If people were not testing, what we'd see is higher case positivity yes.
However, there could be a situation where the very worried well were testing more and more, and keeping the testing numbers up, thus suppressing the figures. So I don't think it's absolutely reliable an indicator. False negative LFT's would still be in proportion to the positives in every other time, and positivity in LFT's is declining, that hypothesis doesn't fit.
The antibody stats from Bordois saying that 88% as a minimum of 16-24 year olds have antibodies is at the likely herd immunity levels for the age group, so as this group has the most contacts, this will have a huge dampener on spread. The only way we could not see a reduction with these levels is if the immunity from those antibodies is poor - now that is possible because all the tests for antibodys, and the vaccine are against the original covid, and delta might be different enough to evade it.