One of my posts last week was about the struggles inherent in balancing new and bizarre effects really bad in games. The prime example was Sideshow Spelleater, who is absolute junk and will not be played. Except, I suppose, by some reluctant players who get him in Arena and have two other even more wretched options at their disposal.
Naked Man dropped a comment on the post asking about why I felt this way about bad cards. He sent a link to an old Making Magic post by Mark Rosewater outlining why bad cards exist and why they *must* exist.
This is a good point, but isn't exactly what I meant to convey. Bad cards have to exist. For one, having variability in card power is important for the company's bottom line and collectability. For two, this actually allows better players to win matches more easily. If cards were all balanced on a razor's edge the game would be a lot more random and that isn't necessarily a good thing.
However! There is a difference between a bad card made with little resources, and a bad card that takes a lot to build and gives nothing in return. As an example, Magma Rager is a terrible card.
This card is SO bad. But it cost very little to make. There is no possibility of it generating buggy behaviour, the coding required to make it functional once the game is working is nonexistent, and it does not confuse players. It is easy to make and so it being bad isn't really costing much. It is actually a great example of a card that might well look fantastic to a new player (It beats down *really* hard!) but eventually people realize that a 3 drop that is trivially killed by any 1 or 2 drop and also dies to hero powers and AOE is TERRIBLE. This card is a nice addition to the learning curve and aside from the art costs virtually nothing to make.
Sideshow Spelleater isn't like that. It is in an expansion so people will be getting it as a card from packs they bought. It is obviously bad, which can be fine if there is something deeper to it, but on further inspection it is still just bad. The cost of making a unique ability in a computer game is a lot of coding and rewiring and bug checking and problems (plenty of people are or did complain about it being buggy and not working properly) and that is a waste if it could have been used on a card that actually would see play. This issue of buggy behaviour is less of a problem in Magic because people can usually just read cards and figure them out and even if a card does something totally new it doesn't require much effort to write that down.
It isn't that bad cards shouldn't exist, it is that if a company is going to be a lot of development time into a concept and sink coding hours into making a new thing work smoothly they ought to make sure that it is actually going to be used and that there will be a good return on investment for that effort. The players appreciate it when their interesting cards are usable and the games are more fun. People aren't sad that their Magma Ragers are staying benched forever, but when they open a new pack to an exciting card with an effect that they have never used before and immediately ignore it that is a real lost opportunity.