Exponential growth and epidemics on HackerTube: recommended by 10 Hacker News commenters across 1 month, 2020.
explainer · evergreen
A short 3Blue1Brown primer on exponential and logistic growth, using early COVID-19 case data to show how epidemics grow and eventually level off. It explains why small changes in growth rate make enormous differences in outcomes.
During the early COVID-19 period, HN commenters linked it as the clearest quick explanation of what exponential growth means for an epidemic.
“Here is a neat video that helps put things in perspective for most any viewer.”
20 Hacker News comments link this video. Each one links back to its thread.
Just watched the 3brown1blue video today, which talks about exponential growth and fixes it to the latest actual COVID-19 numbers. Really helped it all click. ▶ this video Edit: separate HN discussion: [link]
in “Show HN: Epidemic Simulator” · View on HN
To be exact, it's a "logistic curve", which initially is exponential. From this 3Blue1Brown excellent video: ▶ this video
in “Why some governments appear not to be acting on the Covid-19 threat” · View on HN
And those who don't understand, or need a refresher on, exponential growth should watch this great video by 3Blue1Brown. Exponential growth and epidemics: ▶ this video
in “Ask HN: What happens next after a successful lockdown?” · View on HN
I live in Bergen County. There is a lack of panic - although I'm on the north side, opposite of the most infected parts. I went grocery shopping earlier today, there were a few people with masks but no one was really trying to keep a distance from others. Plenty of people there, perhaps a few less than usual. At times there are so many people that they have to restrict entry into the store and create a line outside (possibly to help with social distancing, but there are still too many people in the store for that to matter). It's the opposite of social distancing. The disaster in NYC, and the counties immediately north/east of Bergen (Westchester, Rockland) is going to spill over. We're not going to be able to handle it and we don't have the clout of NY to push for more resources. In a recent video, 3Blue1Brown wrote "The only thing to fear is a lack of fear itself." This is what worries me. My parents are at risk as long-time smokers and if Valley runs out of ventilators and PPE, our major/usual option is out. (excellent video by the way) ▶ this video
in “‘It looks like a war zone’: NJ docs say hospitals under siege, supplies low” · View on HN
regardless of what you hear, china takes outbreaks very seriously and responds aggressively. the general population in china also takes it very seriously. normal behavior response in china looks like over reaction in the us. please watch this video by 3blue1brown ▶ this video china is past its inflection point and we have yet to determine where our inflection point will be. everyone should be changing your behavior now. the fact that sars-cov didn’t get as far in 2002 as sars-cov-2 virus is currently going should be concerning. the us could ignore prior outbreaks because china is aggressive to stop it.
in “Show HN: HN reader that filters out COV**-19 and Coronavir** articles” · View on HN
But months ago they really couldn’t have known. We still don’t know how many will be infected. All we can do is take precautionary measures. The WHO messaging should have been stronger and earlier, but it isn’t their fault for not labeling a pandemic months ago. Granted, there was enough data and research done to squarely label this as a pandemic, and that’s really where I assign blame on them. Here is a neat video that helps put things in perspective for most any viewer. ▶ this video
in “Covid-19 is now officially a pandemic, WHO says” · View on HN
Maybe they should include this 3Blue1Brown video to explain what 'exponential' means in this context: ▶ this video
in “Open letter from Italy to the international scientific community” · View on HN
This is the time to resort to drastic measures to prevent drastic consequences. Have you seen this from 3Blue1Brown: ▶ this video "Exponential growth and epidemics" ?
in “Shelter in place for Bay Area counties” · View on HN
3blue1brown did a good video explaining the logistic shape of pandemic curves: ▶ this video My key takeaway was that once you see the inflection point, you know that you’ll have roughly 2x the number infections by the time it’s done.
in “The Doctor Who Helped Defeat Smallpox Explains What's Coming” · View on HN
3 blue one brown had a great video on exponential growth that detailed how a 10% difference in the numbers is the difference between being halfway there and having orders of magnitude to go. As I'm sure you're aware, the point of the numbers isn't to change the reality of the situation, it's to understand what that reality is. And the people that care are epidemiologists and leaders with decades of experience fighting pandemic after pandemic. You know, people that know what they're talking about? Might be worth trying to spend time actually learning how the people that are spending day and night trying to save your life and the lives of your loved ones do their work before dismissing them as useless. ▶ this video
in “Covid-19 Sanity – Covid-19 papers from bioRxiv/medRxiv searchable and sortable” · View on HN
100k is a rough estimate. It might be higher, but will most likely be lower. The data will need to be pruned and carefully looked at. There are incentives for people to misclassify deaths at COVID19[0]. It will likely not get much higher than 100k as exponential growth doesn't continue indefinitely[1]. It's practically gone in Italy right now. 600k die every year from hearth disease. 700k died globally from mosquito transmitted illness[2]. There are tons of secondary effects. People in America have died of Malaria. Many have committed suicide (Including my best friend's flatmate). Many were told not to come to hospitals. Many hearth attacks were wrongly attributed to COVID if a patient tested positive for it. We've seen massive amounts of people at beaches in Florida, Texas and other places, as well as these protests and riots, and there are no massive spikes in fatalities 2~4 weeks after those events. The WHO said asymptomatic spread was rare, then waked it back (probably because of political pressure from Dr Fauchi), even though it was known asymptomatic spread was probably not a major transmission in February[3]! This entire thing has been one massive media manipulation, and in less than two days, the entire thing flipped from one global narrative to another, like a switch! [0]: [link] [1]: ▶ this video [2]: [link] [3]: [link]
in “Capitol Hill Autonomous Zone” · View on HN
"there are plenty of other data" -- would you be able to educate with some sources? I've been digging in deep and trying to keep an impartial opinion, but have not seen solid evidence that this is true -- or, as you say "there is a study for every truth", I've seen no study to indicate "it is hardly dangerous" (which also feels subjective in those words) > Also, so many deaths are being counted as covid when the patient already had a severe condition Yes... as many scientists talk about, COVID-19 has a much more severe effect in people with underlying conditions. Here is where Dr. Fauci, or the CDC as a whole talks specifically about underlying issues: [link] "the numbers are way inflated" is using generalized language for something that might be true, but even if you take a _very_ conservative view and consider 50% of the reports are false (and I've seen nowhere claiming it's anywhere near that high), the 50% of remaining deaths and complications are significantly high enough that it's far worse than the flu, and far worse if lockdown wasn't in play where some basic math indicates it would be 10x worse: ▶ this video
in “Update to L.A.'s stay-at-home orders” · View on HN
People keep extrapolating that tens of millions will die, and I don't doubt that tens of millions surely would _if_ there were no lock-downs! (The subject of this article!) The whole point of locking down Wuhan/Hubei and then northern Italy and now the whole country and other places and cancelling large public gatherings and events is that local, regional, and national authorities start seeing the numbers soaring and initiate extreme mitigation strategies. We don't know how successful these strategies will be individually/collectively in the long run but if China is anything to go by it can severely retard the spread of the virus. To show I'm not spouting out of my derriere compare the logarithmic infection and death rates of China[1] and Italy[2] So you have to modify your formula: `7.8 billion * 0.6 * 0.03 * mitigation_factor = ?` We just don't know what the mitigation_factor is yet and because the spread is exponential any dent in the rate of spread brings the fatality number tumbling down[3]. I happen to believe this virus is a very serious global threat and I assure you that I am not one given to alarmism. But I also have never seen governments respond to the onset of an epidemic like this before and I'm nearly fifty and my mother says she can't remember anything like this in her lifetime. So yes, tens of millions absolutely could die but I'm willing to bet that between lock-downs and modern medicine tens of millions won't and I pray to, you know, $deity I am not wrong.
in “Italy is extending its coronavirus quarantine measures to the entire country” · View on HN
> That's my point: to observably flatten the curve, you need to go almost all the way there of killing it altogether. no, you don't. killing it completely would (hand-waving) require taking R0 < 1, and holding it there indefinitely. any reduction in R0 at all will flatten the curve, at least a bit. and small changes in the rate of growth can have a huge impact when you're dealing with (even temporarily) exponential growth. ▶ this video
in “Covid-19 update and guidance to limit spread” · View on HN
This is a simple tool for visualizing how massive the impact if minor behavioral changes can be in the spread of infectious diseases. This model is mostly based off of the math discussed in 3Blue1Brown's video on exponents and epidemics. ▶ this video There are tons of ways this model could be improved, so please feel free to contribute. [link]
in “Show HN: Tool for Visualizing Spread of Infectious Disease” · View on HN
Even at a 0.7% mortality rate, you’re talking about a small percentage of a very big number. Up to 70% of the world’s population may become infected. [0] In that case, a 0.7% death rate converts into 37 MILLION deaths. The runs on toilet paper are the irrational panic of humans. I saw some comments on the internet early on in the outbreak telling people to buy anything they can’t live without for two weeks, with particular emphasis on toilet rolls, so that’s probably where that came from. With regards to the extreme measures, my understanding is that this is primarily to try to prevent health care systems collapsing. Around 20% [1] of people with COVID-19 are sick enough to need hospitalisation, and 1/4 of those (5% of total cases) require intensive care. The number of hospital beds is finite, and the number of ICU beds even smaller. Most health care systems already run near maximum capacity, and right now we’re barely past the peak of flu season [2], so there is very little buffer. The rate for new infections is dictated by the number of people exposed to an infected person and the probability of each exposure becoming an infection [3]. As such, the primary lever we have to try to “flatten the curve” is to reduce the number of people exposed to an infected person, and the way to do that from the top-down is to ban as many congregations of people as possible. Again, since this is exponential growth we’re talking about, even a tiny downward shift in the number of exposures res
in “Ask HN: Covid-19 panic – what am I missing?” · View on HN
At the current exponential rate of worldwide growth, by the end of June, we will be in one of these two scenarios: 1) Almost everyone in the world who is susceptible has it, or had it. 2) We implemented a way to slow down the global growth rate, or it slowed down by itself. One way or another, we will know in a just a few months. If you want to check the math, skip to about 3:20 on this video from 3Blue1Brown, Exponential growth and epidemics. ▶ this video
in “Trump Declares National Emergency” · View on HN
All epidemiologists agree with this. All of them. They've been making public statements to that effect and signing letters reiterating the urgency of the situation. So what are you saying exactly? I don't understand what your point is! I'm agreeing with the epidemiologists. Their maths checks out. It's the same maths as my maths. It's the same maths as everyone's maths : ▶ this video The only people I hear disagreements from are people who don't know how to use a calculator, Or they do, but haven't bothered. Nonetheless, they seem to have lots of time argue online, dismissing logical, coherent, and factual arguments without making any defensible statements of their own.
in “Open letter from Italy to the international scientific community” · View on HN
> "Or, multiply it by less if you want to take into account various amounts of social distancing we're all doing. I'll cut that number above down by a huge amount – 90% – on the assumption that we all learn to stay home and self-isolate immediately, just to be super aggressive on my assumptions about how humanity will rise to the occasion" Dropping 90% is not a great way to calculate this. Look at the math for 61 days out (from March 8): ▶ this video I don't think our social distancing and hand washing will be near enough to prevent millions of deaths, but it's important to recognize that a small change in the growth rate does make a huge difference. Estimating how much we can affect the growth rate is hard, but I think models should be based on a iffy guess on that value rather than an iffy guess on the resulting count. (Expert models certainly are.)
in “Please poke holes in my Covid-19 math (SF Bay Area)” · View on HN
you're right if one assumes a linear growth, but the reality is that it has an exponential growth. See ▶ this video for a short intro.
in “Current estimates about the Covid-19 fatality rate may be too high” · View on HN