Mistakes? I’ve Made a Few
This week I thought I’d let you in on a big designer secret. We make mistakes. Lots and lots of mistakes. For every success, I’d say we make twenty cards that fail miserably. Many of those cards are caught during design itself. And luckily there’s an entire development team there to catch the ones we miss. But even the developers can’t catch them all. That means that plenty of our mistakes make it to print. I’m willing to bet that you all have played with many of them. Some of which you may not even have been aware were mistakes.
In our defense, mistakes are a crucial part of the creative process. Part of innovation is exploring virgin areas of design space. New problems are hard to identify as you can’t use past history to catch them. This week I thought I’d take a look at a number of cards I’ve designed that I feel are mistakes. I’m not upset that I made the cards on this list. Hell, many of them I’m proud of. And if I could turn back time, most of them I’d do again.
The important part about identifying mistakes is to learn from them. If I can understand what I did wrong last time, I can keep from doing it next time. Success may bring you acclaim, but mistakes bring you knowledge.
Also, be aware that this column is about design mistakes not development mistakes. Power level is a development issue. Flametongue Kavu, for example, was too good. But this isn’t a design issue. Turn him from a 4/2 to a 3/2 and have him deal 3 damage rather than 4 damage when he comes into play and he’s probably okay. Basically, if development was aware of the problem and could easily fix it by juggling numbers, it’s not a design problem. Design problems go deeper than adding a mana or slightly shifting the card power.
While looking through my design files I found numerous mistakes. For brevity sake, I’ve cut them down to six. In particular, I chose cards that I feel taught me an important lesson. Without further ado, here are six cards I screwed up on: (note that these cards are in no particular order)
Donate
As the story goes, I used to have a Juxtapose/Gauntlets of Chaos deck that won by giving the opponent permanents they could not deal with (Demonic Hordes, Spirit Linked Force of Nature, etc.) While playing this deck, it dawned on me that it would be cool if there was a spell that just gave stuff to your opponent. Flash forward a few years and I’m now designing Magic cards. I designed Donate (then called Take It) and put it into my first design, Tempest. The development team removed it for being too useless. For years we played a game where I would stick the card in design and development would remove it. Then during Urza’s Destiny, the development team decided that I was too stubborn to ever give up (and they were right) and just let it stay.
So what’s the problem? Donate obviously wasn’t too weak. It was a key element of an Extended dominant deck called Trix. The problem was that the card created more design constraints than it was worth. Huh? Whenever we design a card, we have to live with two things. One, what effect will it have on the various environments? And two, what effect will it have on future design? While Donate impacted the Extended metagame, its real impact was on design.
Here’s the problem. There is great flavor in making a powerful spell that has a dangerous upkeep. Let’s take the card Forbidden Crypt. You get to draw from your graveyard but you lose the game when your graveyard is empty. This card is cool. But with Donate in the environment, it now becomes a two card kill combo. Probably the power level of that particular combo was okay, but now R&D has to constantly be on the lookout.
Donate’s existence requires us to be more cautious on dangerous upkeeps. Is that trade-off worth it? Should the game have less dangerous upkeep cards so that Donate can stay in the environment? I think the answer is no. The former has a much greater area of design space. Thus, the mistake of Donate was that the existence of one card cut off an interesting design area capable of making numerous other cards. Now, it’s possible that the one card might be worth more than the combined cut-off design space. But with Donate, I believe this isn’t so.
Humility
This card was inspired by the card Humble. This might seem odd as Humility came out in Tempest and Humble came out in Urza’s Saga. Both were designed during Tempest design (and yes, as this column demonstrates Tempest created a huge amount of material that’s been used for years after) and the development team felt that they shouldn’t exist in the same set, so the less interesting card was pushed off.
So what’s the problem with Humility? It’s also seen plenty of play and it has a cool effect. What’s wrong with it? Just a little problem called “the rules”. You see, there are certain things the game handles well. A lot of things actually. But there are also a few things that the game doesn’t handle that well. One of those things is cards with the phrase “loses all abilities”.
The lesson of Humility is that no one card is more important than the rules. As an innovator, I’ll fight to stretch the rules or to make new rules, but in the end I respect the fact that the rules are important. Making a card that the rules can’t handle hurts the game in the long run. And we don’t have to. Magic design space is immense. We should focus on cards that don’t force the rules people to make rules that the rest of you can’t understand or, more importantly, remember.
Duplicity
If you haven’t been playing the game for a while, I’m sure you’ll have to clink on the link to see what the hell this card does. (By the way, if you’re new to the site, whenever you see a card name, you can click on it to link to a picture of the card as well as a listing of the current Oracle text). Unlike some of the other cards, Duplicity hasn’t seen much tournament play.
So why is it here? Because I think Duplicity is an example of a cool card that I made unplayable in design. Here’s how it happened. Long before I came to Wizards of the Coast, I came up with a neat variant. At the beginning of the game, each player drew two hands of cards. They put one hand aside. At the beginning of each turn, the player swapped hands. That meant that players alternated between having two hands. This was pretty cool. So I decided to design a card to recreate this format.
It was at this point that I made a bad decision that would have effects that rippled throughout the design. I decided that it would be cool if the player could decide each turn which hand they had. The problem this caused though was that the card turned into a good card drawer. I pay some amount of mana and then next turn I get a new handful of cards. It became sort of a one-shot blue Necropotence. To balance this, I lessened the cards drawn from seven to five. Then I added an “at end of turn” cost.
The end result? The card sucked so much that no one played it. Not even the Johnnies that normally try out the non-tournament worthy cards. The lesson of Duplicity is that no idea, no matter how cool, stays cool if enough negative baggage is tacked onto it. And if I can’t make the card without the baggage, the card isn’t worth making.
Dream Halls
This card was inspired by the pitch cards from Alliances. In fact, when I designed it, I called it Pitch World. The idea was simple. It turned every card into a pitch card. Sounds cool, huh? And it was cool.
So what’s the problem? It’s bah-roken. That’s R&D speak for “really, really causes problems in constructed”. A quick aside – This slang came from William Jockusch. During Mirage development, we came across some card (I forget which) that William thought was a problem. To emphasize the point, William stretched out the word “broken”. It was catchy enough that R&D started using the term whenever cards were very problematic.
Why is Dream Halls so bah-roken? Because it breaks a very fundamental rule of Magic. It allows player to play a spell without paying its mana cost. Why is that a problem? Because mana cost is one of the tools R&D has to balance cards. We like making big nasty creatures (and artifacts, enchantments and spells). In order to do this, we have to be able to use mana cost as a balance. This doesn’t mean we can never make cards that get around mana cost, but when we do we have to make sure that they’re harder to use than Dream Halls.
For those of you that have never seen a Dream Halls deck in play, I recommend you take a gander at Brian Selden’s Standard deck performance at the Magic Invitational (then called the Duelist Invitational) in Barcelona. The deck felt like a degenerate Type I deck and it was playable at the time in Standard. Once Dream Halls is out, the game stops being Magic and turns into some new game where mana is completely irrelevant.
The lesson of this card is that the designers have to respect mana cost. Whenever we get around mana cost we have to be extra careful because we’re playing with fire. Dream Halls was just the worst case of us getting burnt.
Mind Over Matter
This card seemed innocent enough when I designed it. It allowed a blue player to turn every card in their hand into a Twiddle. That didn’t sound so bad. To me at least. Development was rightly intimidated by it. Enough so that they raised the mana cost to 
So what’s the problem with Mind Over Matter? I didn’t respect a different aspect of the game. If Dream Halls ignored mana cost, Mind Over Matter ignores the cost of tapping a card. The fact the cards with a tap symbol can only be used once a turn is important. Mind Over Matter allowed you, for no mana (this is very important), to reuse tapping permanents. If nothing else, it allowed you to exchange cards for mana by untapping lands.
The lesson of this card is twofold. First, the designers need to have more respect for the tapping cost. Cards which allow a player to repeatedly untap permanents (one shots seem okay) need to be carefully monitored. Second, effects such as these need to have a resource limitation greater than simply discarding cards. This means effects like Mind Over Matter should at minimum also have a mana cost to use.
Congregate
In general, R&D doesn’t tend to push life gaining cards because they slow down and extend games. But ever once in a while we’ll push them a little bit. Congregate is an example of this. But, this card never saw constructed play, how could it be a problem? Click on the link if you haven’t and see if you can catch the problem.
Back? Okay, what’s wrong? “Target player gains 2 life for each creature in play.” How many creatures can there be in play? Ten? Fifteen? Aha! That’s the problem. You’re thinking in two-player mode. Imagine a game with eight players. Two life per creature gets a little better. In fact, as I’ve been informed by numerous players who play multi-player games regularly, it’s too good. “Ban it in your house rules” too good. The card has single-handedly made many multi-player games not fun.
The lessons of this card is that the designers have to think across all formats. A card that might seem innocent in one arena could cause problems in another.
Lessons Learned
As you can see, there are many different mistakes for a designer to make. That’s why it’s crucial for us to identify and study them. If we do not understand what we did wrong, we will simply repeat the mistake. This doesn’t mean I won’t make more mistakes. I will. My goal is simply to forge into new areas and make brand new mistakes.
Join me next week when I’ll talk about the beauty of the beast.
Until then, may your mistakes require less banning than mine.
Mark Rosewater