Theoretically, this makes complete sense. But in practice, math matters. If you believe that the number of promising startups is within the range of those that

Theoretically, this makes complete sense. But in practice, math matters. If you believe that the number of promising startups is within the range of those that are accepted by top accelerators, then you are right. But if you believe that current accelerators serve a minor percentage of promising startups, then it means a large part of the market (ie. promising startups) is not being served. I'm actually inclined to believe it's the latter considering the supply and demand. The demand for accelerators is extremely high and the supply of open spots is very low. I think current accelerators are just scratching the surface with respect to promising startups which is why we're seeing many more sprout up.