Adam Kucharski
@adamjkucharski.bsky.social
Epidemiologist/mathematician. Professor at London School of Hygiene & Tropical Medicine. Author of The Rules of Contagion and The Perfect Bet. Views own.
New book Proof: The Uncertain Science of Certainty available now: proof.kucharski.io
New book Proof: The Uncertain Science of Certainty available now: proof.kucharski.io
How to be taken seriously as a junior data scientist: www.divingintodata.com/p/how-to-be-...
November 11, 2025 at 8:54 PM
How to be taken seriously as a junior data scientist: www.divingintodata.com/p/how-to-be-...
“Instead of writing a lengthy doc and holding endless meetings to debate what to build, Hannah whipped up a prototype so people could react to something tangible.” Useful piece on agency: www.operatorshandbook.com/p/how-to-be-...
How to be "high agency" at work
Practical examples of how to "just do things" when others are stuck overanalyzing and procrastinating — incl. how to leverage AI
www.operatorshandbook.com
November 11, 2025 at 4:57 PM
“Instead of writing a lengthy doc and holding endless meetings to debate what to build, Hannah whipped up a prototype so people could react to something tangible.” Useful piece on agency: www.operatorshandbook.com/p/how-to-be-...
Reposted by Adam Kucharski
"Schrödinger's causal inference" (n):
The practice of making causal claims or interpretations within a scientific article - typically in the title, abstract, implications, or conclusion - while simultaneously warning that the study design is unsuitable for causal inference.
The practice of making causal claims or interpretations within a scientific article - typically in the title, abstract, implications, or conclusion - while simultaneously warning that the study design is unsuitable for causal inference.
November 11, 2025 at 11:36 AM
"Schrödinger's causal inference" (n):
The practice of making causal claims or interpretations within a scientific article - typically in the title, abstract, implications, or conclusion - while simultaneously warning that the study design is unsuitable for causal inference.
The practice of making causal claims or interpretations within a scientific article - typically in the title, abstract, implications, or conclusion - while simultaneously warning that the study design is unsuitable for causal inference.
Reposted by Adam Kucharski
Over 24 weeks, 55% of the participants had a respiratory virus detected at least once, although they don't always need to symptoms. The youngest (ages 3-8) were most likely to be positive for a virus AND most likely to have symptoms; if staff were virus+, it was likely to be symptomatic.
November 11, 2025 at 11:31 AM
Over 24 weeks, 55% of the participants had a respiratory virus detected at least once, although they don't always need to symptoms. The youngest (ages 3-8) were most likely to be positive for a virus AND most likely to have symptoms; if staff were virus+, it was likely to be symptomatic.
Literal shot: Questionable methodological chaser:
November 10, 2025 at 8:51 PM
Literal shot: Questionable methodological chaser:
Proof has hit 100 Amazon reviews! Thanks to everyone who's liked the book and been kind enough to leave one. (Also nice to see that it remains the highest rated of my 3 books.)
November 10, 2025 at 8:32 PM
Proof has hit 100 Amazon reviews! Thanks to everyone who's liked the book and been kind enough to leave one. (Also nice to see that it remains the highest rated of my 3 books.)
Reposted by Adam Kucharski
Modeling often requires making assumptions and design choices explicit. Experimenters could benefit from such transparent thinking abt the design and knowledge of the range of experiments they are foregoing. They could also benefit from understanding how data produced is conditional on the design.
November 10, 2025 at 4:05 PM
Modeling often requires making assumptions and design choices explicit. Experimenters could benefit from such transparent thinking abt the design and knowledge of the range of experiments they are foregoing. They could also benefit from understanding how data produced is conditional on the design.
This is fascinating, can't believe I hadn't seen it before – 'it is not a fallacy to believe in the hot hand'
November 10, 2025 at 12:58 PM
This is fascinating, can't believe I hadn't seen it before – 'it is not a fallacy to believe in the hot hand'
Reposted by Adam Kucharski
So many nonsense ad hoc pipelines could be prevented by requiring that they work on synthetic data.
I tend to think of experiments as special cases of inference, since most of the problems I work on cannot be studied in experiments. But I get that many researchers see experiments as base analogy.
I tend to think of experiments as special cases of inference, since most of the problems I work on cannot be studied in experiments. But I get that many researchers see experiments as base analogy.
"Validate With Simulated Truth: A first habit is to test whether an analytical pipeline can recover known conditions."
Very good advice below. So much COVID nonsense (e.g. 'immunological dark matter') basically came down to a non-identifiable model that hadn't been properly tested.
Very good advice below. So much COVID nonsense (e.g. 'immunological dark matter') basically came down to a non-identifiable model that hadn't been properly tested.
Modelling Like an Experimentalist
Dahlin et al. (2024) apply experimental thinking to a model of mosquito-borne disease transmissions.
onlinelibrary.wiley.com
November 10, 2025 at 12:41 PM
So many nonsense ad hoc pipelines could be prevented by requiring that they work on synthetic data.
I tend to think of experiments as special cases of inference, since most of the problems I work on cannot be studied in experiments. But I get that many researchers see experiments as base analogy.
I tend to think of experiments as special cases of inference, since most of the problems I work on cannot be studied in experiments. But I get that many researchers see experiments as base analogy.
Some thought-provoking reflections on USAID: www.theguardian.com/us-news/2025...
November 10, 2025 at 12:25 PM
Some thought-provoking reflections on USAID: www.theguardian.com/us-news/2025...
Reposted by Adam Kucharski
Let's play Bucks Name or Name Name
November 7, 2025 at 1:01 PM
Let's play Bucks Name or Name Name
But at least we got a new passport colour (which we could have had anyway).
November 10, 2025 at 9:50 AM
But at least we got a new passport colour (which we could have had anyway).
Reposted by Adam Kucharski
Great post, do read
What happened when I dug into some of the key studies and ideas quoted in The Tipping Point and its sequel… kucharski.substack.com/p/the-real-r...
November 9, 2025 at 1:54 PM
Great post, do read
"Validate With Simulated Truth: A first habit is to test whether an analytical pipeline can recover known conditions."
Very good advice below. So much COVID nonsense (e.g. 'immunological dark matter') basically came down to a non-identifiable model that hadn't been properly tested.
Very good advice below. So much COVID nonsense (e.g. 'immunological dark matter') basically came down to a non-identifiable model that hadn't been properly tested.
Modelling Like an Experimentalist
Dahlin et al. (2024) apply experimental thinking to a model of mosquito-borne disease transmissions.
onlinelibrary.wiley.com
November 9, 2025 at 9:18 PM
"Validate With Simulated Truth: A first habit is to test whether an analytical pipeline can recover known conditions."
Very good advice below. So much COVID nonsense (e.g. 'immunological dark matter') basically came down to a non-identifiable model that hadn't been properly tested.
Very good advice below. So much COVID nonsense (e.g. 'immunological dark matter') basically came down to a non-identifiable model that hadn't been properly tested.
To back up their (incorrect) argument that COVID vaccines didn't affect transmission of the early variants, someone just sent me a paper showing that vaccines 'only' prevented infection...
The thing is, it's pretty hard to transmit COVID if you don't get infected in the first place.
The thing is, it's pretty hard to transmit COVID if you don't get infected in the first place.
November 9, 2025 at 7:53 PM
To back up their (incorrect) argument that COVID vaccines didn't affect transmission of the early variants, someone just sent me a paper showing that vaccines 'only' prevented infection...
The thing is, it's pretty hard to transmit COVID if you don't get infected in the first place.
The thing is, it's pretty hard to transmit COVID if you don't get infected in the first place.
Reposted by Adam Kucharski
Scraped data from the plot ... gist.github.com/bbolker/d6a6... Scraping is imperfect because the plot is a fairly crappy JPG (hard to distinguish overlapping points). Dashed line is 156 particles per liter, nominal normal/superspreader cutoff (points fall on either side due to scraping error)
November 9, 2025 at 5:46 PM
Scraped data from the plot ... gist.github.com/bbolker/d6a6... Scraping is imperfect because the plot is a fairly crappy JPG (hard to distinguish overlapping points). Dashed line is 156 particles per liter, nominal normal/superspreader cutoff (points fall on either side due to scraping error)
Reposted by Adam Kucharski
“When it comes to understanding how things spread – and what to do about it – we need good science and careful reasoning, not just great stories.”
Couldn’t agree more. And it’s always worth reading @adamjkucharski.bsky.social because you usually get both…
Couldn’t agree more. And it’s always worth reading @adamjkucharski.bsky.social because you usually get both…
What happened when I dug into some of the key studies and ideas quoted in The Tipping Point and its sequel… kucharski.substack.com/p/the-real-r...
November 9, 2025 at 2:05 PM
“When it comes to understanding how things spread – and what to do about it – we need good science and careful reasoning, not just great stories.”
Couldn’t agree more. And it’s always worth reading @adamjkucharski.bsky.social because you usually get both…
Couldn’t agree more. And it’s always worth reading @adamjkucharski.bsky.social because you usually get both…
What happened when I dug into some of the key studies and ideas quoted in The Tipping Point and its sequel… kucharski.substack.com/p/the-real-r...
November 9, 2025 at 1:44 PM
What happened when I dug into some of the key studies and ideas quoted in The Tipping Point and its sequel… kucharski.substack.com/p/the-real-r...
Latest post, on digital twins and digital spin: kucharski.substack.com/p/can-langua...
Can language models tell us what populations are thinking and doing?
Digital twins and digital spin
kucharski.substack.com
November 8, 2025 at 2:51 PM
Latest post, on digital twins and digital spin: kucharski.substack.com/p/can-langua...
How common do you think something is? When asked to judge proportions, people's estimates can be drawn towards their prior expectations of what the 'typical' value is. The result can be an overestimate of low proportions, and an underestimate of high ones... link.springer.com/article/10.3...
November 5, 2025 at 11:28 AM
How common do you think something is? When asked to judge proportions, people's estimates can be drawn towards their prior expectations of what the 'typical' value is. The result can be an overestimate of low proportions, and an underestimate of high ones... link.springer.com/article/10.3...
"ARRRRGGGHHHH WHAT'S THAT BEEPING NOISE AND WHY'S IT DARK??"
or
"Oh, I'm awake now"
Reminder, as mornings get darker, that a sunrise alarm clock is one of the best things I've ever picked up for dealing with the winter run in.
or
"Oh, I'm awake now"
Reminder, as mornings get darker, that a sunrise alarm clock is one of the best things I've ever picked up for dealing with the winter run in.
November 5, 2025 at 7:45 AM
"ARRRRGGGHHHH WHAT'S THAT BEEPING NOISE AND WHY'S IT DARK??"
or
"Oh, I'm awake now"
Reminder, as mornings get darker, that a sunrise alarm clock is one of the best things I've ever picked up for dealing with the winter run in.
or
"Oh, I'm awake now"
Reminder, as mornings get darker, that a sunrise alarm clock is one of the best things I've ever picked up for dealing with the winter run in.
Writing a book is a bit like doing a jigsaw. Except, unlike an actual jigsaw, you have to carve out all the pieces from scratch then work out how they fit together.
1000 hours, give or take
Some lessons from the ups and downs of writing three books
kucharski.substack.com
November 4, 2025 at 5:46 PM
Writing a book is a bit like doing a jigsaw. Except, unlike an actual jigsaw, you have to carve out all the pieces from scratch then work out how they fit together.
An important point to remember in the face of (often financially-motivated) modern falsehoods: it's not a new issue, and remarkable improvements in health have historically been achieved despite this challenge.
This a million times. Denial never really changes, just shifts to update its language with new discoveries and new terminology. We see the same with evolution and HIV denial. Plus, people were sick of antivaxxers even in the late 1800s.
November 4, 2025 at 11:56 AM
An important point to remember in the face of (often financially-motivated) modern falsehoods: it's not a new issue, and remarkable improvements in health have historically been achieved despite this challenge.
One thing I do miss about old Twitter (apart from all the useful info and interactions – some of which have moved, some not) is that it had a very good search function. Realised I basically used it as a public store of interesting things I'd seen. Now very hard to retrace elsewhere...
November 4, 2025 at 8:52 AM
One thing I do miss about old Twitter (apart from all the useful info and interactions – some of which have moved, some not) is that it had a very good search function. Realised I basically used it as a public store of interesting things I'd seen. Now very hard to retrace elsewhere...
Got curious about what was happening* on Threads these days.
*or not, it seems
*or not, it seems
November 3, 2025 at 9:05 PM
Got curious about what was happening* on Threads these days.
*or not, it seems
*or not, it seems