The general problem that the article raises is still valid even though the pictures don’t match.
The article is pointing out that the increasing circle sizes represent variation of centerline paths, not the actual size of the hurricane (which is a common misconception). Just because the plots don’t match doesn’t make the article irrelevant.
I get your point, but there is another angle here. People have a strong tendency not to trust forecasters, despite relative accuracy. This article implies a lack of accuracy that wasn't actually the case.
It would be incredibly easy to come away from this article with the wrong lessons.
Where do you see the article implying the forecast was inaccurate? To me the author is only saying that how people interpret it is inaccurate which is the entire point of the article.
Yes, exactly. The cone is as accurate as it can be, by which I mean: accurate but with a huge degree of uncertainty. The challenge is that some readers tend to read it as deterministic, and get it wrong by:
1. Thinking "If I'm inside of the cone I'm in danger, if I'm outside I'm good" (wrong! 1/3 of the time the storm will be outside of the cone!)
2. If I'm inside of the cone AND under its center line/dots of I'll worry a lot, but if I'm inside and in between the center of the cone and its boundary, I can relax (I'd be VERY careful with this, as it isn't true either).
And your article drives this point home really well. I honestly don't understand the criticism - the hurricane example you picked from 2017 is just to make a point - which people in this thread are unnecessarily complaining about.
> People have a strong tendency not to trust forecasters
It's possible a lot of this bias is just outdated, up until very recently most hurricane (cyclone in my case) warnings came with a disclaimer like "cyclones or notoriously unpredictable and can make sudden, unexpected course changes" and for the most part this was true even in the short term. I don't think I've seen a wildly incorrect forecast in my area in 20 years but my mother still believes in this inherent unpredictability.
The challenge is that, eventually, the forecast of one storm will be wildly off. This is inevitable. But that doesn't mean that the forecast was necessarily incorrect. I think that this is what it's crucial to explain to the public: the general meaning of uncertainty. We need to learn to deal with it.
Why use a real-world example if it's not going to be based on data? Surely if the numbers are correct that there's 60-70% accuracy with the cone (on any particular day? or possibly early on) then they could use real data.
It also starts with "The National Hurricane Center publishes graphics like this one" (with "graphics" directly linking to a different graph than the one shown [1]) which would lead one to believe it was an actual NHC example being challenged.
The general point can still get across but its just important to choose your anecdotes carefully if you want to be taken seriously.
Especially in the hyper-critical social media age when everyone is looking to kick anything popular down a peg or two, for their own personal satisfaction of being smart (or w/e the motivation is).
This is extreme. Why use a real-world example? Because people can relate to it. People have seen these types of plots. It could have been a completely fabricated example of an arbitrary hurricane in Florida or Canada or whatever - the point would still be the same. I feel like people are completely missing the point of the article.
Also, my question is : what are the wrong lessons that this article teaches?
A whole lot of people who are the target market for this article would likely have experienced the same hurricane. Not giving the peanut gallery easy reasons for dismissal is something to be considered.
> Surely if the numbers are correct that there's 60-70% accuracy with the cone (on any particular day? or possibly early on) then they could use real data.
I've noticed this as well. In reality, outside the cone seems to happen pretty regularly. So far outside the cone that it is significant as a practical matter is much more rare, at least in terms of the 48 hour cone.
> Especially in the hyper-critical social media age when everyone is looking to kick anything popular down a peg or two, for their own personal satisfaction of being smart
The article is pointing out that the increasing circle sizes represent variation of centerline paths, not the actual size of the hurricane (which is a common misconception). Just because the plots don’t match doesn’t make the article irrelevant.