Okay, I picked out the invalid assumption before the paragraph, "It's noticeably bigger."
See if you can! :) #foresight #futurist
See if you can! :) #foresight #futurist
The answer is, of course, the assumption of stationarity.
Stationarily means, in this context, that rates don't change over time.
Under the model of stationarity, the odds per year of a violent revolution in the 1800s is identical to the odds this year.
Stationarily means, in this context, that rates don't change over time.
Under the model of stationarity, the odds per year of a violent revolution in the 1800s is identical to the odds this year.
Of course, this isn't likely true. For instance, if you take the years 1750 to 1850 as the reference period, the probability of a violent revolution is 2% per year.
In the next century, it's 0% per year.
In the next century, it's (so far) 0% per year.
In the next century, it's 0% per year.
In the next century, it's (so far) 0% per year.
Now in practice, we don't treat statistics like that, because it doesn't much make sense. A better model using internal data only is a logistic model, which would have the revolution rate be a stochastic variable that may change over time.
A simple model for these kind of stochastic events is a logistic one, one where the rate is logistically distributed in time.
Event rates, like polling outcomes over time, tend to follow linear relationships in logistic probability space.
Event rates, like polling outcomes over time, tend to follow linear relationships in logistic probability space.
If you fit both a constant and a linear model, the version where the underlying revolution declines over time is a much better fit than the one where it is constant. It's the curve labelled "Linear rate" here. 
The next issue is that most preppers aren't going to live 79 more years. A true estimate is very hard, but a quick guess is 79 - their age.
Let's assume they average 40 years old.
Let's assume they average 40 years old.
The odds of a revolution in a prepper's remaining lifetime is 20.6% according to the Constant rate model, but only 6.1% in the Linear rate model.
Of course, any student of geopolitical modelling or of history knows that revolution rates don't change linearly with time. They have different and variable time courses. You could probably estimate a rate considering peer countries and getting more data... but this is a 1st cut.
Also people will probably live longer than that in practice, because technology isn't stationary either, but I digress.
