Should we be worried about what the COVID-19 data is telling us in the UK?
There is a one word answer to this - "yes" - but as with most one word answers it fails to do justice to the complexity of the data.
For while it's possible to look at certain charts of case numbers and hospital populations and make some very scary noises - and while it's also possible (if considerably more difficult) to take some of the data and make the case that there's nothing to fear whatsoever - the reality lies in that annoying grey area which defies simple explanations.
Before we venture there, and on the basis that not everyone reads to the end of these articles, let's expand somewhat on that one-word answer. The majority of warning lights over the virus are certainly flashing red at the moment.
COVID-19 growth is rising at a rate we haven't seen since the spring, far faster than in the autumn, when it prompted the second English lockdown. Deaths are already high - higher than in almost every winter in previous decades, even after you adjust for population - and the likelihood is that they will rise further in the coming weeks.
Still: to judge from some of the analyses and nuggets you might have seen online you might have assumed that we are facing a situation even more dangerous than the one we faced in the spring.
This seems, well, extreme, for two reasons: the first is that the data doesn't support it.
Consider one of the bits of data that supposedly terrified the prime minister into initiating this third lockdown: the revelation that more than 80,000 people tested positive for COVID-19 on one day alone, 29 December. Not only was this the highest total ever, it was a terrifying leap from the previous days and weeks.
But here's the thing (and apologies if you've already spotted this) there is something special about 29 December that means we were always likely to record an outsize caseload on that day: it was the first working day after a long Christmas break.
In other words, 29 December was a backlog day, when a lot of people who might have liked to have a test during the Christmas period finally had their tests processed.
Now it's possible that as more data comes in, it turns out that the Tuesday in question turns out to be the first of many mega days for new cases, but it's also quite likely that it turns out to be a big bump which distorts the medium term picture of the growth of the disease.
As it happens, the latest data suggests that the daily total for the following day dropped to 70,000. That's still a scarily high figure in the grand scheme of things, but it's probably too early to understand how much of that is still a Christmas backlog effect.
Anyway, while it's certainly true that cases are higher, much higher, than they were during the spring, this is in part a reflection of the fact that we are testing far, far more than we were back then (indeed, the UK tests more than almost every other developed economy, which tends to make UK absolute case numbers look a lot higher than in most other nations).
A far better measure of the spread of the disease, in theory if not in practice, is to look at the percentage of people whose COVID tests come back positive. Look at that statistic - running at around 18% in England - and you'll see that while it's certainly high now, it was considerably higher in the spring, peaking over 40%.
I say "in theory not in practice" because even with this data there are some big question marks. Was the level in the spring a true representation of that outbreak or, given testing was relatively scant and was limited mostly to hospitals where many were already sick, might it overstate the peak back then? In much the same way, might the recent data be somewhat overstating the speed of the spread?
For when you drill down into the data you see something interesting: the number of tests being taken leapt before Christmas (presumably as people prepared to mix with their families during the festive season).
That pushed down the proportion of positive tests as, one presumes, a lot of those people were asymptomatic and wouldn't otherwise have been tested.
So while the recent numbers are probably a fairer reflection of where the country is, the steepness of the rise looks scarier than the reality. Another Christmas bump in the lines.
You see the issue here: the more one explores the data, the more question marks there are. Consider another scary statistic mentioned frequently in recent prime ministerial press conferences: the fact that there are more COVID patients in hospital now than in the spring.
This is technically true - that is what the data tells us. But then again: we were testing far less in the spring, so it's quite plausible that the COVID patient levels from back then are an understatement of what was really going on.
In short, we might still be some way short of the effective spring intensity, even if the numbers suggest otherwise.
Now as I say, when you take a step back from the data and adjust for all these discrepancies and question marks, even so the outlook looks concerning, in a way that it hasn't since the spring.
The spread of the disease looks real, even if the pace is somewhat less terrifying than the numbers might on the surface look.
Anyway, the data is not the only reason why this lockdown and this outbreak is somewhat different from the spring.
This time around the vaccine is not an idea which might materialise at some point in the future; it is a reality. In other words, while the lockdown of spring 2020 was a siege, with the country battening down the hatches to fight off the virus, this time around it is more like a race, with the vaccination rate rapidly ratcheting up.
But here too we have some data problems. While NHS England provides some detail about the ages of those getting the vaccines, we need to know more: do they have medical conditions, have they previously had COVID-19? Which vaccine are they receiving?
The more of this data we have, the better a sense we will get of how many people in the country are likely to be immune from this disease.
The smaller the susceptible proportion of the population, the more the R number should fall and the closer we should get to normality.