I’m stealing this from Spectator, which has done science right. It shows the UK’s SAGE Expert models for covid. In each case, the gray and others colors are the model predictions, and the red what happened in Reality. Except in the “‘Winter scenarios’ for 2020 Covid deaths, assuming no lockdown”, where Reality is thick black.
What can we say? We can say exactly what I was shouting in fury since the beginning of the panic. Yes, day one. Some of us from the very beginning urged calm. Alas, the world seemed to want panic. Experts surely did, as we documented again and again.
We can say that there was no reason to blindly trust Experts or their models. That there is never a reason to blindly trust them. We can say Experts see what they want to see. We can say that Experts, above all people, want to be needed. We say that all models, even Expert models, only say what they are told to say.
We can say that panic destroys the ability to think. That lust for power and recognition makes bad science. That confirmation bias always happens, even to Experts who have insured it will not happen to then.
What else can we say? Stay with me as I interject an anecdote.
I used to play petanque in Central Park with a slew of French restaurant workers, way back in the old days. I can still count to thirteen in French. And swear.
The idea is to get your manly steel balls as close as possible to a small wooden one. Either your ball is closer to the “little one”, or the other team’s is.
Often it wasn’t clear whose ball was closest. All of us would hover over the balls, viewing them from every angle, trying to see who won. Now I promise I did try to be objective in this, especially when no money was involved. But I cannot count the number of times I was certain sure our team’s ball was closer, only to have a measurement disprove me (we used collapsible radio antennas).
This was just a game. Nothing really was at stake. Even so, tempers could flare. This is why the decision was left to a neutral measurement.
I never could figure out why I was seeing the wrong thing. That is, was I standing too close, or too much to the left, or whatever. But I did know I was seeing the wrong thing after the decision was taken out of my hands.
The lesson is obvious enough, for all love.
What’s the legal dictum? Nemo judex in causa sua. No man can be the judge of his own case. This principle is sound for all the obvious reasons.
Those who have an interest in the outcome of a game cannot be trusted to declare which side has fit the rules best. We employee referees for this. Even professional referees do not get it right every time, but their presence all admit is necessary.
Why isn’t it the same in science? Professional referees are desperately needed. No scientist should be the judge of his own theory.
In the covid panic, the Experts, like the appalling Neil Ferguson, were trusted because they were Experts. They did math. On computers! It helped immensely that the message their models were selling was that rulers needed to seize power. Nobody was encouraged to look too closely at guts of the science they were selling.
Some of us saw the flaws, but our warnings went unheeded. Because we were not Experts. Which are those with expertise given authority and control. We had no authority and no control, and thus had no standing to produce formal criticisms. Some who questioned Expert wisdom even ended up being investigated by “counter-terrorism agencies“!
We are asked to believe not just covid models, but “climate change” models, economic models, models, models galore. And we put in charge of evaluating those models the same groups of people who create them. This should not be.
“Briggs, peer review does exactly what you want.”
Come over here, behind the shed, so I can show you something.
Peer review forsooth. It is not independent. And it never, and cannot, judge how predictive models are. Open public professional refereeing of models, which are are asked to follow, must be mandatory. The process must be adversarial, like the jury system.
Science, they say, is “self correcting.” It is not. But it could approach something like it by requiring independent proof.
This will not happen, of course. Imagine having independent referees of science models which say slicing the breasts off of young girls is good for their “mental health.” Those referees would have to go into hiding.
If they lived.
Subscribe or donate to support this site and its wholly independent host using credit card click here. Or use the paid subscription at Substack. Cash App: $WilliamMBriggs. For Zelle, use my email: matt@wmbriggs.com, and please include yours so I know who to thank.
I’d like to see someone take the real world data and back engineer a model that would mirror the actual outcomes. Perhaps these arrogant assholes would be able to determine where they went wrong in their initial models and learn from it. But nah, there’s no money in that. Who wants to “know”? It’s better to be wrong and paid, than right and poor.
Best line in this Substack - which I will squirrel away for future use -
“Because we were not Experts. Which are those with expertise given authority and control.” Simple yet profound beyond concise appearance.