Updating the model

I blogged back on January 13 about the latest "modelling data" from the Ontario Science Table, the expert group that has been providing advice to Ontario Premier Doug Ford about how to combat the pandemic.  Ford had said ahead of a report from that august body that the fresh projections would make Ontarians "fall of their chair", and the numbers were indeed scary: 20,000 cases of COVID per day in the Province by mid-February.

I felt at the time that the numbers represented extrapolation rather than modelling.  Here's part of what I wrote:

I don't intend to disparage the Science Panel here, but this kind of modelling can produce some very weird results. One member of the Panel commented that a faster growth rate would potentially lead to a 40,000 per day case rate, as indeed it would. Then again, it happens that the number of new cases reported this week has actually been slightly below the levels seen a week ago. Is that a new trend? Maybe, maybe not, but simply projecting it forward would lead to a much different outcome from the one the Province is using to justify its decisions.   

Well, here we are in mid-February, so let's have a mini-inquest. There are some issues with data collection, but in round terms the number of daily cases in Ontario seems to be close to 1000, which means the Science Table's scary projection was off by a factor of twenty! I'm not saying this just to have some finger-pointing fun: after all, as the man said, forecasting is difficult, especially when it's about the future. But being this wrong can be damaging. Ford and his team constantly claim to be guided by the science, and here's a piece of that science that has been spectacularly wrong. It makes much harder for the Government to justify its policies -- you were dead wrong last time, Doug, so why should we believe you now?  

On Thursday the Science Table produced its updated modelling results.  They seem to have learned some lessons. The report now looks at two elements of the scenario going forward: the future of the original  COVID strain, which seems to be under reasonable control, and the risks posed by the so-called UK variant. It looks as though infections from the original strain will continue to fade across the Province, but the overall outbreak could start to ramp up again if the UK strain gets out of control. Keeping it under control means getting the R-number, which is currently around 0.9, down to about 0.7. If that does not happen, the overall case count could head towards 5-6,000 per day by the end of March.

That all looks a whole lot more like actual modelling and less like extrapolation. It's also not fall-off-your-chair scary, and it requires a little time and thought to understand. So, needless to say, it has hardly been reported in the media at all.