Just last weekend, Prof Neil Ferguson told the BBC it was "almost inevitable" that Britain would soon see 100,000 Covid cases per day and possibly even 200,000 daily infections.
For months now, modellers have been warning of a devastating summer spike that could exceed the highs of previous waves. Sajid Javid, the Health Secretary, asked the country to brace for six-figure daily infection numbers.
Yet there are early signs that Britain has already peaked – and has done so at a far lower point than anyone expected. On July 15, the country recorded 60,676 cases, and infections have been largely falling ever since.
Although daily case numbers can be erratic, the seven-day rolling average is telling the same story.
The slowdown began on July 19, when there was just a 1.7 per cent increase compared to 5.5 per cent the day before. The increase was smaller again the following day, at just 0.5 per cent, before falling by 2.6 per cent on July 22.
On Friday, 36,389 cases were reported – a fall of 3,517 from Thursday and a 42 per cent drop from the previous Friday, when 51,870 were announced.
"We started to see a weak signal a few days ago and we’re now seeing a considerable dip from when cases hit 56,000 last week in England," said Prof Carl Heneghan, the director of the Centre for Evidence Based Medicine at the University of Oxford.
"This seems to be how viruses work – you see this exponential rise and then they seem to drop at a rapid pace."
How Covid infection modelled data in England misread the third wave
It is inevitable that hospital admissions and deaths will continue to rise for several weeks before they catch up with the fall in infections, but it does suggest the end of the third wave may be on the horizon.
It is already clear that Scotland has peaked, with a high of 3,930 recorded on June 28.
Curiously, that was just six days after the country’s football team was knocked out of Euro 2020, and could signal that the surge was partially caused by increased social contact during the tournament. It may explain why England, whose team reached the final at Wembley on July 11, peaked later, at 56,196 infections on July 15.
Paul Hunter, professor in medicine at the University of East Anglia, found there had been a "significant acceleration" in new cases about eight to 10 days after the first England game in the Euros and then after the quarter-finals.
"In Scotland case numbers started to fall after they left the championships, and case numbers have continued to fall since," he said.
"If such a perturbation as the Euros caused only a temporary acceleration in the increase in case numbers despite games continuing, this may bode well for the impact of [ending Covid restrictions on] July 19."
Prof Heneghan said: "We know significant events can act as a promoter of infection, and it’s interesting to note that Scotland peaked earlier. The fall in numbers is reassuring because we’ve had quite a lot of opening up already, as well as the football, and I can’t see that a great deal has changed in behaviour since then.
"It’s now time to overhaul the data and publish cycle threshold information so we have a better understanding of how infectious people are, and find out how many people are being admitted to hospital with Covid rather than being tested when they go in with something else. That would really help us understand what is really going on."
UK Covid infections and peaks
Overall, England has done far better coming out of Covid restrictions than any of the models predicted.
Scenarios suggested that even with the one-month delay to "Freedom Day", daily hospital admissions would be around 1,300 a day by now. Instead, 870 were recorded on Friday.
Modellers also forecast hospital occupancy of between 8,000 and 12,600 now and 200 deaths a day, yet there are currently 4,861 people in hospital, and 64 deaths were reported on Friday.
All this must be taken with a healthy dose of caution, because the July 19 unlocking may still spark a further upward trend that is not being seen in the figures yet.
The latest data from the Office for National Statistics (ONS) shows infection rates are continuing to rise, with one in 75 people now carrying the virus. However, the ONS data only runs up to the week ending July 17 and so will not have captured the recent falls.
This week’s ONS data also shows a rise of just 28 per cent between the week ending July 10 and July 17, compared to a 73 per cent rise between the weeks ending July 3 and 10.
It might be argued that the fall is due to reduced testing. The number of tests carried out recently has also been sporadic, ranging from 1.2 million on July 14 to 784,000 on July 17. But fewer tests may be indicative of fewer symptomatic people, and there is no sign of a delay in processing tests which could be causing a backlog of case data.
So it may be that there are finally enough people vaccinated to keep Covid under control. Put simply, the virus may be running out unvaccinated people to infect.