Will Stancil Profile picture
Jan 31 4 tweets 1 min read
I'm literally being told by a notable account with 80k followers that it is impossible to calculate the share of breakthrough COVID deaths, given the following two numbers for a period:
-the number of COVID deaths
-the number COVID deaths of vaccinated people
Look, this is not fancy math. You divide the first number into the second one. It's trivial.

The fact that people don't feel comfortable making this calculation on their own is really weird! Doing basic division to reach an obvious conclusion should not be beyond the pale!
I just hope at some point in the near future some semblance of sanity returns to this discourse, and all the people who refuse to see the obvious thing sitting right in front of them feel as stupid as they look right now.
That's right, figuring out percentages of things - no man alive can do this calculation. They whisper about it in biostatistics departments. The forbidden art.

• • •

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

Keep Current with Will Stancil

Will Stancil 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!

PDF

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 @whstancil

Feb 1
Let's talk about breakthrough deaths. It's honestly been a tough couple of days, with everyone yelling at me that I was lying, or worse, turning anti-vax, because I've said 20%-25% of recent COVID deaths are vaccinated. I worried I was way out of line so I checked more numbers.
I'm going to go through what I found in detail. Like, far too much detail. Like, it's going to be a long thread. But first, a few notes.
Okay, so this keeps tripping people up: THE EXISTENCE OF BREAKTHROUGH CASES, EVEN IN LARGE NUMBERS, DOES NOT SHOW THAT VACCINES ARE INEFFECTIVE.

The vaccinated population is much larger and older than unvaccinated population, so it can produce many deaths even if safer overall!
Read 35 tweets
Jan 30
This is a good illustration of what I mean about RATES versus OVERALL TOTALS. This chart shows comparative hospitalization rates for NYC. If you look at it you might think twenty unvaccinated people are hospitalized for every vaccinated person. But that’s not true!
For an individual, this chart is useful: you are way, way safer if you get vaccinated.

But for a policymaker it can be misleading: it suggests that this is almost entirely a “pandemic of the unvaccinated,” so there’s no reason to worry about the rest of the population.
Both kinds of information are useful, in different ways. Comparative rate data shows the massive effectiveness of vaccines, but gives a poor sense of total harm. Overall data shows total harm better, but can make the vaccines look less effective than they are.
Read 4 tweets
Jan 30
People are going to accuse me of being an anti-vaxxer for even raising the question, but it's incredibly frustrating how difficult it is to get raw data on numbers of COVID deaths broken down by vaccination status. Almost all data is published as a share-of-population rate.
One might think you could use the share-of-population rate to calculate the raw death totals, but you can't - it's been heavily adjusted and everyone has different population-level vaccination figures anyway.
I'm pretty sure the reason for the strange absence of this rather important data is because people think if Americans see the raw number of vaccinated people that have died, it'll give ammunition to anti-vaxxers. I'm sympathetic to these concerns.
Read 4 tweets
Jan 29
This is the weirdest account. I followed it many years ago when it just seemed like some dude posting funny old newspaper stories about the dangers of telephones. But it’s turned into weird Big Tech propaganda. Image
Like in response to Neil Young’s pretty successful campaign against Spotify for carrying Rogan, you know what this account has been doing? Going after Neil Young over and over! Almost like you’d expect a crisis PR campaign to do. ImageImageImage
I mean, doesn’t this feel like a not-particularly-subtle attempt to defend Spotify more than an attack on scientific misinformation? Image
Read 4 tweets
Jan 27
"The economy is growing explosively but the president's approval is imploding" drives a freight train through every theory of politics that the Democrats and the media political class has used for 40 years, and they're responding by saying "Of COURSE he's unpopular, because X"
Our theory about what drives presidential approval is completely broken, we need a new one pronto, but most of the putative experts and practitioners are instead desperately adding epicycles: "It's still the economy, stupid, but 'the economy' specifically means gas prices"
Here's my proposal: what people think about Biden does not reflect what they already BELIEVE about his policy proposals, or what they are SEEING in their day-to-day lives, but mostly reflects what they are HEARING about him from each other and from the news
Read 4 tweets
Jan 26
I think politics makes more sense if you assume voters see money stuff as boring and don't really care about it the way nerds do
Some years I do better financially, other years I don't, on some level I understand this reflects public policy but emotionally it just processes like "the vicissitudes of life" and a pretty esoteric thing to vote on, instead of the sweeping moral narratives of partisan conflict
here's a clue that everyone misses: even among people who are strongly supportive of the child tax credit, the support is not because they want the money themselves, but because for them, the CTC is a values issue about protecting children that instills strong emotions!
Read 6 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

Don't want to be a Premium member but still want to support us?

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

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

:(