South African cases in the JHU data show a big spike on Nov 23. This is likely to be a simple artifact or an error, rather than the number of cases detected that day. For now, we've removed that data point from the 7-day average on our charts. (Issue:…)
In case that's not obvious, this doesn't mean the conversation & concerns around B.1.1.529 aren't relevant. It simply means that the cases specifically reported by JHU on November 23 shouldn't be linked to that conversation – because we don't know yet if those cases are "real".
There we go: it's a backlog of antigen tests added to the cumulative total. Importantly, "The estimated number of new cases for 23 Nov 2021 is 868"
(A little unfortunate to add that backlog right on the day when everyone is talking about the variant 🤷🏻‍♂️)
Here's what the 7-day rolling average of new cases looks like, once corrected for Nov 23 (to the 868 actual cases confirmed that day).
I'm guessing that people will see whatever they want to see on the previous chart, so for the record, here's what I see: cases are still pretty low, but there's definitely a strong inflection point on November 18.
See here in context, with the time series since November 1st.

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Edouard Mathieu

Edouard Mathieu Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @redouad

25 Nov
Media reporting on COVID vaccination should focus much less on fully-vaccinated people, especially when comparing countries.
It's a metric that has many more flaws than the rest of the vaccination figures, and we now know that it doesn't have as much epidemiological relevance.
Flaw 1: the original definition of the protocols was sometimes very arbitrary. J&J doses give you a full vaccination, but only because that's what the company tested. But since J&J seems to give lower protection, many countries recommend people get a 2nd dose of another vaccine.
Flaw 2: definitions vary from country to country. A few countries in Central Europe consider that people with a previous infection + only 1 dose are fully vaccinated. It isn't necessarily a bad thing, but it makes it very hard to compare full vaccinations accurately.
Read 6 tweets
23 Nov
NEW: To understand how the pandemic is evolving, it's crucial to know how death rates from COVID differ by vaccination status.
In this post we explain why rates are the key metric to use for this, and we show the latest data for the US, England, and Chile:… Image
You may sometimes see headlines like “Half of those who died from the virus were vaccinated”.
But it would be very wrong to draw conclusions about the vaccines based on this headline, because we also need to know how many people in this population were vaccinated & unvaccinated. ImageImage
In the post, we walk you through an example to illustrate how to think about these statistics in a hypothetical case.
The same logic also applies in the pandemic. Comparisons of absolute numbers, as some headlines do, ignore the fact that one group is much larger than the other. Image
Read 8 tweets
13 Sep
We've added charts on country-by-country donations to COVAX:…

Donations are broken down by whether they have been only announced, actually donated, or delivered to the recipient countries.

Chart 1 shows absolute numbers; chart 2 shows donations by GDP. ImageImage
In absolute numbers, the largest pledges have come from:
🇺🇸 US: 290 million doses announced
🇬🇧 UK: 80 million
🇫🇷 France: 54 million

But so far only a fraction of these doses have been delivered:
🇺🇸 US: 77 million doses donated & delivered
🇬🇧 UK: 5 million
🇫🇷 France: 4 million Image
Once we adjust for the size of each country's economy, the largest pledges were made by:
🇬🇧 UK
🇨🇦 Canada
🇫🇷 France
🇺🇸 US
🇪🇸 Spain

And the largest vaccine deliveries have come from:
🇺🇸 US
🇸🇪 Sweden
🇬🇧 UK
🇫🇷 France
🇳🇴 Norway Image
Read 4 tweets
13 Aug
There is a rising debate in Spanish Twitter following this tweet by Pedro Sanchez.

The question being: is it true that Spain is first in the world in vaccinations, or is it just the first country in the default selection on this chart?
Due to limited space on our charts and the number of countries in our data, we can't show all countries. We make a default selection based on a mix of criteria, the main one being population (if you can't show everyone, the second-best thing is to show as many people as possible)
Now, are there countries that have vaccinated a higher % than Spain?

For a population > 20 million, there aren't.

For a population > 10 million: Chile, Portugal

For a population > 1 million: UAE, Singapore, Denmark, Uruguay, Qatar

For a population > 100,000: Malta, Iceland
Read 4 tweets
23 Jan
Regarding vaccination speed and comparisons between countries (especially in the EU), some people say that "it doesn't matter how fast you start, what matters is crossing the finish line at the same time".
But that's not at all how an epidemic works. Quick explanation below.
This claim can be summarized by the following chart: both the yellow and the black country arrive at 100% coverage by the same date.
But the country in yellow started very quickly and slowed down, while the country in black started very slowly and then strongly accelerated.
The idea that those two trajectories are equivalent works for many things in life.
For example, if two marathon runners follow those trajectories during a race, they will indeed cross the finish line at the same time, and neither of the two strategies can be said to be "better".
Read 8 tweets
20 Jan
After several years contributing in various ways to the wonderful resource that is @five_books, I officially sent my last newsletter today!

Here's a list of some of my favorite (and most related to my work at @OurWorldInData) interviews that I published between 2018 and 2020.
[Computer science] How do computers work? What is well-crafted code? How do you write an algorithm? @anabellphd, lecturer in the Electrical Engineering and Computer Science Department at MIT, chooses the best books to learn computer science & programming.…
[Data science] This new discipline has risen to extreme popularity in the last decade. @rdpeng, Professor of Biostatistics at Johns Hopkins University and founder of one of the largest data science online courses, recommends the best books to dive into it.…
Read 8 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Thank you for your support!

Follow Us on Twitter!