5 Comments
Oct 20Liked by Michael Weissman

Thanks for the great analysis! One element that I don’t find entirely convincing is the part about considering the uncertainty in the log of the prior odds, e.g. assuming a 3-df-t-distribution with an SD of 2.3. What is missing is an upper bound on the likelihood ratio. I think P0(ZW)/P0(LL)>100,000 is unjustifiable. But these absurd cases do not have a negligible weight in the t-distribution. My calculation is that the odds change from 300/1 to 1000/1 if the upper limit of P0(ZW)/P0(LL) is 100,000.

Expand full comment
author

Thanks. That should be easy enough for me to check with the numerical integration.

Meanwhile, I've made numerous tweaks, made at least one error and fixed it, and used new information in the current version:

https://michaelweissman.substack.com/p/an-inconvenient-probability-v57

It may be revised again soon because there's a new idea about a possible natural origin for the FCS that is worth considering.

Expand full comment

What's the likelihood that a very significant % of the scientists working on the most closely related research, and with the most do moan-specific background knowledge, would conspire together and lie about their beliefs and their activities to conspire to hide the truth about their views and the state of the scientific probabilities - with none stepping forward to blow the whistle?

The problem with conspiracy theories is always the probabilities. Seems a through Beyesian analysis would have to factor that in.

Expand full comment
Jan 20Liked by Michael Weissman

Oops. Moan-specific should be domain-specific. One of my better auto-corrects.

Expand full comment
author

Joshua, not directly relevant, but it would be good to step over to my current V5.

" none stepping forward to blow the whistle" ?

There's dozens of leading experts who have spoken out in public. just as the Proximal Origin authors did in private. It's not just the obvious ones- Ebright, Kinney, Chan, Nickels,.....but also ones who will only speak privately.

If I had to name the most outstanding scientist of those who have done a lot of highly domain-relevant work, it would be Bloom. It's pretty clear from his writings that he thinks the zoo arguments are crap and leans toward a lab explanation. Of those most deeply involved in the general WIV-linked research group, Baric has maintained a discreet silence.

I'm not saying authority counts for nothing but given that scientific communities have similar social dynamics to other communities, authority is a weak putty to fill in gaps not directly accessible to evidence and logic.

Here's a true story. Two personable and well-spoken researchers from prestigious universities were writing up a Bayesian analysis similar in form to mine, although not using uncertainties. Their odds favored zoo, largely because of overwhelming priors (10^5) and naive belief that Worobey was strong evidence. They haven't published, perhaps in part because we had a long discussion, but more explicitly for a completely different reason. They said they could not get a single relevant scientist to speak with them. They had legal papers drawn up by their institutions guaranteeing confidentiality. Not good enough to loosen tongues. This is not normal science.

Expand full comment