Can we finally stop believing computer models now?

We’ve long known that climate models are (to use a modeller’s own words) “useless junk”. But climate models have been a paragon of accuracy compared to pandemic models.

As a Guest Post on The BFD noted recently, computer models can be very useful. But only insofar as they are “tweaked to match the real world”. Which sounds an awful lot like post-hoc fudging to me. Still, some companies place great store by commercial computer models. Academic models, much less so.

Part of the problem is that no academic ever gets sacked for being wrong. Prof. Neil Ferguson has been wrong with his computer modelling more often than I’ve not guessed the winner of the Melbourne Cup, yet, astonishingly, he hasn’t been sacked and sent to earn a far more productive living as a street sweeper.

When Britain declared its “Freedom Day”, public health troughers shrieked that it would unleash a Biblical carnage. Their modelling said so! But what really happened was that “case numbers” (by which they actually mean, not sick people, but apparently positive results to a PCR test) plunged.

Australians and New Zealanders are being hectored by the same bureaucratic bullies wailing and tearing their hair that ending undemocratic, authoritarian Covid restrictions will see us dropping like flies.

Once again, reality has other ideas.

Vaccination is working to reduce the number of people requiring hospitalisation and ICU treatment with Covid-19, and now the pressure on the hospital system looks set to be markedly less than predicted by government-commissioned modelling.

At current lockdown settings in NSW, government modelling predicted there would be 560 people suffering Covid-19 in intensive care units across the state by late October […]

Victoria’s hospitalisation rates are also well below modelling estimates.

In fact, the number of ICU Covid patients has halved before October has even begun.

It’s the same story in supposedly pestilential Victoria: just 75 Covid patients in ICU – just 15% of its ICU beds (that’s without the 4,000 extra beds Dan Andrews promised but still hasn’t delivered).

For all the scaremongering over “Delta”, the numbers tell a very different story: the hospitalisation rate in Victoria is half what it was during last year’s disastrous outbreak. Deaths are just 15% of what they were in Victoria last year.

So why the continued hysteria? It’s almost like there’s an agenda at work.

Australian National University microbiologist and physician Peter Collignon said he believed that modelling from the Burnet Institute – which has been a vocal proponent of “Covid zero” – was too pessimistic. “It does look like the case numbers and the hospitalisation numbers are quite different to the modelling,” he said.

“A lot of the modelling is very pessimistic. The people who are making the worst predictions are the ones that are pushing zero Covid the most.

Funny about that.

But don’t expect a pandemic dictator to admit the truth.

However, Victorian Premier Daniel Andrews said he believed modelling predicting a 10 per cent hospitalisation rate was “holding up”.

The Australian

Hey, who are you going to believe? Reality, or politician with a computer model?

Please share this article so that others can discover The BFD

Punk rock philosopher. Liberalist contrarian. Grumpy old bastard. I grew up in a generational-Labor-voting family. I kept the faith long after the political left had abandoned it. In the last decade...