I'm posting to reserve my place, and to predict that my entry will be the most stupid-simple of all the HN-ers.
Truth be told, really pulling off even "simple" graphics takes skill and an artistic eye. But often times for small indie games "good enough" really is. There are plenty of indie games with obviously developer created graphics that have done well: iShoot is a big one.
Even Minecraft's graphics have a very down to earth, homebrew feel to them. Which can both be easy to pull off and work well if you can find that sweet spot.
In general, this was trying to offer concrete advice that people could use right away.
Here's what I'd suggest (and what I'm doing for my games):
- Make sure you pick something that doesn't need a ton of art.
- First, try and make all the art yourself. For certain styles and games, this is not as hard as you think.
- If you're not happy with your programmer art (and you should be very careful here of falling in love with your bad art just because you drew it), then find an artist online and give them your programmer art to redraw. That way you don't have to "spec out" the art too much, the bad art is it's own spec and can drop right into the game hopefully.
You couldn't possibly afford to pay someone
Another option would be to use resources like Open Game Art - http://opengameart.org/ and supplement as needed (but of course be mindful of licenses).
Third you could use primarily geometry, or procedurally generated content, or ascii.
I think finding an artist willing to partner up is the best bet.
Even if you're looking at other platforms I'd suggest trying to find an artist there, there's some great talent.
You can also find art guys at sites like http://www.newgrounds.com and http://www.deviantart.com and whatever deal you make can vary, some will do it for % only, some for ridiculously little amounts of $$$, some will want ridiculously big amounts of $$$.
I tend to use the same guys (http://hardcircle.com/) unless it's a little project I can do myself. Not cheap but good quality:
I for one, may just do this.
If you're making HTML5 games check this out too:
So go for it! The only thing you have to lose is a bunch of time and your self respect if you fail!
The latter type of game looks better for marketing purposes and can command a certain kind of audience that wants couch-potato entertainment - but the first kind is ultimately more efficient and profitable, because it can exploit the user's imagination to a far greater extent.
But if someone has the numbers, I'd love to see how its profit stacks up against, say, the number one best-selling PC game -- The Sims (and number two: The Sims 2). On the one hand, WoW benefits from consumer psychology; it's like buying a new AAA title every three months, and it takes effort to cancel. On the other hand, WoW must have huge ongoing administrative, overhead, and content generation costs.
For example, I think Minecraft is a great example of a MVP, although I admit it's an unusual case. The developer started on it in May 2009 and as far as I know it was available to play or buy within a few weeks of that. It's now making a lot of money, as I'm sure you're aware.
Valve's efforts with TF2 were a step in the MVP direction. It was in development for years and years before a massively marketed release upon completion - so in that sense nothing like MVP - but the fact that they then kept on developing it and new people kept on buying it showed that games don't necessarily have to be this big "release and forget" deal where a slow start dooms the product to complete failure.
Realtime Worlds recently collapsed after taking $100m in venture capital and then releasing APB to mediocre reviews and terrible sales. What if they had been selling alpha versions of the game all along? Perhaps they would have slowly picked up fans over a long period rather than relying on the huge marketing push to find a critical mass of fans in a narrow time window before the world moved onto the next big thing. Perhaps the earlier feedback could have alerted them to the game's problems and given them the chance to pivot. Perhaps they could have seen the game's failure coming and given up before wasting quite so much money.
They were also part of the huge Humble Indie Bundle, so it's fair to say they're doing pretty well on the marketing/money front.
Maybe you have a puzzle game and you launch with 10 levels instead of 50. You can always add more as updates later.
Maybe you're doing a side-scroller and your character can only shoot horizontally right now. You can add shooting at angles in later updates.
Maybe you write a racing game, but you only include one car type right now. You can add more later.
The point was to get people to actually finish a game and turn them into 'people who have sold a game'. They're a couple of important ego boosts and carry some lessons.
That said, I wonder if it's possible to make a "this does nothing" game in 10 minutes and still sell one copy for $1 assuming it gets past the Apple censors. I think I have an idea for my submission.