Pushed Cards - Neon

This article was originally published November 2016.

Over the time that I have been in the Eternal Beta there have been some recurring conversations about specific cards being “imbalanced” or “too good”. “Why is Sandstorm Titan so good? Don’t they know it is better than every other 4 drop?!” “Why is Siraf a billion times better than Ijin? Surely DWD could better balance them!”Eternal has a number deliberately pushed cards that are important in shaping the overall environment, but some people take issue with this.  Today I want to address some of the reoccurring themes within these conversations with the hope of advancing these discussions through discussing principles of card game design. The point is not to determine if any given card is “imbalanced”, but rather help develop the understanding of the community of why some cards are better than others.

Before I launch into this, I will give a bit of a preface. Firstly, nothing that I am saying has been endorsed by Dire Wolf Digital. Although I have probably had more direct conversations with DWD employees than the average player, they have not given me special information on these subjects. I would classify myself as a “game design enthusiast”, and have consumed tons of content on the subject, but I do not have any formal training. I am also interested in taking input from others, and really welcome feedback. There is still plenty of room to develop my own thinking on this subject. With all that out of the way, we can get started.


Many of you may have heard that XYZ is a ‘pushed’ card, but what does this term mean, and why do game designers deliberately push cards? A good definition of the term might be “a card that is obviously and deliberately powerful”, or maybe “a card that is significantly above average on purpose”. One of the most obvious examples in Eternal is Sandstorm Titan. This card has more stats than any other 4-drop, has no downside, and has two great abilities. This is pretty obviously pushed. It is also possible for spells to be pushed. Wisdom of the Elders may not be as obviously pushed as Sandstorm Titan, but this spell is incredibly powerful. The ability to threaten a removal spell or draw 2 on your opponent’s turn for only 3 power is extremely strong. There are many other examples we can come up with in both Eternal and in other games such as Smuggler’s Copter in Magic: the Gathering, and Tirion Fordring from Hearthstone.

The concept of a pushed card is pretty easy to grasp, but the reasoning why creators intentionally make pushed cards might be harder to understand. Don’t game designers want to create a “balanced” environment? Some people may have the callous response that the purpose of pushed cards is to “sell packs”, where players need to spend tons of money before they have access to competitive decks. I don’t think this is a fair explanation. In Eternal, two great examples of pushed cards are Oni Ronin and Torch. These are extremely powerful cards and they are commons. If everything were about selling packs, there is no universe in which Crownwatch Paladin is an uncommon while Retribution is a Legend.

Properly pushed cards are actually very good for card games. Mark Rosewater, head designer for Magic the Gathering (easily the most influential card designer of all time) had this to say about developing sets to have a flat power level:

It’s one of those things some players think they want, but would be miserable if we actually gave it to you. – Mark Rosewater, 20/8/2016, Blogatog

Now, let’s start breaking down some of the many things pushed cards do and why they are good. I am going to begin with the most practical reasons I am aware of, and move to more theoretical reasoning (these are not in a hierarchy of importance but of “obtuseness”).


How do card game developers sell you their game? They show you sweet cards! Whether it is big dragons or awesome swords or massive fireballs, they are going to try and hook you on the coolest cards they have to offer. Powerful cards are exciting to people, and are an important tool for selling their game.

Above I said that “selling packs” isn’t the primary reason to create pushed cards, and here I seem to contradict myself. Let’s be real here though – obviously the game developers are looking to sell their product, but I don’t feel the rational people use for “selling packs” is quite right. Many people will say “Siraf is a Legend because DWD wants us to open a billion packs to play top tier decks”, but what I am saying here is “Powerful flagship cards like Siraf are an important part in attracting people to your game.” There is a subtle distinction here that I feel paints the developers in a more favourable light.


Pushed cards can provide signposts for people to find the strategies they like. If you are coming from a card game background and had particular interest in being more aggressive or controlling, it probably didn’t take you too long to get an idea about which decks fit that strategy best. Do you like aggression? Oni Ronin and Torch are clear signposts of where to go. Do you like control? Wisdom of the Elders and Harsh Rule are great places to start. These types of pushed cards are important for people to feel comfortable in the environment and get a sense of where to begin.

It should be emphasized that comfort is a useful element of game design, and is a recurring aspect in the design of most games. First person shooter games usually have a sniper rifle, a shotgun, a big bazooka and so on. Real time strategy games all have small swarming units, the fast flying units and the huge tanks. Think of any genre of game and their will likely be some recurring themes between different games within that genre. Are there exceptions to this? Yes, obviously, but the general rule is still true. These reoccurring motifs are especially important in card games, as they tend to have more “pieces” than most other types of games, and therefore having pushed cards helps guide you to the various available strategies.


One game that uses pushed cards as signposts very consistently is Magic: the Gathering, especially in the case of limited. They often have cycles of pushed multicolour cards at uncommon to give a loud message to players on what a certain color pair is “about”. Take for example the card shown here. You basically don’t need to know anything about Magic or the format to be able to guess that white-red may have a “vehicle matters” theme. Clear signs such as this found on pushed cards are a huge help to new and experienced players alike.


Card games rely on a variety of play styles to be available to the players. Above I said that pushed cards act as signposts for finding archetypes and providing the comfort of having certain types of effects, but that is all about finding the archetype. Here I am talking about actually playing the archetype. Pushed cards do a spectacular job in defining the play style of a given deck. Sandstorm Titan makes the game about big dumb monsters and board stalls. Harsh Rule encourages unit light strategies that are looking to go long, and forces both players to contort their play. Smuggler’s Stash encourages playing a mix of both units and weapons, but with grindy mindset. In order for games to not be repetitive it is important that different cards actually feel different, and pushing cards in specific directions helps in creating that identity.

It is also important to remember here that balance happens at the level of decks rather than the level of cards. We want a healthy ecosystem of a variety of deck with different strengths/weaknesses. Individual cards can be a bit over the line as long as the decks that tend to play these cards still have some inherent weaknesses. A classic example of this is Combrei, which has a collection of truly remarkable cards, but also has several systematic weaknesses it struggles to overcome.


Next, let’s talk about “play patterns”. The term “play pattern” refers to sequences of plays in the course of a game that share certain characteristics. These may involve the same exact cards, or use interchangeable pieces that accomplish similar goals. For example, the Rakano “pants” deck is built around a very specific play pattern. First you play a cheap unit (usually with Warcry), which you then equip it with Weapons, and compliment with a few removal spells to clear the way. This exact sequence could be Oni Ronin + Elder’s Feather + Vanquish or Rakano Outlaw + Shogun’s Sceptre + Torch. There are obviously different strengths and weakness to each of these exact lines of play, but both follow a similar “play pattern”.

Players and developers have slightly different incentives when it comes to the preferred play patterns. On the one hand, players are simply attracted to the most powerful play patterns. Although a subset of players will choose to do the “fun” thing, most will gravitate towards whatever wins the most. Developers, on the other hand, want users to engage in play patterns that are fun and interactive. It is actually on the developers to make sure the “fun” play patterns overlap with the “powerful” play patterns. If the most powerful thing to do in a game is not fun players will gravitate towards the winning strategies despite this. In addition, they will blame developers for making a bad game because the “winningest” way to play isn’t fun, even though there may be more fun (but less winning) ways to play the game. The developers must therefore lead the player to ‘the fun’ by making sure it overlaps with what is powerful.

This section is going to come off as mighty subjective because of the word “fun”. Fun is a vague term, but I am not going to go into it here as that will be a bit of a tangent, but let’s pretend that we agree on what the term “fun” means. A “fun” play pattern is usually one that allows for interaction and decision-making. Although some games involving Rakano Aggro may not require a ton of meaningful decisions, the Rakano player is often faced with tough decisions on how to sequence threats, how much to go “all in”, and how to utilize their limited removal optimally. The opponent needs to make tough and interesting choices about how to spend their limited removal, when to chump, multi-block or counter-attack. Interesting decisions are being made on both sides, so it seems like this is a play pattern you would want to encourage!

Another good example of the type of play style developers will want to push is having games end. The best examples of this is clearly Siraf. If you have read my Combrei article (also known as my 5000 word love letter to Siraf), you should understand how important she is to finishing midrange mirrors. There is only so long an opponent can battle against an endless parade of beefy monsters before they are beaten into submission. If it were not for her, many Combrei mirrors would devolve into a miserable staring contest. The “activate Siraf and bury your opponent with oversized monsters” play pattern allows games to actually finish!

I want to take a moment here to emphasize the difficulty of a game developer’s job. Designing some cards that are interesting and do original things is challenging, but developers have to actually mould these cards into a game that is fun. It is a complex task of determining which play patterns are the most fun for the most people, which cards best promote those play patterns, and then balance the cards properly to best achieve this fun. That is an exceptionally difficult!


As much as developers want to encourage fun play patterns, they also don’t want any type of play pattern to get out of hand. It is important for an environment to have pushed answer cards to deal with potential threats from your opponents. When you were reading the section above there were probably some of you that thought “Hey, I don’t actually think it is very fun to play against Rakano. There are a ton of games where I just get run over by a turn 1 Oni Ronin wearing 10 tons of armor!”. If that is the case, then you should be thankful that there are a number of pushed answer cards. Annihilate is pushed. Scorpion Wasp is pushed. Harsh Rule is pushed. You get the idea. These answer cards are a great way to attack the various threats that exist in the format.

One card that I see people complain about often is Desert Marshall, which I find surprising. People will argue that it is too good, and stops too many “fun” strategies. I certainly understand that it sucks to have your neat synergy invalidated by this pesky 2/2, but he is an important part of keeping the format stable. Dawnwalker based strategies would be much more powerful if Marshall were not around, and killer Dawnwalker mirrors sound terrible. The Haunting Scream deck would be totally intolerable if there were no efficient answers to it in the form of cards like Desert Marshall. I have sometimes said that given the developers did not truly know what the meta game would actually look like, it would be best to push cards that err on the side of making the game “too fair” rather than “too unfair”. Given where we are now, I wouldn’t be shocked if there were not an abundance of super-efficient void-hate and silence effects released in set 2 as the game currently seems well saturated.


The last reason for pushed cards actually relates to increasing the skill intensity of the game. According to many, the opposite is true, so I am going to run through this argument carefully. Here I am going to need to describe 3 possible worlds. The first is a world in which the power level is flat, the second is a world with variable power level, and the third is a world in which there are cards that are overpowered.

A flat power level implies that all cards are balanced to be roughly similar in quality. Mark Rosewater has said that during design the power level of cards is flat, balanced such that everything is at the correct power for limited (draft) play. Such an environment is very low in skill, since it will almost always be correct to just cast the most expensive card in your hand. You don’t need to make as many hard choices about card value relative to board state and deck composition; everything is close to equivalent. There is always skill in sequencing and assessing minor advantages that can be gained from keen observation, but this is greatly diluted. Deck building also becomes low skill, as card choice becomes irrelevant.

An environment where there is a variable power level becomes much more skill intensive. Firstly, deck building becomes more challenging. Of course you want to jam as many pushed cards into your deck, and leave out the weaker cards, but there is a limit in the number of excellent cards that are hanging around. You need to identify what is best, which becomes harder and harder as you move down the list. In addition, you need answers to the best cards from your opponent’s deck. It isn’t enough to just have a random sample of cards that are good, since your opponent’s best cards may be much better than your answers if you are not careful. Game play becomes more interesting too. The value of cards will tend to move up and down much more depending on what is happening in the game. In flat power level world the most important cards will generally just be the highest cost card in play. In variable power world, this is often not the case, as context becomes more important. The skill of the game goes up as a result, since you must keep track of what matters and why, rather than just following what costs the most. You are also forced to come up with creative lines of play when you draw your less powerful cards and you opponents draw their more powerful cards. Sure they have the advantage, but is their a way to combine these low power cards to actually counter my opponent’s strongest threats? That leads to some of the most interesting games.

Overpowered world is an overextension of this. A card is overpowered (OP) when it has a substantially higher power level to anything else, and there are insufficient answers to the card. When some cards are significantly better than everything else the format warps around it and makes the format low skill. Deck building becomes all about playing your OP cards, answering your opponent’s OP cards and nothing else. Games are decided by who draws the most OP cards and nothing else. A good sign that a card is OP is that, regardless of match up, everything revolves around these same cards. In Eternal, the best example I can recall is Morningstar in draft back when it was a 3/2 Weapon at common. The game was all about turn 1 idiot turn 2 Morningstar. It was very hard to beat almost regardless of what your opponent was playing. These are not games that are decided by tight play or crucial decisions. These are games decided by drawing the right cards in your opening hand. In addition, it is essentially impossible to combine lower powered cards to oppose the most busted cards from the opponent.

With this breakdown, I hope I have convinced you that in order for the game to be as skill testing as possible we want cards to have a range of power level, but not be OP. Once again, this is hard to achieve when creating a game. Developers don’t get a clear sign of when they cross the line to OP, and there will always be some segments of the player-base that disagree with your choice.


As a closing note to this relatively theoretical topic, I am going to discuss a larger philosophical point. Many of the things I have described ultimately feed into subjective assessments. What play patterns are “fun”? What is the appropriate power level of answer cards? Where is the line between OP and merely pushed? What is best for the game? Anyone who has hard-and-fast answers to these questions hasn’t thought hard enough about them.

Not only are these questions challenging, game designers have an added level of difficulty as they must answer these questions for everyone. You can obviously do market research on some of these issues. Magic, for example, has done extensive research to determine play patterns and card types that are more or less popular with the player base. At the same time though, if you look at that quote from Mark Rosewater, people don’t really know what is best for them. A classic example of this that he cites in many places is how people say they want more and more powerful cards, but in reality this will eventually lead to serious problems with power creep. This doesn’t mean that we should always take a stance of “game creators know best”, as some cards end up being improperly balanced despite the best efforts of creators. We have already seen a number of balance changes in this game. At the end of the day, a DWD’s job is very difficult, and they are trying to make the best game possible. Hopefully though, the next time someone writes about how Sandstorm Titan should be nerfed just because it is overstated, you can respond with “Well, it is more complicated than that…”

That’s all for today folks! Big-picture theory articles are difficult to write, so I hope some people found this interesting. I would be more than happy to engage with these ideas in the corresponding Reddit thread. If you enjoy this kind of content, please let me know! I am always looking for feedback to help direct my efforts.