mobile.twitter.com/bristoliver/status/1439966843381460992?s=21
I'm not sure the Orkney and Shetland data actually says what he claims, if false positive were simply by chance, then it might, but there's every reason to believe false positives are not by chance, but influenced by other viruses, by the acidity of recent things you've eaten, and all sorts that mean they is no reason to assume you can extend that data. Certainly not uncritically based on a simple number as done there.
Obviously any positive is more likely to be a true positive when prevalence is high (although it's also more likely to have been a past infection rather than a currently infectious one as it's more likely you'd already had it) And again what it's missing is the time factor, as LFD's and PCRS are not taken at the same time on the same sample, but the simplification on the thread is that they are, some people will have stopped being infectious between the tests, not that many 'cos the sensitivity question ('cos PCR is more sensitive, some will have recovered).
Even in people ill enough to be hospitalised with covid by day 9 1/3rd of people are not positive on PCRs. In non-hospitalised vaccinated (shortens infection) individuals there's lots of evidence that the course of the virus is even shorter, so it's likely some of the individuals will have recovered in that time, and as mrex said above, isolation is not a punishment for catching the virus, it's about preventing spread. If you're not PCR positive 'cos you've cleared it then there's no reason to be harmed by isolating.
If you're worried about false negative PCRs, you need to demonstrate harm, not simply play around with numbers above on after LFD's. The corollary to the twitter statement that in times of high prevalence the positive LFD should be taken as positive is surely that in times of low prevalence and no contact the positive LFD should be taken as more likely to be a false? That was the case in march/april, so why not just make it a rule?
PCRs taken too early 'cos of the "test if you're a contact" are more likely to be a harmful false negative than any negative after and LFD.