Is there a better way to fund research?

As I submit yet another low-probability grant bid that took up too much of my time, once again thoughts that “there must be a better way” come to mind. It seems that many colleagues feel the same way. Some interesting reading:

I have always felt that adding a random component to the grant award system, so as not to waste so much time trying to distinguish between very similar proposals, and giving out more, but smaller, grants, would be improvements. Reading comments on the above articles shows that several others agree. Perhaps it is time to try out some more modelling!