metagame just for the sake of stirring things up, and they prefer increasing opportunities for counterplay rather than using nerfs. The first point seems to specifically focusing on balance patches rather than Promo cards. Promos are clearly trying to push the meta a little. I appreciate the focusing on finding opportunities for counterplay. I have spoken in the past about how some cards are “nerfs by proxy”, such as the printing of Throne Warden being a major nerf to Rakano Warcry and Burn. Balance adjustments are obviously important on occasion, but I think we all generally prefer this approach.

A second point – Chapin shares another qualifications for fun play patterns – “satisfying number of decision points”. What does that mean? Games that functionally end on turn 3 or 4 certainly don’t qualify. On the other hand, too many decisions can be repetitive. I would argue that some decks from Magic the Gathering have too many decisions to make, most of which are very obvious. A classic example is “going off” with storm in Modern, or playing Miracles in Legacy (once the Counterbalance-Top lock is firmly established). Although there are a lot of decisions to be made, most of them are fairly obvious, and require little skill.

2. Variance

2.1 The write-up for the redraw rule change did cover some interesting points about the “philosophy” behind redraw rules, but I wanted to dive deeper. I addressed the mathematics of the change in a recent article, and the results (to me) seemed to suggest that running 25 power was usually best, with any additional power being made up largely by power-searching effects. Although I doubt my analysis of the change was 100% accurate, I am interested on what exactly DWD was trying to achieve with redraw rule change in terms of deckbuilding. Do you want people to play more power in their decks? Do you want people to be running monuments? Or do you just want people to complain less about screw/flood? Did the change to the redraw rule meet those objectives (at least so far)?

The changes to the redraw rule were primarily about increasing satisfying gaming experiences, while preserving depth and diversity in how the games play out. While the new redraw rule is less punishing to people playing 25 power at the wrong times (since their deck will at least function); it definitely doesn’t mean it is always right to play 25. For instance, a deck with 4 copies of a key card and 25 power has a 22.55% chance of seeing a copy in their redraw hand. However, if they had 26 power in their deck, they would actually have a 22.98% chance of drawing their key card in their opener.

Although Chapin’s answer was exceptionally eloquent and had more depth, it seems as if his response was basically, “we just want people to complain less about flood/screw”. That’s fair. He points out this neat interaction related to running more power to amplify the chance of drawing a key card in the opening hand. This is an interesting take on the subject, and may be particularly relevant to anyone building a combo deck.

2.2 Drawing too much or not enough power is probably most annoying factor in Eternal (and e.g. Magic). Adjusting the mulligan is a good step in the right direction, but even then you could end up with 2 power (or wrong color) for a while or the other way around.

How do you feel about a system like Spellweaver? Would anything like the following – potentially adjusted – ever be discussed as an option?

Once every turn you have the option to discard a card from your hand to look at the top 4 cards from your deck and put a single power card (sigil) into your hand.

The resource system is an extremely deep subject, but suffice it to say, we have discussed just about every resource system under the sun and have no plans to change it in any kind of a fundamental level like that. Variance is fun!

Variance is fun. If there are people out there that don’t buy this and are interested in discussing it, it can be the subject of a future article.

2.3 How do feel about the design decision to make the deck size 75? To me personally I feel like the added randomness that comes with the deck size detracts from both play and deck building. Trying to create a level of consistency in certain strategies is hard to achieve with the extra large deck size.

Using 75 cards makes for more variety of challenges, more diversity of puzzles to solve, more personal customization, and more reward for in-game skill and improvisation (rather than just memorization). Ultimately, fun and replayability are the most important goals – people (including ourselves) have played and enjoyed The Eternal Throne set much longer than we originally anticipated.

3. Counterspells

3.1 Counterspells feel very weak in Eternal right now due to their limited scope, but adding response windows to unit and attachment summoning would make the game slower and perhaps more awkward. Are you happy with the state of counterspells currently? Are there plans to expand their functionality or introduce similar concepts in the future?

The nature of our response windows has a big impact on what kind of Negate effects we do, but there is also a lot of online only space that will let us do some pretty cool things. While we value letting players play their cards, we also want to provide players that can correctly anticipate the threats they face to have a variety of satisfying answers at their disposal. I would definitely expect the range of Negate effects to increase, not decrease; but we’re happy with this as a starting point.

Neat! I have said in private conversations that I don’t think DWD would implement counterspells in the manner they are used in Magic. They would require a fundamental rework of the response window rules. Chapin hints here that some future cards might encapsulate the counterspell experience while avoiding the priority system issues. If I had to guess what this might look like are Secrets in Hearthstone or Traps in Magic that cast automatically. In The Elder Scrolls Legends you get to draw a card when you reach 25, 20, 15, 10, and 5 health or lower for the first time in a game. If that card is a “Prophecy” card you get to play it for free. Prophecy cards range from being a little to a lot over-costed relative to their effect but can sometimes be free. This is obviously a big bonus, so they are a common inclusion in many decks, despite being a touch below the curve. I could easily imagine a similar effect for Eternal. Take for example the following cards:

Squash – 2S – Spell

Destroy unit that costs 3 or less.
Trap – if an opponent casts a unit that costs 2 or less, this spell is cast targeting that unit.

Pigify – 3PP – Fast Spell

Target unit becomes a 2/2 Pig.
Trap – if an opponent casts a unit with a Summon or Entomb ability, this spell is cast targeting that unit, and they lose all Summon effects.

Time Warp – 2T – Spell

Target unit is shuffled into its owner’s deck.

Trap – if an unit with 5 or more attacks you, this spell is cast targeting that unit (if more than one is attacking a unit is chosen at random).

Hooru’s Blessing – 2PJ – Fast Spell

Target unit gains +1/+1 and Aegis
Trap – if an opponent casts a spell targeting a unit you control, this spell is cast targeting that unit.

I obviously haven’t put much thought into balancing these, but it seems like a possibility that something in this space could be created. It may be possible to design them so they are optional rather than mandatory as I have spelled out, but that adds in response window issues. There is a reasonable amount of space in this area, and I am looking forward to see what DWD comes up with.

3.2 Hi, Patrick! 1) Can we expect a real counterspell effect to be introduced to the game? 2) Can we expect an introduction of “choose one” effects with set 2 or later? 3) How wide are the options for instant counter play (fast spells/effects) should be according to DWD’s vision? 4) Are there plans or ideas to introduce new card types to the game, or the design team feels like there’s just enough?

There are some very important things being accomplished with the response rules, and we don’t want players to have to pass back and forth on every single card. That said, I would expect the range of options for how to interact with cards to increase in some surprising ways… We’re a big fan of cards with two or three options, and the “choose one” style of design you mention works excellent in Eternal. Spoiler(!): There is a cycle of uncommon multi-faction spells in the next set that each have two modes…

More hints at new mechanics and/or card types as pseudo-counters. In terms of “choose one” cards – I don’t quite understand why people asked this. It seems obvious to me that they will explore it in the future, as modal cards are incredibly popular in other games and are both fun and skill-testing. We also already have 2 “choose 1” cards in the game – Accelerated Evolution and Marisen’s Disciple.

4. Forms of Interaction

4.1 Concerning entomb and void hate – Most people would expect Entomb effects to fire before a unit reaches the void. But in Eternal, Entomb effects fire after the unit has entered the void, which means Entombs can be cancelled out by void hate cards. Which makes void hate more powerful than would be expected. Why did the Eternal team make the decision to allow Entomb effects to be affected by void hate?

The most common ways to interact with cards is to kill them, which necessarily doesn’t work with Entomb; so we wanted to make sure there were a variety of ways to interact with Entomb cards (though they are still less interactible than most types of cards).

I feel like there is a weird disconnect between appreciating the power level of threats versus answers. People seem frustrated when their powerful entomb units are hosed by Steward of the past, but forget that the only reason these cards exist in the first place is because they can be countered. I know we all want to have our own toys without interference, but that is just not how the world works. At present, if it were not for Steward of the Past and Statuary Maiden there would be very limited answers in either Feln or Stonescar for Entomb effects. I know it sucks when you have your fancy Reanimator deck and your 2 Vara’s you dumped in the Void are silenced by your opponent’s turn 4 unit, but things would be busted if this were not the case.

4.2 Attachments are popular and almost all relic removal is terrible to have in your deck for the ladder and even the best one furnace mage is very lackluster. Why the lack of versatile attachment removal cards?

Regarding attachment removal, we agree there is less than we’d like in general. Making a game from one set is very different than adding cards and sets to an existing format. We wanted to leave big gaps early on to give the format room to radically evolve with the release of new cards and sets that flip a lot of previously basic assumptions.

This is great to hear, as I think Obelisk is a little oppressive. This card has the most warping effect on the format of all available cards, and is a major pressure on control. It is certainly tempting to develop things such that attachments are a core element of the metagame, and then introduce a powerful attachment hoser to generate a big swing in the meta. I just hope it comes sooner rather than later, since I am very bored of the attachment-heavy metagame. Some example cards that could fill this role (once again, not a lot of thought put into proper costing):

Stonescar Vandal – 3FS – 4/3


Summon: You may destroy target attachment and give Stonescar Vandal +2/+2

Cat Burglar – 2S – 1/4

Infiltrate: Steal a random attachment from your opponent’s hand.

4.3 Given that there are still a vast number of possible unique mechanics (some that might only be possible due to the digital format ala Warcry/echo), with many more yet to come, do you feel that the silence mechanic, particularly when pushed (I.E. Desert Marshal, Valkyrie Enforcer, Steward of the Past to a lesser extent) constrains the possibility of creating cards in the future? I.E. if a mechanic or particular card’s unique ability is so good that its cost is factored into the cost of the unit, which then loses a great deal of value when silenced, does it then run the risk of becoming unplayable? Will we see a change in pushed silence-type cards (particularly units) to allow more unique mechanics or individual cards to flourish in the future?

We made a conscious decision to push Valkyrie Enforcer and Desert Marshal in set 1, so as to help provide a safety valve for units with amazing abilities, like the Champions. I would expect we’ll see a wide range in the future, with an increase not only in mechanics worth interacting with, but a wider range of ways to do so.

Similar to my discussion about Steward of the Past, the power level of silence effects is an important safety valve of keeping the format healthy. If silence were not pushed then silence-able units could not be pushed. Although I understand some people enjoy formats where everyone is just trying to throw haymakers at one-another, and see who can pull off the most outrageous shenanigans, it is not the kind of gameplay I would enjoy for long periods. The question also discusses the premium on an ability being factored into the cost, and how silence brutally punishes this kind of card. In fact, I’m sure DWD pushes the power of these special abilities exactly because powerful silence effects exist in the format, so the cost-to-effect ratio includes silences in its calculus.

5. Complexity and Play Patterns

5.1 What is DWD’s philosophy about game/card complexity? This is clearly a …complex… topic, but there have been a number of game changes already that are at least partly a function of reducing game complexity (e.g. Excavate, Crown, Withering Witch). Some argue that these changes drive the game to be completely unit-focused, and this “dumbs down” the game. What logic/standard does DWD apply to identify effects that are too complex? What levels of complexity do you think is appropriate for various skill levels?

Complexity in games gets a bad rep. We are big fans of complexity, just the right kinds and in the right places. Strategic complexity is extremely satisfying, and can frequently be hidden beneath the surface, so as to not overwhelm players that aren’t ready for that level of the game, yet. The number of possible 75 card Eternal decks using just cards available right now has about 100 digits more than the number of atoms in The Universe. It is mind-blowing to consider just how much depth there is, and that’s just in deckbuilding.

There are actually an incredible number of ways you can use Finest Hour, Torch, or Teleport. Those cards lead to a lot of strategic complexity. While we want to create a wide variety of cards, we’re most interested in cards that have complex potential consequences and implications of using them; rather than complex legalese leaving players trying to understand what the card’s text even means. We’re also very interested in fun per minute and skill per minute. For instance, a card with a lot of on-board complexity, rather than deep strategic complexity, might reward a certain type of repetition and maintenance skill. However, if it slows the game down enough, it can easily decrease the amount of skill per minute.

For instance, some players note the auto-passing of priority revealing partial information about players’ hands, a rule that obviously speeds up gameplay tremendously, compared to any system involving players spending lots of time just clicking back and forth to each other, over and over. In this particular case, there are actually a variety of other skills being measured, that many players do not yet realize. Balancing complex systems of partial information and having the potential to preserve or secure information is a very skill-testing pursuit. Anybody can bluff a Torch when there is no autopass. However, the card Temper takes on whole new meaning, when it can actually bluff a Torch for you. Even deciding how to balance the possibility of wanting to Torch something on the first turn, vs the risk of revealing you probably have a Torch is a very important and difficult challenge presented to many players on the very first turn.

Cards are kind of on a case-by-case basis; but it really comes down to trade-offs. Is the card adding enough fun, enough strategic depth, for the costs that go along with it, the risks? Re: Crown of Possibility – It’s a super fun card, but the old way was an experiment inconsistent with our rule set that proved more problematic than it was worth, making each individual skill less meaningful (and exacerbating spots where the onboard complexity dragged on the game). The biggest thing, however, is that in cleaning up our rules, Crown of Possibilities was a strange one-off exception that really shouldn’t have been. Re: Withering Witch – This is an oddball that uses very different rules than just about any of our other cards, leading to some pretty non-intuitive scenarios. The card’s strategic implications, however, are not actually that complicated. Most of the card’s utility can be distilled down to one of the following two options: Kill wounded units. Shrink units and then hopefully kill them, whether on this turn or another.

This is the longest answer in the AMA. There is a lot of meat here to chew on, so let’s dig in. The first paragraph of the answer discusses deckbuilding complexity. Although I know Chapin is correct in saying there is an absurd number of possible decks in Eternal, but I am honestly not a fan of this answer. The number of playable, competitive decks that are substantially different from one another is less than 1% of 1% of 1% of the total possible decks, and I feel that this much smaller number is more important. I still feel there is a ton of interesting space left in deck design, and love solving the problem of the weekly meta even after playing with the same cards for almost a year.

As much as I am lukewarm on Chapin’s optimism on deckbuilding, I couldn’t agree more on his comments about strategic complexity. This answer was a little rough given his time constraints, but it closely relates to another thread on Reddit from last week that asked about real decision points in a game. I watch a lot of Eternal streams, often without the player knowing, and will notice play choices I disagree with. Although it may feel like you can play on autopilot a lot, I promise you are giving up some wins as a result. That being said, this is just talking about the player’s perspective rather than the game designer’s perspective. It appears as if Chapin has a strong preference for effects that are exceptionally straight-forward on face, but multifaceted in its application. These kinds of clean designs are great, and I would love to see more of them in the future.

In some respects I feel this question relates to DWD’s vision of “fun” gameplay. Many games follow common patterns, but the most interesting games are those where you must use your cards in innovative ways. The first time you realize that Torching a turn 1 Oni Ronin is sometimes wrong because you must save it as a pseudo-Ruin for a turn 3 Sword of Icaria or to pop the aegis of a Silverwing Familiar are fun moments. Improvisation is fun, and these versatile spells are a fascinating element of card games. This interplay between pattern recognition and improvisation is especially loud in draft.

Chapin does not offer much additional reasoning for the nerfs/card changes from the past, and repeats the answers that have been given elsewhere. This is fine, but a little disappointing.

5.2 A lot of the reasoning I’ve heard about the relative strength of threats vs. answers (I.E. why we don’t have draw-go yet, for instance) is that “people want to be able to play their cards” and other comments about “fun play patterns”, and so on. For DWD, what constitutes these patterns? How close are we to that ideal world? Can you better explain, from your perspective, what play patterns you’re seeking out, and how pushed cards (I.E. Sandstorm Titan, Steward of the Past, Oni Ronin, Desert Marshal, etc.) play into creating these ideal patterns?

Fun play patterns is, not surprisingly, a deep and nuanced subject. To start with, fun is about having the experience of improving your position from an imperfect state because of an increase in your mastery and good fortune.

One example out of the list is dynamic gameplay. Cards playing differently, but satisfyingly, from one game to another can be a pretty good thing. Knight-Chancellor Siraf is a very powerful card that gets activated a lot. Fortunately, her ability is very dynamic and makes games interesting, turn after turn, unlike most repetitive actions.

Once again, we don’t get a full answer to the “what makes something fun”. It is a deep topic, which I would like to hear more about. Chapin’s responses to this question here and in the past have referenced Erin Hoffman’s work. I suppose one day I will need to get my hands on her book on the subject. I really like the example he uses of a fun card. Siraf is a blast to play. Chapin says he has probably activated Siraf more times than anyone else, and I believe it, but I have probably activated her more times than anyone who lives outside of Denver Colorado. Although activating Siraf is a roll of the dice where you can high-roll or hit snake eyes, she is usually going to get you ahead in a game. I like this a lot more than random effects that are extremely high variance, like Ragnaros from Hearthstone. I’m hoping we get more cards in this spirit in the future.

5.3 The nerfs that came in 1.18 mentioned “Improve balance by pulling back on some of the more frustrating play patterns.” This doesn’t really explain whether the strategies were too powerful and winning too often. Is the intended design to take decks that implement these “frustrating play patterns” and ensure that they have lower average winrates than decks with play patterns that you wish to promote?

A card’s win rate is really about the relationship between its popularity and power level, so it’s actually a much more complicated subject than this. In general, however, we wanted to promote an increased diversity of gameplay.

Not a lot to say here, other than a reminder that DWD gets a cross-section of data that is much more nuanced and deep than what we see.

5.4 What have been the biggest surprises from the community so far? Are there specific decks that we created that RnD did not consider? Are there certain types of feedback that you have been unprepared for? Are there things we have “missed” that you are surprised about?

I’m not sure I would say surprises, as much as there being tons we’ve learned from the community. One strategy we did not play as much, or in as much depth as the community were the various Echo-scavate lock decks. There are some cards and strategies we suspect are better than the community realizes; but no need to spoil the fun, now. Maybe the biggest surprise stems from the community missing Sandstorm Titan. Have you seen that one? From what I hear, I guess it has some bigger numbers on it than some other cards have. It’s weird that no one ever noticed before, but I guess sometimes cards fall through the cracks. Just recently, however, someone submitted some feedback on it, so it’s on our radar now and we’re going to start looking into it. I’m thinking Rakano Banner might be the card to go.

A couple points to bring up from here. First, Chapin says there are some under-explored decks hanging out in the card pool. I have heard this compared to the “Unicorn Priest” from Hearthstone, where the HS developers hinted about there being a strong Priest deck in the card pool, but the community felt Priest was bad. Since I am the classic DWD cheerleader I am inclined to believe Chapin and have continued to brew, but BOY could I use a hint.

The second thing I would note is to just talk about Echo-scavate lock decks, and explain a little why they are so toxic, since some people still pine for the days of Excavate-prison. I should make clear there is nothing “wrong with you” about enjoying playing a deck like this. I liked playing these decks well enough, though I was never crazy about them. Decks like this are problems because they break one of the cardinal rules of game design – playing your game should be fun, and as long as the game continues you have a chance to win. Mark Rosewater talks about how players believe the game incentivizes you towards fun. If the game “encourages” unfun play patterns, the players will blame the designer for coming up with a crappy game rather than blame themselves for choosing to play the game improperly. A good example of this for non-card games is Monopoly. We have all seen a game where one of the players is in a near-losing position, but can barely squeak past Go with 2$ to their name only to have their lunch money taken by the bullies building up actual real estate empires. They are miserable playing the game, but they continue to play since they have not technically lost yet. This is one of the many flaws in Monopoly as a game.

If your opponent has you in the echo-Excavate/echo-Copperhall Blessing lock there are a lot of decks that are literally unable to win unless their opponent makes a mistake. Now the “locked” player has a weird incentive structure. There is technically a non-zero chance they win this game (“opponent is hit by a meteor” argument). There is also a chance that the locked player doesn’t understand how the loop works, and hasn’t put-together that it is impossible to win. Also, conceding violates this unspoken rule that playing a game is supposed to be fun, so this locked player stays on as a mix of protest and being stubborn. A player may also force there opponent to deck them as an attempt to frustrate their opponent (possibly waiting the entire turn time before passing). These are all reasons that a player may stick out a game for 30+ turns until they are decked, miserable the entire time. Given there are several reasons for a person to play-it-out, and the magnitude of discomfort experienced by the locked player, I agree with DWD’s decision to nerf these decks. (This is all aside from the Excavate-lock mirror problem, which has the potential to end in 2-hours of players gaining 2 health a turn with no cards in deck).

Also, I appreciate the Titan jab. Games are supposed to be fun right?

6. Designing Cards

6.1 Can you explain how you handle the stats on units? Like usually in cardgames you see a 4 mana unit with around 9 stat points (so a 4 mana 4/5 vanilla) and depending on rarity and abilities it gets modified. But then Eternal has Sandstorm Titan, a 5/6 for 4 power, with one ability that completely shuts down a strategy (flying units) AND Endurance, which is a monstrous ability on units with high defense value. So I feel like Sandstorm Titan is overstatted, which leads me to the question how you arrive at your unit design with costs, stats & abilities when planning a new set?

Will we see some more removal to punish players from playing big units in Set 2?

The way we select stats to put on a unit is actually a very rich and complex subject. Different cards have different purposes. Not every card has the same rate of efficiency. Also, not every card’s utility in the game is the same as its rate. The existence of powerful cards is fun. Some cards really are better than others. Figuring out what you think they are is fun. Additionally, figuring out how to leverage what you know about the cards and strategies other people like is also fun. Sandstorm Titan, for instance does have bigger numbers than some other cards. What does that tell you about Time? Every faction has a number of cards that help define what the faction does well.

Additionally, it has abilities that impact the game in a dynamic and meaningful way. It can single-handedly change the pace of a game when it comes down, even one someone was losing badly. It can also lead to a radical change in the game, if it’s killed. We believe games of Eternal are more fun because of the existence of Sandstorm Titan, even games where neither player has it in their deck(!) The existence of Sandstorm Titan might encourage a player to play Annihilate instead of Permafrost, Vanquish instead of Suffocate. The changes in removal people play with leads to more opportunity for meaningful decisions. The variety of removal increases diversity of gameplay experience for everyone. If it ever seems like too many people are playing Sandstorm Titan, players have lots of satisfying answers for cards to add to their decks to improve their percentages against it. It’s a very strong card; but by design. It helps set a bar of the sorts of things people do. There will definitely be a variety of new answers in the next set to choose from for finding the right way to interact with the threats you anticipate facing.

For people who might want to see some further discussion on this topic, I would encourage you to check out my “Defending Darude” article, or my “Pushed Cards” article. Pushed cards make for interesting deckbuilding, varied gameplay, faction identity, and more. Games where Sandstorm Titan is played feel different than games he doesn’t show up, which helps make the game fun! Darude is clearly a controversial card, which I have said my piece on in the past, so I don’t want to re-hash it here. What I can say is that this response suggests Titan is not going anywhere. It’s a du du du du du du du world and we are just living in it.

6.2 Will we have the power level of extremely expensive cards raised? For instance, there seems to be little justification in playing extremely expensive top-end cards such as Channel the Tempest, Shadowlands Feaster, Curiox, and so on that leave you very vulnerable early due to their inability to be played at all. In contrast, we have more flexible cards that serve as both value finishers and cards that can be played early on to get you to that point, such as Great Parliament, Siraf, and Xenan Obelisk. How do you feel about the state of high-cost cards in Eternal on a whole, and do you think that they may need to see their power level increased in order to be better compensated for the risk of sitting dead in your hand in the context of more flexible cards that generate a guaranteed advantage later down the stretch?

I think we’re going to see an ebb-ing and flowing. We started in a pretty good spot, for what we’re trying to do with this set. The goals will be different in future sets. When you want to make a format fun, it’s good when the distribution of costs people play is very flat. It’s nice when there are about as many playable 3s as 4s, about as many playable 2s as 5s. However, the more cards you add to a format, the more power-efficient the best cards and synergies are. This leads to a distorting of the value of cards’ rates. Ideally, the distribution of powerful cards should be thicker in the middle (lots of good 3s, 4s, 5s, and to a degree 6s), with 2s less common, and 1s even less so, just as 7s will likely remain less common and 8s, even less so. With a more bell-shaped distribution, the aggregate effect of several sets is a more flat distribution (tailing off a little at the top). There will be powerful cards at all these spots; it’s just that if we pushed too many of the high end cards too much, too early, they would have a very substantial limiting factor, crowding out the space to do other designs. We also are leading with an above average number of good 1s and 2s to start with, helping get the “full format” experience, with just a single set.

This answer is very interesting, and touches on some aspects of set design that I had not considered. Chapin rightly identifies that formats tend to get more compact as time goes on. Being able to play 2 or 3 spells a turn while your opponent is only able to play 1 is so powerful that we see formats like Modern and Legacy in Magic getting compressed down to the point that “free” spells are more popular than 4-drops. If DWD released sets with roughly the same power level distribution as set 1 the format would become totally warped quickly. We have some excellent cards that cost 3 or less, and it would not take long before that was practically all we saw.

Now, the question is actually related to 6+ drops. In a smaller format the standard for the high-end finisher is going to be a relatively low bar, since surviving to turn 6+ is not as difficult. As the format matures, the pressure on top-end finishers increases. They will not only need to trump more powerful 1-to-5 drops, they also need to be worth building your deck around. Although I actually like the design of many top-end finishers in the game at present (Icaria, Sword of the Sky King, Vara, etc.) I expect they will not be able to keep up for very long. If 7-drops just won the game right now, than there would be a lot of decks that only revolved around playing a 7-drop and riding it to victory. By allowing a card like Icaria to be “powerful but mortal”, it means the game continues when she is played, rather than ends on the spot. If the format becomes compressed, I expect 7-drops will need to be devastating to remain playable.

6.3 Design Theory question: Are the various abilities clearly divided between the factions in terms of strengths and weaknesses, or is it more a case of moment to moment decisions rather than rigid structure? For instance, we originally saw ambush primarily in time/primal, but with the addition of Cabal Countess and Shadowlands Feaster it seems to be firmly within Shadow’s sphere.

The abilities are divided between the factions internally, but exceptions are also possible. Regarding Cabal Countess and Shadowland Feaster, we originally experimented with Ambush in Primal and Killer in Shadow, but eventually decided we preferred the game play from swapping them.

These realignments have worked well, and I’m looking forward to seeing how they evolve.


That was a long one! Obviously had the help of Chapin’s magnificent answers and your magnificent questions to work from. Lots to discuss here, so I look forward to reading all of your thoughts! Were there any important points you feel I missed, or that you would like to hear my thoughts about? Let me know! I specifically welcome any commentary from DWD employees on any of this, as I’m sure the community would love to continue the dialog on the design of Eternal. This AMA was a real blast, and I learned a ton about card/set design form it. Looking forward to the next one already!




  1. I critique where appropriate, but I am just sharing my thoughts. If I had legitimate issues with his answers I would have shared them, but I really don’t.

  2. I liked this article and appreciate the effort and time it took to compile and respond to Chapin’s answers. I’m a fairly new player so I especially like when Neon goes over the history of Eternal, like with the Excavate and Blessing lock. My one complaint I have is Neon doesn’t ever criticize an answer given by Chapin or even give a differing opinion. The author is constantly agreeing, cheering, and explaining in every response and the article would be vastly improved if Neon had taken a more critical approach to some of his responses.

Leave a Reply