Because it highlights the futility of the lockdown?
Summary of UK advice and lockdown:
2 Feb: Matt Hancock issues handwashing/sneezing advice;
13 Feb: Chris Whitty says no behaviour change needed;
25 Feb: those returning from Italian quarantine zones advised to quarantine for 14 days;
29 Feb: Chris Whitty says school closures not needed;
3 Mar: Boris shook hands with everybody at a hospital;
7 Mar: Boris shook hands at Twickenham;
9 Mar: Boris issues handwashing/sneezing advice;
12 Mar: those with symptoms told to quarantine for just 7 days (
WHO advice is for the whole time you have symptoms plus 14 days after);
16 Mar: Boris advises to work from home and against travel, pubs and theatres; those with symptoms now told to quarantine for 14 days (so anyone already quarantined since 12 Mar would not be out yet);
17 Mar: Foreign Office advisory against nonessential travel abroad - but travel numbers reportedly already falling fast;
18 Mar: schools to close 20 Mar - widely reported that many parents had withdrawn their children before that;
20 Mar: entertainment and leisure venues ordered to close but it takes a day to make it legal;
23 Mar: lockdown announced but it takes two more days to make it legal;
27 Mar: Boris and Hancock test positive, Whitty develops symptoms.
If the peak really was obtained at 8 April and
the median time for symptoms to show is 5 days and
the median time between symptoms and death is 18.5 days, that may put the key change 23.5 days before the 2pm, so 2am on 16 March, which would suggest the key change was telling anyone with any symptoms to quarantine, rather than even the not-quite-Swedish lockdown bankrupting hospitality without a bailout/furlough scheme.
BUT! The counterpoint to this is that these are all probabilities and estimates and it's a long time between taking an action and seeing the reaction, so rather like steering a heavy boat, a heavier adjustment to the rudder may have been needed once we started to drift off course in order to bring us back sooner. It would have been a brave or foolhardy captain who stopped after the quarantine order and the expected cost of a small probability of crashing into full hospitals would probably have been higher than the expected cost of a probably-unnecessary lockdown.
Correction (29 April): I totally lost track that we are talking about the "peak" which is basically the "mode" and something like time-to-death will be very skewed, so the mode will be less than the median. In line with
https://www.cyclechat.net/threads/peak-infections-occurred-before-full-lockdown.260300/post-5975500 - this means that peak infection probably occurred at some point between 16 March and 3 April and we need more information about typical time-to-death before we can decide when is most likely. Any claims that it was one side or the other of 24 March probably tells you more about the claimant than the virus!
Statisticians and epidemiologists will probably argue about this one for decades. I think that, at best, we need more information about this virus and illness than we have even now to decide reliably when and how to unlock, let alone whether we locked down optimally.
(edited to add missing words and line breaks - edited to add correction)