"3. Marginal Niche
Most of the groups that apply to Y Combinator suffer from a common problem: choosing a small, obscure niche in the hope of avoiding competition."
---
Usually the problem isn't that the market is too big, it's that the company is too scattered in focus. If you know your product well and you know its target, big playing fields are not a problem. You can be a small niche player and still be all over the place in delivery.
Now this does not mean that you can go ahead and invent some "obscure niche". A startup that serves a market too small and of little interest may never gain the traction to reach a tipping point.
Competition is a good thing. It takes a George Foreman to make a Muhammad Ali. Strong competition validates the smaller player. I will go as far as to suggest that without Digg, Reddit would've had a difficult time gaining wide angle attention from the blogosphere. However, competition should be avoided where it makes sense. In other words, don't consciously look for it.