I believe the statistical test for "surviving" would be a propensity score matching study, but these have many issues. A simple example would be to interview 1,000 NBA players and ask them to detail all the actions they took from the 8th grade in order to "survive" and make it to the NBA. Carefully document and compile these secrets to surviving high school and college basketball, thus making it to the NBA. Then do a controlled experiment in selecting 1,000 8th grade basketball players and incentivize them to follow the exact actions of the NBA players preparations and control group without these "prepper secrets" of surviving to the NBA. Most would understand the treatment and control group would roughly have the same success rate of surviving and making the NBA. This is the same thing with the prepper community reading stories of "survivors" of past dislocations.jacob wrote: ↑Sat Mar 12, 2022 11:09 amAnd yet ... if you look at N>2 "surviving" articles and recognize commonalities in the methods of the survivors maybe a statistical signal can be detected under some confidence interval. A problem with traditional statistics is that variables are often treated as independent when corrected for. Frequentist statistics does not allow for contingencies beyond A.and.B, A.given.B, or A.or.B... essentially various Venn overlaps and exclusions between sets of outcomes. There's no way to detect an algorithmic method. It's only possible to test for inputs and outcome in terms of "trait" variables.
To give a better idea of what I'm talking about, consider how it's possible to "hide" an algorithmic approach so it still hides within a Gaussian distribution and thus appear random from a lens of simple statistical tests.
I can describe a perfect experiment to estimate the propensity score for prepper treatment and no-prepper treatment group which would reveal all the issues with a prepper strategy but would have to use several dog whistles (Gates, Soros, Fauci, Lizard/Pangolin people, WEF, Barack-Biden-Bioweapons lab in Ukraine, many others) to explain the study. The underlying issues are being able to determine when the real threat occurs and the nature of the threat, duration of threat and all interactions between these variables and unintended consequences of prepper actions. Only in movies does the star of the movie understand when the crisis starts and not another false positive.
The only protection against collapse is to do everything one can to prevent collapse. I can poke holes in every prepper strategy to reduce the survival rate to 0%, unless one is living on Picairn Island or other water independent island in the Pacific Ocean (assuming there isn't a Tsunami or random collapse of food supply that the modern world protects against during the duration of the crisis).