Comment by sashank_1509

This has to be one of the worst chain of reasoning I've seen and it is depressing to see this come from a Stanford Professor.

In 1978 a sports columnist joked that the Superbowl could predict the stock market. It went like this: If one of the 16 original National Football League teams — those in existence before the NFL's 1966 merger with the American Football League — won the Super Bowl, the stock market would rise throughout the rest of the year. If a former AFL team won, it would go down. This works with a 74% success rate till 2021. What does this prove? Literally nothing. The fact of the matter is the stock market and games are irreducible complex systems with thousands of factors contributing to their outcome. With such a system you cannot figure anything out empirically because such complexity will bring out thousands of correlations over any timeframe.

There are only 2 ways to make any sense of complex systems (human bodies is also one of the most complex systems we know). One is to have a causal chain of reasoning. For example say identify some protein that causes Alzheimers and we also find that shingles modifies this protein. Something like this would be the gold standard. Now even if this does not work on a person, we can be guided to deduce that some other condition is preventing the vaccine from working or some other protein is also being damaged etc etc.

If we want to be empirical we need a very carefully controlled experiment (which might be impossible even if ethical). We need two identically healthy humans, some way to induce Alzheimers in both of them and then inject one with shingles vaccine and see if it works. The fact is the medical establishment does not trust our ability to find two identically healthy humans, so we instead do this over thousands of people in a hope of extracting a causal relationship (a randomized control trial). Notice the one major aspect in this trial that the professor does not demonstrate? In his trial there is no inducement of Alzheimers or Dementia. Without that this whole exercise is meaningless. He seems to drastically underestimate the complexity of human life, maybe the group before 1933 in his dataset also did not take a host of other vaccines which actually stops dementia and not this to name a simple example.

Even worse he never tells us the number of humans being examined, if its in 100k range maybe there is a small chance that there is something here but I suspect this is in 1k range at which point this whole study is a joke, you can find thousands of correlations in a group as small as 1k humans. Heck even randomized control trials require thousands of people and this is most definitely a much worse trial than this. In what way does this create a “Clean”, “Causal”, “Without confounders” relation is beyond me and to me is a failure of the academic establishment that someone can think like this after becoming a professor in Stanford. I would be shocked if a graduate student talked like this, much less a professor.

Edit: Upon reading the paper this conclusion seems to be based on 5% of women in that period which is roughly 5000 women. Needless to say there can be thousands of similarities between these 5000 women that is not explained by the vaccine.