[native_samples]

It wouldn't be exponential. That's the discredited "viruses grow until exponentially until the entire population is infected" model that keeps embarrassing epidemiologists.

Epidemics come and go in waves. They always did. They don't grow until everyone has been sick, they exhibit far more complex patterns than that, hence why the term endemic is now coming up so often. It doesn't seem impossible that there were waves of COVID before 2020 and they just didn't get detected as such, given the large number of asymptomatic people, the vague symptoms that can be easily confused with other diseases, etc. I don't personally think it's likely, but continuous exponential growth is definitely not a safe assumption.

Re: phylogenetic analysis. Well, the phylogenetic tree they are building classifies something as COV2 if it can be linked into the phylogenetic tree that is rooted at the samples taken in Sept 2019, and was detected after that point. If it can't be, then it's not COV2 by definition. There's a risk of circularity there.

[jsnell]

It is completely impossible for Covid to have been present in humans in 2017. It did not take a lot for Covid to be discovered originally: a cluster of dozens of undiagnosed pneumonia cases was enough to trigger a deeper investigation and rapidly pinpoint the new virus. Why did this never happen before? Either it would have to be the case that the progenitors did not cause these symptoms (and if it didn't, well, how was it Covid?) or that the authorities everywhere in the world were incompetent for two years until suddenly we got lucky and actually asked for the right investigations to be done?

And then you'd still need to explain how the actual spread of Covid, after it was discovered, was a gradual process rather than being found to be massively prevalent all over the world the moment it was discovered.

I don't really understand your point about phylogenetics being inaccurate here. Are you claiming that samples of Covid have been artificially dropped from the tree to make it build with such a recent common ancenstor? Or that we've got a huge proportion of Covid cases that were incorrectly never diagnosed as such because nobody was considering some earlier. The latter is total nonsense.

Your theory just doesn't fit any of the facts.

[native_samples]

Well, to be clear I agree that it's very likely it originated in Wuhan in 2019. I believe the evidence for a lab leak, the EcoHealth stuff, Fauci's involvement etc is well beyond the balance of probabilities at this point. I just don't think an argument about phylogenetic trees can prove it, because of the way those trees are defined.

To play the devil's advocate:

- "a cluster of dozens of undiagnosed pneumonia cases was enough to trigger a deeper investigation and rapidly pinpoint the new virus"

A cluster of undiagnosed pneumonia cases ... that occurred right next to a large coronavirus research lab. Which presumably the hospital staff knew all about. If I worked at the local hospital of a vast virus lab then I'd be quick off the mark to test people showing weird symptoms for novel viruses too.

- "Either it would have to be the case that the progenitors did not cause these symptoms (and if it didn't, well, how was it Covid?)"

That's easy. COVID isn't defined by symptoms, is it? The definition of COVID being used by governments is "tested positive on a SARS-CoV-2 PCR test". The clinical symptoms of COVID include all possible symptoms of any respiratory illness including no symptoms at all, which is useless.

- "you'd still need to explain how the actual spread of Covid, after it was discovered, was a gradual process rather than being found to be massively prevalent all over the world the moment it was discovered"

It comes and goes in waves. If the start of testing was triggered not by the actual new emergence of the virus but by the proximity of some cases to a massive lab with hyper-vigilant doctors, then as test capacity ramped up we'd be able to observe the next seasonal wave and we'd think it was the first (incorrectly).

- "Are you claiming that samples of Covid have been artificially dropped from the tree to make it build with such a recent common ancenstor?"

COVID is literally defined as "the presence of a viral RNA sequence that can be fitted into this phylogenetic tree". Anything else isn't SARS-CoV-2 and thus not COVID, by definition. Consider that the tree has two roots! It has to have two roots because right at the very start COVID did have a clinical definition and the viruses they were finding didn't trace to a common ancestor. The re-definition of COVID as testing positive only came later with the introduction of mass testing.