We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

Samizdata quote of the day

The Ferguson – or Imperial – coronavirus model is a load of Hooey. But not, or not alone, for the reasons generally given that it’s a tangled mess of code that doesn’t even produce the same answer each time. Nor because its output was so useless that even the originator wouldn’t obey the implied rules from its use when seeking a shag.

No, Ferguson failed because his model failed to include human beings in it. Which is really very weird indeed when attempting to model, erm, human beings.

Tim Worstall

22 comments to Samizdata quote of the day

  • APL

    “No, Ferguson failed because his model failed to include human beings in it.”

    It didn’t fail, it was very successful as part of a portfolio of terrorist instruments. Along with the BBC ‘The total number of deaths in the UK has reached an all time high of ..” blather, day after day after day. But never, ‘x number of people have died in the UK today from COVID-19, but as a proportion of the population that is nearly negligible’.

    If there is one lesson from the COVID-19 apart from, don’t believe an academic or his model. It’s the BBC should be disbanded, it’s buildings raised to the ground and the land they stood on salted.

    We can no longer afford the BBC. And anyway it has been a highly corrosive cultural instrument.

  • djm

    Clears throat…….. razed

    Apart from that, an entirely reasonable summation of why the BBC has to go

  • Snorri Godhi

    There is a much more important quote in the same article:

    Lawson […] point[s] out that the damage to the economy has not all been caused by the lockdown but at least in part by that change in behaviour. […]

    The obvious implication of this being that the lockdown has been very much less costly than generally thought, because the effects of the lockdown are only those additional losses that stem from it, we must subtract the losses from our initial, unforced, changes in behaviour from the total to get to the additional costs of the lockdown.

    In other words, the Imperial model has been useless, rather than disastrous. A distinction that people who can only think in black vs white will fail to appreciate.

    Although Tim Worstall seems to contradict himself in the very last sentence:

    We run the entire nation on models that we know are wrong?

    No, “we” have NOT run the nation on any model, we have run ourselves on commonsense.

  • Snorri Godhi

    A review of articles on the lockdown debate can be found on National Review (h/t Instapundit).

    The articles are conveniently divided into 3 sets, depending on the conclusions:
    Lockdowns work;
    Lockdowns don’t work;
    It depends on what the meaning of ‘lockdown’ is.

    So you can avoid articles reaching conclusions that you don’t like.

  • APL

    djm “Clears throat…….. razed”

    Thank you. There was an ” it’s ” there, where it had no business, too.

  • Stonyground

    Razed to the ground.

  • Stonyground

    Isn’t failure to take account of human nature the reason why socialism is always doomed to failure?

  • Snorri Godhi (May 25, 2020 at 2:18 pm), I guess Tim’s point was that the imposed rules, where they caused changes beyond already-chosen public behaviour, were based on the models, which were wrong. I’d say the models, and the propaganda they induced to impose the rules, alarmed the public beyond what the virus reporting was already causing. 500,000 dead is more alarming than 50,000 dead – and any order-of-magnitude error is likely to affect the public domain in unhelpful ways.

    Its a bit like being a fire-walker: if you feel so confident it will work that the soles of your feet stop sweating with fear, then you will be burned. If we had been so blase that no-one changed their behaviour at all, then some number (only randomly related to Ferguson’s rubbish guess, but larger than today’s) of people would have died from ChiComCold and some other number would not have died from deferred health checks and hospital, economic or mental side-effects. Left to themselves, people would have adjusted behaviour in the face of both threats, to avoid the bug at one time and then backing off to avoid the side-effects at a later time. Instead we have the government managing these changes in its usual cluncky way and now a bit hindered by the excessive alarm it has generated.

    I’m not in the least an anarchist: you can say more for government’s role including pandemic response coordination than much else it does. We’ve had a chance in this quarter to see how it partakes of the general fallibility of government.

  • Outstanding summation of the situation, Niall.

  • Snorri Godhi

    Niall: it seems to me that you (and Tim Worstall, if your interpretation is correct) are trying to have it both ways.

    As i understand, you accept that the change in people’s behavior is spontaneous, and yet you blame the Imperial model for changing people’s behavior, and the government for the damage resulting from such changes.

    And how can you say that the predictions were rubbish? You don’t know how many people would have died w/o a ‘lockdown’, either spontaneous or mandated. Extrapolating from Bergamo province (where any behavioral changes were not motivated by any model) i’d say that about 400K excess deaths is a realistic prediction for the UK — and it could have been worse than that, because there is more obesity in the UK than in Italy.

    For me, the real division is between countries that acted early and countries that acted late or not at all. There is little difference between Italy, the UK, and Sweden in my eyes. And there will be little in your eyes when you’ll look at this.
    Oh, and the Swedish economy is suffering too: we’ll have to wait and see whether it suffers more or less than those of Sweden’s neighbors.

  • Chester Draws

    Extrapolating from Bergamo province

    Cherry pick much? You have taken the worst affected place in the world, where they did most everything wrong* (mixing kids with grandparents, sending infected to rest homes, letting people spread it when they fled to home villages and country homes, not actually stopping many work places, not isolating properly in hospitals) and then blown it up a bit more for effect.

    Why not extrapolate from Florida? Even Naples?

    * they weren’t to know many of these things were wrong, but they still did them.

  • Snorri Godhi

    Why not extrapolate from Florida? Even Naples?

    DUH! Because Florida and Southern Italy acted early (relative to time of first deaths) and Lombardy didn’t, and i was specifically estimating what could have happened if Boris acted even later.
    Or rather, what would have happened if British people got scared even later.

  • Nullius in Verba

    “I guess Tim’s point was that the imposed rules, where they caused changes beyond already-chosen public behaviour, were based on the models, which were wrong.”

    Possibly. Tim said:

    “But there’s a corollary to this too. We changed our behaviour without being forced to – therefore we cannot attribute the changes in behaviour to, entirely at least, the lockdown. Which is why Ferguson’s model was wrong. For it had only two states, lockdown or normal. It did not contain reality, changes in behaviour without lockdown.”

    The model of course had lots of states besides ‘normal’ and ‘lockdown’ – interventions modelled included: “PC=school and university closure, CI=home isolation of cases, HQ=household quarantine, SD=social distancing of the entire population, SDOL70=social distancing of those over 70 years for 4 months (a month more than other interventions).” But then Tim, like most people, isn’t basing his opinions on having read the paper, looking at the model, or understanding what was done. Most of this seems to be based on memes and media misunderstandings of what happened. (That’s a persistent feature of events.)

    In addition, the model was not even attempting to model people’s motivations, or what the precise rules were to be. The already observed speed of spread implied we were headed for a very rapid peak that would overload the health service and kill more people. To prevent that, the R number would have to be reduced very rapidly by more than 60% to get R below 1, neither of which conclusions depended on the models. The model was then used to figure out what combinations of social contacts would have to be cut off to achieve that, and indicated that it would take a pretty drastic degree of separation/isolation. (And that has been confirmed with hindsight, since even with the lockdown the number of deaths is declining only very slowly.) The model says nothing at all about how that is to be achieved – whether voluntary or by government dictat. It just says: “if people behave this way, that is likely to happen”. And of course the model has huge uncertainties about that, which is why the paper itself said you couldn’t rely on the models for this, you would have to guide events based on trying stuff and observing the effects.

    Figuring out how to get the people to behave that way was largely down to the politicians. I think it would have been obvious from the first they didn’t have the security infrastructure to enforce it. Which is why they used engagement, persuasion, and a “we’re all in this together” sort of message to get voluntary compliance. The rules were imposed I think largely to convince people this was serious, and to stop all the other rules getting in the way. For example, people may want to voluntarily stop at home, but their employer wants them to come in to work. They’ve got contracts and commitments. It’s a lot easier to get out of that contractual obligation stuff if you can say to your employer or creditor “the government told me to”. People follow the rules, so if the rules are stopping people complying voluntarily, you’ve got to change the rules officially.

    But it seems Tim has just spotted what I thought was obvious from the first – the ‘lockdown’ is based primarily on voluntary compliance by the largely willing and convinced, not on state dictat taking away an unwilling people’s freedom by force. It’s a ‘prison’ where none of the doors are actually locked.

    “500,000 dead is more alarming than 50,000 dead – and any order-of-magnitude error is likely to affect the public domain in unhelpful ways.”

    It’s not an order of magnitude error. Roughly 1% of people who catch it die (so long as they get hospital treatment). If 8% of the population catch it, 50,000 would die. If 80% of the population catch it, 500,000 would die. The paper did NOT claim that after 8% of the population caught it, 500,000 would die!

    This isn’t an order of magnitude error in the paper, it’s an order of magnitude error in people’s reading comprehension! That it still persists – even here, after lengthy discussion of the error – is testament to the universality of human nature. But they also say that to forgive is divine, so I’ll desist from expressing my feelings about it any more strongly. 🙂

    “We’ve had a chance in this quarter to see how it partakes of the general fallibility of government.”

    It’s because the government is made up of humans. The fallibility is not specific to government.

  • It’s because the government is made up of humans. The fallibility is not specific to government.

    Sure but in most cases, only governments gets to require you follow their fallible notions at gunpoint.

  • Nullius in Verba

    “Sure but in most cases, only governments gets to require you follow their fallible notions at gunpoint.”

    And lynch mobs of vigilantes. And criminals. And rebellious citizens standing up for their revolutionary rights against a tyrannical government.

    But I agree. The fallibility of humans in government is usually a more serious affair, because they have more power, and because they are less subject to the correction of the market. When private enterprise screws up, which it often does, it goes bankrupt and stops doing it.

    What I mean is that it is an error to think government is the cause of the fallibility, and so replace government with some other way of organising humans to make collective decisions that is equally fallible. Like “stuff I got sent on Twitter”. Or “I read it on the internet”, or “in the newspapers”, or whatever.

    Institutions of government like representative democracy and constitutional limits makes some flawed attempt to mitigate our universal human fallibility. Whatever you set up in its place has to be aware of the problem and do a better job.

  • Marius

    i’d say that about 400K excess deaths is a realistic prediction for the UK — and it could have been worse than that

    I’d say that was bollocks on stilts.

    Roughly 1% of people who catch it die

    Talking of nads on poles. ‘Roughly’ is doing a lot of work there.

    This isn’t an order of magnitude error in the paper, it’s an order of magnitude error in people’s reading comprehension!

    Nope. The model claimed 500k would die without some sort of intervention. This is tripe, regardless of how you read it.

    If we are throwing random numbers out, then I predict that when the dust has settled it will be clear that – once the deaths by PHE/NHS are taken out of consideration – COVID-19 is as deadly as a nasty flu, but by no means the nastiest in living memory. I also predict that lockdown morons will claim any difference between what actually happened and the end of the world was thanks to the lockdown.

  • Snorri Godhi

    Following up on Nullius @10:28am:

    The already observed speed of spread implied we were headed for a very rapid peak that would overload the health service and kill more people. To prevent that, the R number would have to be reduced very rapidly by more than 60% to get R below 1, neither of which conclusions depended on the models.

    This is an important point. People like me, who used a spreadsheet to make a semi-log plot of numbers of deaths over time, are people who (a) have some mathematical skills, obviously, but more importantly (b) do not trust government to tell them the level of danger. Once you have a semi-log plot, you can estimate both the danger from the virus and the danger of a lockdown.

    I knew that Boris would impose a lockdown before Boris himself did. If you didn’t, then you are reduced to railing impotently against the government, the way that a stereotypical Irishman is supposed to rail against the rain.

    You should not trust the government, either to protect you from the virus or not to impose a lockdown. That is independently of whether a lockdown is good, bad, or irrelevant; and of which you think it is.

    But it seems Tim has just spotted what I thought was obvious from the first – the ‘lockdown’ is based primarily on voluntary compliance by the largely willing and convinced, not on state dictat taking away an unwilling people’s freedom by force.

    Actually:
    (A) Tim did not spot it, it was spotted for him by Dominic Lawson.

    (B) I did not think it obvious, because it does not explain why countries whose governments acted early seem to have suffered less than countries which acted late.
    I can think of alternative explanations for that; but still, it is not obvious that government action is of little relevance.

  • Snorri Godhi

    Still expanding on Nullius’ comment:

    This isn’t an order of magnitude error in the paper, it’s an order of magnitude error in people’s reading comprehension!

    That’s a good way of putting it.

    That it still persists – even here, after lengthy discussion of the error – is testament to the universality of human nature.

    I don’t think it’s human nature: i really think that it is the modern Western diet that is making us prone to delusional insanity.

    Faulty reading comprehension is a form of delusional insanity.

  • Nullius in Verba

    “Talking of nads on poles. ‘Roughly’ is doing a lot of work there.”

    You want me to be more precise?

    So far, about 6% of the UK population has caught it, which is about 4 million people, and around 37,000 have died, which is 0.93%.

    In New York, about 21% of the 8 million population caught it and 16,000 died, which is about 0.95%.

    It’s about 1%.

    “(A) Tim did not spot it, it was spotted for him by Dominic Lawson.”

    Good point.

    “(B) I did not think it obvious, because it does not explain why countries whose governments acted early seem to have suffered less than countries which acted late.”

    I agree that there are other reasons for that.

  • Paul Marks

    Both the American “Tony Heller” (not his real name) and the Irishman David Cullen have produced videos on this – very good ones.

    There may, in March, have been an excuse for supporting the “Lockdown” – good people were on both sides of this debate.

    HOWEVER, now that the “evidence” for the various “Lockdowns” has been exposed as a fraud, and a fraud motivated by Collectivist objectives, the Red-Green Agenda 21 and Agenda 2030, there is no excuse for continuing to support “Lockdowns”.

    Some people, I will not give any names in this comment, seem incapable of saying “I was mistaken” or even “I was mislead”.

    Fair enough, perhaps some people are just incapable of admitting that the “Lockdowns” were a terrible blunder and were pushed by fraud – fraud motivated by the darkest of motives, the desire to spread tyranny.

    We must move on without these people – if they choose to live in a river in Egypt (continuing to support the insane “lockdown”) that is their affair.

    “You would sing a different tune if you were dying Paul”.

    No I would not. And I may well die of the virus.

    My death is not relevant – what is relevant is that the “lockdown” must end – there must be no totalitarian “New Normal”. People who support the totalitarian “New Normal” must be defeated.

  • Plamus

    Niall Kilmartin:

    I’d say the models, and the propaganda they induced to impose the rules, alarmed the public beyond what the virus reporting was already causing.

    Relevant. Money quote: “Third, the more infectious people believe that COVID-19 is, the less willing they are to take social distancing measures, a finding we dub the “fatalism effect”.”