A. Life is a bowl of cherries. Really:
Three-in-one cherry tree, from the front yard orchard. Yes, the could be riper, but the birds are eating them as soon as they get really red. Plus, while the Bings should be almost black, the other two varieties don’t get much redder than those above. And they taste good.
A young lady we’ve known for years came by every day to feed the cat and water the gardens. She did a good job. While we were gone, the cherries hit their stride. It’s only one tree, so we’ll only get a few bowls worth per season – but fun. Next up: apricots and peaches, probably end of the month.
B. Back from the Epic Wedding Trip. 7 days, 6 nights, 4 states not counting airports and home. Some pics:
C. In New Hampshire, the spell of the magic mask talisman has been suspended – one can go about bare-faced and walk up to people, and the gods, we have been assured, will not be offended; cross the state line into Massachusetts or Vermont, however, and the wrath of the gods will descend upon any who dare sally forth with undiapered visage.
For now. Our betters are pumping the brakes, mixing it up, because, as any animal trainer will tell you, being predictable with your rewards does not get as eager a compliance as keeping the animal guessing. To add to the hilarity: when the New Hampshire folks decided to remove restrictions, they didn’t just announce: “OK, nobody’s dying of the Coof anymore, so go ahead and take off your masks and feel free to walk up to people and shake hands.” Nope, that would be too easy. Instead, it was *scheduled* for Monday, May 31. As in:
D. Speaking of terrified, scientifically illiterate rabbits doing as they’re told, I’ve got a massive post to drop in the next day or two about analyzing risk. Sometimes, I think I’ve been uniquely prepared for the COVID hysteria:
- worked in the actuarial department of a major life insurance company, picked up some basic knowledge of how risk is measured;
- worked as an underwriter and and underwriting analyst for a few years, so I know how the pros apply those risk models;
- used and helped design mathematical models for 25 years, and taught people how to use and understand them (I can literally say: I wrote the book (well, a fat pamphlet) on a couple fancy models used by thousands of people to do fancy financing).
- analyzed and cleaned up data for these models so that it was useful. Unless you’ve had to do this sort of clean up on real-world data, you simply have no idea how much sheer judgement goes into what gets measured and how. E.g., financial reporting systems are about as well defined, well-tested, and well funded as any data systems anywhere. Every company has one or more, with trained professionals inputting data, and have been doing this for decades. Yet, a data dump of the raw inputs is chaotic, unclear, and confusing. The question I had: what cash flows took place when? Surprisingly hard to answer! Correcting entries are ubiquitous, and often raise their own questions. And so on.
- read a bunch of medical studies. When our kids were babies, I, like every other new parent in America at the time, was constantly ordered and shamed to not let the baby sleep in our bed with us. But I knew that this practice, called a family bed, was common everywhere else in the world. So I searched around, found the studies, and read them. Insane. Bad methodology, dubious data, poor analysis, no criticisms and answers (meaning: a study should address the obvious criticisms and answer them – it’s called science.) Just out and out junk. Yet – and here’s the real eye opener – a protocol had been developed from these two junk studies, and every freaking pediatrician in America was pushing the no family bed nonsense. It’s Science! It’s the medical consensus! Also read a few studies on salt and blood pressure, and was likewise unimpressed. Then noted how nobody did studies on drug interactions until it was clear such interactions were killing people – who’s going to pay for such endless studies? I reached the conclusion, since backed up by all the failed attempts at replication, that medical studies are mostly – useless? Wildly overconfident? Wildly over cautious? Not to be taken at face value?
With that background, and an amateur’s love of the scientific method, I was not buying the claims of pandemic, the outputs of models, the cleanliness of the data, and the ‘logic’ for panic and lockdowns. Looking into it, it was puke-level idiocy. And yet, here we are.
E. Briggs captures a good bit of what I’m trying to say in my upcoming post on risk analysis in this week’s COVID post:
Many people sent me this Lancet note about the difference between relative and absolute risk reduction. I’ve warned us many times to use absolute numbers (in any situation, not just this), because relative numbers always exaggerate (unless one is keenly aware of the absolutes).
Here’s an example. Suppose the conditional (on certain accepted evidence) risk of getting a dread disease is 0.001, or 0.1%. A drug or vexxine is developed and it is discovered (in update evidence) the risk of getting the disease is now 0.0001, or 0.01%.
The absolute risk reduction (ARR; conditional on the given evidence) is 0.001 – 0.0001 = 0.0009, or 0.09%.
The relative risk is a ratio of the two risks, and the risk reduction ratio is 1 minus this, or 1 – 0.0001/0.001 = 0.9, or 90%.
That relative 90% reduction (RRR) sounds much more marketable than the actual 0.09% reduction; indeed, it sounds 1,000 times better!
Here from the the Lancet piece are some numbers using published results, recalling, as the authors do, that everything is conditional on the evidence, which is always changing.
|Johnson & Johnson||67%||1.2%|
For instance, the CDC says only 300 kids 0-17 died with or of coronadoom (a terrific argument kids don’t need to be vexxed). Population of this age group is about 65 million. We don’t know how many infected or exposed or this group, but you can see that differences between vaccinated and unvaccinated kids would be very small.
Read the whole thing. I only dare write anything on something the esteemable Briggs has already written on because even this level of math is off-putting to some people. I focus on the narrative part – why is it that huge reductions in risk might be meaningless, when the underlying risk is originally very small, as in the COVID risk to kids 17 and under. When pestered by a friend about why I’m not getting the vaccine, I replied: I will not take experimental drugs to lower my risk of death from COVID from something like 0.01% to 0.005%. She immediately changed to the ‘protect others’ tack, so I let it drop.
Alas! If information mattered, we wouldn’t be in the state we’re in.
F. And then there’s this. And this. I tend to go data=>analysis=>political speculation, or perhaps claims=>evidence=>reasons/explanations=>politics. Therefore, I have only really lightly touched on the politics/corruption/coup aspects of the Coronadoom – because I foolishly keep expecting people to care about the truth of the claims first. Yet ‘truth of the claims’ is nowhere to be found in the thought processes of the many, who instead substitute ‘whatever belief maintains my good standing in my group.’ Most people seem to go my social group’s position=>politics. Don’t ask why you need to raise your hand and get permission to go to the bathroom – JUST DO IT, DAMMIT! That sort of training, where group position is paramount and approval is always contingent on mindless obedience, is a large part of what got us to this point.