Y Combinator picked Pollground. And one of their best-performing picks currently was their thirtieth and last pick of the 30 "most promising" they invited for

Y Combinator picked Pollground. And one of their best-performing picks currently was their thirtieth and last pick of the 30 "most promising" they invited for interviews that round -- if they had assigned it #31, it would have gotten axed. One can infer from this that some who missed the cut would likely do better than some who made it. The low-hanging web 2.0 fruit may be harvested soon, in which case they would have to expand into other types of ventures. Also, there are some technologies that are vital but don't make good start-ups. Work done on the Linux kernel (including virtualization); Compiz/Beryl; PostGreSQL; and so forth. The complexity may be an order of magnitude greater than doing a website, but you can't sell ads in Xen source code. YC's modus operandi also rules out start-ups that need more time and/or capital. In short, they are pursuing a specific niche. Don't be discouraged -- their validation is orthogonal to whether your work is technically novel or highly profitable.