Imagine someone said: “Every year, a certain number of people die in traffic and it doesn’t matter if you educate 6-year olds or teenagers and how well-designed driving lessons are. The number stays the same.” Would you believe her? Of course not! But if she was a professor, an expert in statistics, and had a dozen graphs to prove her point? Still not!
If the professor had a Ph.D. in Philosophy and a BSc in Physics and Math, was an Army Major General, chair of a Space Agency and a National Council for Research and Development, head of a Security Studies Program at a University, and a former Member of Parliament? Still not!
Now, if you find that she was not pained by the loss of life in traffic but rather by all the costs of these educational efforts? Then you would have the shameless motive for saying such utter nonsense.
In real life, we have in Israel a professor with a Ph.D. in Philosophy and a BSc in Physics and Mathematics, who was a Major General in the IDF, chairman of the Space Agency and a National Council for Research and Development, head of a Security Studies Program at Tel Aviv University, and a former Member of Knesset for the extreme left Kadima Party.
He prior declared that he’s pained by … not the loss of life or the potential loss of life from COVID-19 but by the losses in the economy. “We are paying NIS 100 billion a month because of this closure.”
And then he comes with a theory (that he doesn’t test) that, no matter what one does, lockdown, social distancing or nothing, the epidemic will die down in 8 weeks. A dozen graphs must prove his absurd point.
The greatest “sin” in statistics is cherry-picking: that you select examples that make your stats come out the way you want. So, I only have to show one prominent example that does not fit his wishful “thinking.”
The example is quickly found. China. Followed by two more.
COVID-19 is short for CoronaVirus Disease 2019. The first infection was at the end of November 2019. The Chinese extreme 10-week lockdown lasted until the end of March 2020. Not 8 weeks. 4 months = 17 weeks!
Another glaring example that the professor leaves out. Taiwan. Six cases. Not 6%. Six cases on 23.6 million people!! Nothing 4 weeks exponential growth of the epidemic, 2 weeks plateau, 2 weeks disappearance. Why? Good anti-epidemic measures. No curve to flatten.
Israel is also an example where the disaster was contained early by the professor’s political opponent, Netanyahu. Of course, the lockdown did produce an enormous saving of lives. You don’t need statistics for that. One example suffices. If I’d have continued my normal life, I would have been infected and would have had a reasonable chance to die from it.
Now, it is possible that in Iran and Turkey, where the authorities did nothing to stop the epidemic, the infection ‘also’ dies down in 8 weeks. But that is not because it mysteriously would have no ‘power’ anymore to infect but because so many people then have been infected (and sent to bed to heal or die) that the virus won’t find uninfected people so easily anymore. But at what cost! Thousands of unnecessary deaths.
Reminds me of a joke our Dutch peerless politically genial standup comedian Wim Kan made during the Cold War. “Russians are experimenting with a sheep and a lion in one cage. It goes very well. Really. It goes very well. It cost a lot of sheep but it goes very well.”
You don’t need lockdown. An epidemic tempers out in 8 weeks anyway. Yeah. And then we would have had, like in Italy, Spain, the UK, and the US, thousands of deaths. For the economy, so it’s worth it? Scandalous!
The author knows statistics and their tricks but is totally clueless about 101 epidemiology, so he treats countries as uniform unites. Anyone with a little sechel (brains) can understand that, now the infection numbers in Israel are falling, some Arab villages in the North still need to go on total lockdown immediately because they just had an infection explosion.
Well-respected news outlets should not blindly trust anyone’s statistics, even from people with the highest qualifications. Rather, journalism is to ask uninvolved statisticians what they think of the statistical proofs.
I know that most journalists are great at language and for some reason by default often have trouble with 1 + 1 (It must be about 3), let alone, sophisticated math and statistics. That’s when you ask specialists. A journalist doesn’t just report. Not racist opinions, not capitalist opinions, and not scientific opinions — especially not when uttered by politicians who are opponents of the government they criticize and have self-declared financial and likely self-promoting motives for their positions.
But maybe I showed above that the greatest statistical nonsense doesn’t need a specialist to debunk it. A little common sense will go a long way.