Ask HN: What is the distribution of startup success?

In the Platonic world there is a distribution of the success of startups that achieve $500K (or more) of seed funding. For instance maybe 1 in a thousand achieve 100:1 returns on the seed investment in 5 - 7 years, 1 in ten goes completely to zero, and the whole range in between. It is pretty close to impossible to actually gather the information required to derive this distribution, but I think a survey of folks knowledgeable in the start-up funding area would be a decent proxy. If you think you can contribute to this survey, what is your estimate of the distribution? Let's establish the following ground rules:

1) By startup we mean largely software-driven company of interest to the HN community. It can either go on to VC financing, or not.

2) Startup achieved $500K+ seed capital on or after 1/1/2002 (we want to eliminate the dot-com bust as much as possible).

3) Within 7 years there is either an actual valuation event, or good subjective evidence to support a valuation. (This rule leaves many younger companies in a nebulous state. Use your judgement.)

4) Return is return on the seed investment.

5) Estimate for distribution points of 100:1, 50:1, 10:1, 5:1, 2:1, 1:1, 0.5:1, 0.1:1, 0:1, summing to 1.

I realize this is not a substitute for due diligence, but simply a data point similar to total stock market return.