[fbbwsa]

Sort of a terrible example by the author (Gladwell, I guess), I think with regards to the wealth creation window being limited to those born between 1831 to 1840.

don’t see why its surprising at all that 14/75 of the richest people in the country came from a given 9 year time period.

Since we don’t know what the distribution looks like, lets grind through this.

If we had no a priori reason to suspect that that particular 9 year period is different from any other, we might think that 75/14*9 ~48 years encompasses all 75 of the richest people 75 people in the country. If we expected that 9 year birth period to show more wealth creation then we should see MORE than 19% representation.

In a time where most people didn’t accumulate their wealth quite as early, lets imagine that people could even imagine to START getting rich at 25 (remember this is a time where instant wealth creation via the internet didn’t exist). Add 48 years to that, and we’re at an age of 73.

Current life expectancy is 78 in the United States and it was lower back then.

So… the claim that 14/75 of the richest people were born within a 9 year period is not inconsistent with the null hypothesis that being born in that 9 year time bracket shows absolutely no proclivity towards wealth creation than any other equivalent 9 year time period.

[jeggers]

Two issues with your calculation: (1) You are considering contiguous 9-year periods across which the 14-person sets can be applied with the 48 years + 25. Because the span of time is recorded history, and with a brief look at the Forbes list, at least 1000 years are covered... or 966 sets of 9-year periods (1000-9-25), rather than 5.

(2) Life expectancy globally across this period of time gives a much wider range.

Overall, your calcs are too narrow to disprove Gladwell's assertion. We'd have to dig more deep into the actual numbers to get the real stats to show how likely/unlikely this is.

[fbbwsa]

you are correct, I skimmed/misread the post. I thought it 75 richest LIVING people (and could not for the life of me figure out why they were using such an old Forbes list...), not in history.

Thats why I was assuming we could assume dead people and those under 25 wouldn't be as relevant and fit all potential candidates to a 50-ish year birth-year block.

Taking any arbitrary 9 year period in a 50 year period and finding that it captured ~20% didn't seem odd to me, which is why I screwed up with my bad argument.

Good catch.