I've said for a long time that one of the major problems with games is how much we're spending to produce them. If you spend $20 million making a game, you need to make more than $20 million back. This means the game needs to have broad appeal, low barrier to entry, and can't take many risks. This doesn't leave much room to experiment, and to makes it very difficult to convince a studio to back a project with niche appeal.
So we turn to indies. Braid. Frozen Synapse. Bastion. Terraria. Atom Zombie Smasher. Good stuff. These are all excellent games, but there's a limit on how much you can do with two or three people working in their garage.
Publishers have shown they don't actually understand the core concepts of gaming. Bobby Kotick and John Riccitiello are not gamers. They are businessmen. The people green-lighting projects at EA and Activision don't know how to tell a good game from a bad one. They can't listen to a pitch and think, "Yeah, I'd like to play that!" Because they wouldn't. They understand branding, marketing, and following trends, so they look for what's already working and copy that.
Publishers have basically been gambling for the last few years. They dump tens of millions of dollars into a project. If it goes well, they turn it into a franchise and make copies until the public is sick of it. If it fails, they close a studio, fire the developers, and blame it all on used games.
Right now we have two tiers of games: The $20,000,000 games, and the $2,000 games. This leaves a strange gap in the middle where the one or two million dollar games could go. The odd thing about this is that I think this is a bit of a sweet spot. It's not hard to make back a couple of million. If you're a publisher, that's a low amount of money to put at risk, provided you actually understand the hobby and aren't just trying to make the next Modern Warfare 2. (To be clear: I'm not knocking MW2, I'm knocking the people knocking out knock-offs. Okay?)
Back in the 1990's, most developers operated in this budget range of a couple million dollars. There are certain kinds of games that can only exist in that range. You couldn't make the original Fallout in your bedroom, and if you're spending twenty million then you're making a fully 3D game aimed at the shooter market, which is a fundamentally different sort of experience.