Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
Extract from WSJ article: "Chicago officials this week said they had succeeded in slowing the spread of the virus, with new Covid-19 cases now doubling every 12 days—down from every two to three days a month ago."
I've seen this same metric used by Cuomo and others and the logic escapes me. Suppose 100 new cases daily are being detected through more aggressive testing (not a reliable metric, since it depends on testing, not the spread of the virus). However, suppose the rate stays steady at 100. If 400 cases have been detected, it would take 4-days to double at 100 per day. But, when the total number of cases detected reaches 1200, it will take 12-days for the number of detected cases to double.... at the same 100 cases per day! If testing is doubled, then the rate of new cases will also likely double.
-- All I'm suggesting is that "number of days to double" creates a false sense of security that the rate of spread is slowing, when it really isn't.
Another metric being used is "death rate" as a percentage of cases detected. Again, IF everyone was tested ... or at least more than 3-percent of the population (3 million tests vs 300+ million population), the 'rate of death' figure would have some reliability. (Likewise, if the same number of tests were administered in different locations, the comparative numbers would be more credible. As it stands, it has little to do with the actual morbidity rate, and everything to do with the number of tests administered. In this case, the rate is probably much lower than what is being touted.
I understand that we are in uncharted waters and people/officials are doing the best they can. But, if these questionable statistics are being used at a national level to make serious decisions, a greater effort needs to be made to make them statistically accurate, rather than politically correct.
I agree that the numbers are fuzzy because we have not done enough testing and so many people don't have symptoms. You can see the number of people dying per day has dropped from 800 to 500 in NYC. At this point that's the best way to show the rate of spread has slowed. https://www.nytimes.com/2020/04/20/n...rk-update.html
That's just one example of funny math, but the problem is most Americans at least don't really understand math. If I post a story about "Joe Bob, 37, dies of Coronavirus" and that's the headline, everyone gets out of it: OMG anyone can die of this thing! When I post a link to the USC Study that in fact about 200,000 to 400,000 people in L.A. County alone have had the Coronavirus, and only 600 have died, that's a 0.02% fatality rate. Extremely low. The Santa Clara study said the same thing. The Iceland one wasn't far off that. How about Singapore? There the testing is fantastic, I was just over in Singapore last month in fact, and they have 8,014 confirmed cases with just 11 deaths. Is it because they are using super advanced top secret technology? No, it's because the Coronavirus just isn't very deadly. I don't blame anyone for being wrong on that, we just didn't know, it is after all called "the novel coronavirus" because it's new. It was better safe than sorry, and when we thought the mortality rate was 3%, we acted in the manner you'd expect for a fairly deadly virus.
I don't know if it's a reading comprehension thing or just lack of education, but people don't really seem able to read a study like that and understand what it's actually telling them. They will immediately jump to, "But... I saw pictures... and... what about New York? And Italy? That can't be right." I guarantee if the media ran as many stories about the average seasonal flu as they did Coronavirus, we'd all be terrified of catching that, too.
That aside, the virus is more than twice as contagious, or at least twice as contagious rather, so since nobody has ANY immunity to it whatsoever, there was no "seasonal flu shot," and that means WAY more people are bound to get a virus like this without any social distancing measures or any other measures. That higher rate of contagiousness adds up fast spread across a population and has exponential effects. I think people don't really understand that 200,000 people could die of coronavirus and only 40,000 for the flu but that doesn't mean "the virus is 5 times more deadly!!" Deadliness is only determined by mortality rate, by the actual number out of 1,000 people who get it who die of it. Not by the raw number of deaths, which doesn't mean anything without knowing how many people actually got it. If 7 billion people got a really weak flu that only killed 0.05% of people, it would still result in more deaths than any flu.
The Govt is scrambling right now to find a solution, at the same time, maintaining control and order, they wont be able to do this forever though.
I have a feeling there is ALOT going on right now, within the Govt, that they are not telling the public about. I expect one of these days, Im going to wake up to find banks and ATMs closed, and not see a cop anywhere.
Extract from WSJ article: "Chicago officials this week said they had succeeded in slowing the spread of the virus, with new Covid-19 cases now doubling every 12 days—down from every two to three days a month ago."
I've seen this same metric used by Cuomo and others and the logic escapes me. Suppose 100 new cases daily are being detected through more aggressive testing (not a reliable metric, since it depends on testing, not the spread of the virus). However, suppose the rate stays steady at 100. If 400 cases have been detected, it would take 4-days to double at 100 per day. But, when the total number of cases detected reaches 1200, it will take 12-days for the number of detected cases to double.... at the same 100 cases per day! If testing is doubled, then the rate of new cases will also likely double.
-- All I'm suggesting is that "number of days to double" creates a false sense of security that the rate of spread is slowing, when it really isn't.
You couldn't be more wrong.
Your hang-up -- what 'escapes' you, as you put it -- is clearly failure to understand the concept of rate.
You are comparing linear growth (where the increase is unrelated to the base quantity) with exponential growth (where the increase is a function of the base quantity). Simply put, a consistent increase of 100 (your example) is very clearly, and by definition, a decreasing rate of increase. Exponential increase is not.
Tell you what:
I'll give you one cent. You'll give me one cent.
Tomorrow I'll give you one more cent. Tomorrow you'll give me two more cents.
The day after tomorrow I'll give you one more cent. The day after tomorrow you'll give me four more cents.
We'll do this for a month. Every day I'll give you one more cent. And every day, you'll give me twice as many cents as you gave me the day before. Your increase in cents will thus be linear, while mine will be exponential.
At the end of thirty days, you'll have 30 cents. I'll have over five hundred million cents (to be exact, $5,368,709.12).
There's nothing fuzzy about the math behind exponential growth compared to linear growth. This is stuff we all learned in high school. And you don't even need to be able to remember the math itself to have a grasp of the concept.
It's the simple fact that the true numbers of infected and dead are under reported. Not just here in the USA, but worldwide. With such inadequate testing, relying on "confirmed" numbers just leads to a false sense that this is not as serious as the harsh reality, which may or may not be publicly reported at some point in the future.
2x3x29x41 - excellent explanation. I have tried explaining this many times in person and here - but I can see my own wife's eyes glaze over as soon as pen hits the paper.
jghorton - using just the info you posted (I didn't go farther) the rate IS slowing...because previous to now, it would never have been 100 per day over and over again. As 2x3 suggested...it would 100 today, 115 tomorrow, 127 the day after that...for instance. The rate was growing...and accelerating. In the case where the day - to - day total is roughly the same - that's a heck of a slow down. It means it's unrelated to the amount that's gone on before, and that's a great sign. Ok, it's a decent sign, better than a "bad" sign, it remains to be seen if it's a "great" sign.
Your hang-up -- what 'escapes' you, as you put it -- is clearly failure to understand the concept of rate.
You are comparing linear growth (where the increase is unrelated to the base quantity) with exponential growth (where the increase is a function of the base quantity). Simply put, a consistent increase of 100 (your example) is very clearly, and by definition, a decreasing rate of increase. Exponential increase is not.
Tell you what:
I'll give you one cent. You'll give me one cent.
Tomorrow I'll give you one more cent. Tomorrow you'll give me two more cents.
The day after tomorrow I'll give you one more cent. The day after tomorrow you'll give me four more cents.
We'll do this for a month. Every day I'll give you one more cent. And every day, you'll give me twice as many cents as you gave me the day before. Your increase in cents will thus be linear, while mine will be exponential.
At the end of thirty days, you'll have 30 cents. I'll have over five hundred million cents (to be exact, $5,368,709.12).
There's nothing fuzzy about the math behind exponential growth compared to linear growth. This is stuff we all learned in high school. And you don't even need to be able to remember the math itself to have a grasp of the concept.
Ve your explanation, which makes sense. But no, I did not get this in high school, or in college. I’ve to think hard to try to understand exponential growth. Thanks for the clear explanation of the difference between linear growth and exponential growth.
...The rate was growing...and accelerating. In the case where the day - to - day total is roughly the same - that's a heck of a slow down. It means it's unrelated to the amount that's gone on before, and that's a great sign. Ok, it's a decent sign, better than a "bad" sign, it remains to be seen if it's a "great" sign.
That is our present condition. In the US the daily new-case count has been hovering at around 30,000 for about three weeks now. It is, if anything, sightly decreasing. The daily new-death count has been hover at around 2000, for about two weeks now... if we disregard the "reporting spike" in NYC (they went back and redefined possible earlier deaths as coronavirus deaths). The total case-count and casualty-count keeps rising. The casualty count will keep rising indefinitely, unless people start resurrecting. Plotted with a logarithmic ordinate axis, the curve of cumulative deaths vs. time has been concave-down, for about a month now.
The only number I rely on is the death count. Other than the one day peak of about 3000, the number is lower but steady between 1500 and 2500. To me this means the infection rate post-shutdown has been steady for the last month, probably reflecting the number of people not practicing adequate contagion control. If you guess at a fatality rate of 1%, that's about 200,000 new cases a day. The real number is probably closer to 500,000 new cases a day, since many cases are never identified.
At that rate, it will take about 2 years to hit 80% herd immunity, and the final death count will be around 600,000 to 900,000. That's assuming 1000 to 1500 deaths a day. End the shutdown and ignore contagion control and we will accomplish that in six months, the medical system will be overwhelmed, and many people will die that didn't have to.
Count corpses. The other numbers are suspect, and don't matter anyway.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
