- furman believes (or claims to believe) the original tweet was being deceitful by presenting a curve fit as a curve fit but leaving enough ambiguity that casual viewers might interpret it as a prediction of future deaths dropping to 0 by mid may

- philipson believes (or claims to believe) furman thinks that curve fitting and forecasting are the same thing, does not address whether the original tweet was attempting to be coy or deceitful

- article believes (or claims to believe) that philipson believes that a curve fit actually is a good prediction of the future, then goes super wide with it to make some vaguely related point about academia

I believe no one is lying or stupid, but everyone is finding the worst possible interpretation of events so they can dunk on each other.

"Rashomon (羅生門, Rashōmon) is a 1950 Jidaigeki psychological thriller/crime film directed by Akira Kurosawa...

The film is known for a plot device that involves various characters providing subjective, alternative, self-serving, and contradictory versions of the same incident."

Now that is an esoteric reference!

my only regret is that now that I understand it, i realize it's not interesting (to me personally)

The problem they are ridiculing is mainly that an observed time series that is going up and down with no clear trend is "modeled" with a polynomial that necessarily goes down, and that the tweet claims there is some value in the fitted polynomial.

Secondarily, the fact that the polynomial is described as cubic, when it appears to be quadratic.

Added together it makes it extremely explicitly clear that at the highest levels of American government, even in situations with the involvement of elite academic advisors, the actual technical content is utterly incompetent; childish; totally illiterate from a statistical point of view; bad even at high school.

I don't know what that says about any of us.

If an analysis is making a reasonable effort to use the best or even some sort of relevant methodology in the field maybe. There is an entire field built around the learnings and failures of previous work that informs how we do things now.

In this circumstance the economist came in with no understanding or desire to understand any of that and through some random excel function at the problem to get the answer out they wanted for political expediency.

Not to mention the prediction was ridiculously stupid and anyone with common sense let alone an epidemiology degree could see it was going to be wildly incorrect within days.

THAT is the model being talked about at the highest levels of government when you have the entire field of the world class epidemiologists at your fingertips.

Trying to “both sides” this one because scientists who have spent their whole lives trying working on this got a little snarky is just missing the boat.

There are plenty of other places where epi Twitter is having productive cordial discussions.

wut. a quadratic only has one inflection point. that plotted curve clearly has 2.

The problem of today is not most of America have never fitted a curve and not understand the difference between training accuracy and prediction accuracy. But that they choose to not believe or even hear out those who did.

Who are they supposed to believe here?

There's a right-wing academic stroke political appointee, and a left-wing academic stroke political appointee. Both have genuine academic credentials. Both are saying something that supports their political masters. One appears to have been ousted by the other so is probably bitter about that.

If you don't know about statistics, which as you say is most people, this is two equivalent people having an unseemly fight on Twitter.

Did you notice neither presented an actual argument? Just abuse.

Like climate change, but people chose to not to listen to the scientific consensus anyways.

p.s. The fact being that comparing training accuracy to prediction accuracy is something that is simply unsound. You can have hundreds of ways to spin a statistic that is backed by academic research, and comparing those two are not. It also happen to be the first thing you're taught to avoid.

But you said people don't know the facts themselves! So how do you want them to know which of these two people has the facts?

I mean they literally say the same thing about each other - 'new low...'. There's no information to action here if you don't know about statistics! Even if you decide to check their authority and motivation there's still nothing to divide them on!

Science doesn't work by consensus. Science works by having a track record of accurate predictions. So when you see people talking about "scientific consensus", that should immediately be a red flag. Valid science doesn't talk about "consensus" at all; it just points at the predictive track record--which requires not just "facts" but a series of accurate predictions, made before you knew the facts, that match with the facts--and lets you draw your own conclusions.

Human beings are at the very best arrogant and fallible by nature, incapable of truth. When you want to improve your model of the world by including some heretofore unknown to you concept,fact, or set of facts you can opt to learn everything from the ground up in order to develop a deep understanding of the topic or accept or slot in some preexisting truths and models as described by others that you presume to be true.

This presumption of correctness is typically based on their standing with yours or preferably with their own peers combined with your assessment of them based on how well their statements comport with things you know or at least believe to be true. This is typical because the world is incredibly complex and our time here is finite.

Even smart skilled people have to lean on option two a lot outside of their particular area of expertise. When intelligent people do this they ask themselves whose views do people skilled in a particular area think are worth listening to or what on average do people skilled in this area say about something. This is what is meant by scientific consensus. Unintelligent people ask themselves what do my fellow unskilled peers think about this or what do I already think is true and are their any experts who confirm what I already want/believe to be true.

When people say that the scientific consensus is that cigarettes cause cancer they mean I haven't fully examined the complexity of the human lung and the effects of carcinogens on same but I accept the fact that many experts have done so and are telling me that If I keep smoking I'm more likely to die of cancer. This is converse to the person who also doesn't have time to understand how lungs work who eagerly looks for someone with credentials who says its OK if I keep smoking.

People talk about scientific consensus precisely because in a broad population of users you can find at least one party with any given credential willing to espouse any given stupid thing for money or for kicks. It's especially useful if you can get someone who actually IS smart and therefore respected in one area to believe he knows something about a field totally outside his area of expertise and lend existing cred to a stupid idea that an actual expert would dismiss. This strategy is very commonly on display in the discussion about climate change for example.

No, I am talking about how groups of people discover increasingly true pictures of the world. They don't do that by consensus; they do it by finding models that make more and more accurate predictions, as shown by the actual track record of accurate predictions.

> When you want to improve your model of the world by including some heretofore unknown to you concept,fact, or set of facts you can opt to learn everything from the ground up in order to develop a deep understanding of the topic or accept or slot in some preexisting truths and models as described by others that you presume to be true.

Or, instead of making any assumptions, you can look at the actual predictive track record to see which "preexisting truths or models" actually work and which don't.

The reason this isn't obvious to most people is that most people don't stop to think about how much of their everyday experience, particularly in this age of computers and GPS and other technological marvels, actually gives them a huge track record of accurate predictions for our fundamental scientific theories. If our predictions based on models using General Relativity were not accurate, GPS wouldn't work. If our predictions based on models using quantum mechanics were not accurate, computers wouldn't work. There are countless other examples. Most people don't stop to think about this so they don't realize how high the bar actually is for having a track record of accurate predictions. They think of GR and QM as esoteric physics, not as everyday realities. They don't realize how huge a volume of evidence from their direct experience they already have for these theories being correct, so they think they have to take physicists' word for it, when they actually don't. Which means they also don't realize how much other people, who seem to be just as sure of themselves and their predictions as physicists (if not more so), actually are just overstating their case, often by many, many orders of magnitude.

So I reject your model of how people should actually assess claims in areas where they don't have expertise.

> When people say that the scientific consensus is that cigarettes cause cancer they mean I haven't fully examined the complexity of the human lung and the effects of carcinogens on same but I accept the fact that many experts have done so and are telling me that If I keep smoking I'm more likely to die of cancer.

When people assess the probability that if they smoke they will increase their risk of dying of cancer, they have no need to rely on any "consensus". They can just look at the data.

The original graph showing the "cubic" curve is just - pathetic and sad. It's the equivalent of using a sharpie to change hurricane trajectories. And then Philipson defending this nonsense by calling someone else an "economist turned political hack"????

How the hell can so many of these people have no shame? I totally agree with a follow-up tweet - doesn't matter what Philipson did in his whole career, he will be remembered for completely abdicating any decent sense of professional ethics.