bluepenquin
Hall of Fame
4-Time VolleyTalk Poster of the Year (2019, 2018, 2017, 2016), All-VolleyTalk 1st Team (2021, 2020, 2019, 2018, 2017, 2016)
Posts: 12,447
|
Post by bluepenquin on Aug 21, 2020 9:02:58 GMT -5
It looks much similar to the US at large. If we look at the states that make up the South (7 day rolling totals). The peak was on 7/19 with 197,918 cases. Yesterday it was down to 106,460 cases for a 46% decrease. The 7 day rolling cases was roughly equal to 6/26. Pretty much a very even bell curve from 6/26 to 8/20 - with a peak on 7/19. If we look at Florida. The peak was 7/17 with 83,090 cases (7 day rolling). Yesterday, it was 31,465 - down 62% since the peak. Yesterday's rolling average is roughly equal to 6/26 with a near perfect looking bell curve. I have some doubts about that data... covidtracking.com/data/charts/all-metrics-per-stateGraph for Florida. Can look at graphs for each of the states.
|
|
|
Post by dc on Aug 21, 2020 9:06:02 GMT -5
You are manipulating data though. The new wave of cases are coming predominantly from the South, such as from Florida. So... What? It is very true that the new wave of cases is coming form the South although you conveniently didn't mention the west which has had big surges in AZ & CA. But the data is just the data. You can try to spin it but even after this second wave the South will be in much better shape than the Northeast. So having a bunch of Yankees lecture the South is just a bit rich. hilarious
|
|
|
Post by ironhammer on Aug 21, 2020 9:06:17 GMT -5
I don't mean the accuracy of the data itself, I mean whether it shows what is really going on. This is from last month, but it did show clearly a spike in cases from some Southern states, in somes cases those that re-open too early.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Aug 21, 2020 9:22:33 GMT -5
I don't mean the accuracy of the data itself, I mean whether it shows what is really going on... To your point, testing in has decreased over the last 7 days in every SEC state except TX, AL, and TN.
|
|
|
Post by bigjohn043 on Aug 21, 2020 9:24:19 GMT -5
You are manipulating data though. The new wave of cases are coming predominantly from the South, such as from Florida. So... Here are cases per 1M since the beginning: New Jersey - 21,225 New York - 21,990 Connecticut - 19,949 Massachusetts - 17,947 Florida - 27,405 Georgia - 23,239 South Carolina - 21,232 North Carolina - 14,292 Cases paints a very different picture. It also shows just how much more deadly this was early on (still deadly). On a positive side - overall cases in the US have been dropping with the peak being 7/22. 7 day average cases is 31% less than the peak on 7/22 and about the same as the 1st week of July. Not sure where the next outbreak will occur - so far, pretty much any place that got hit hard hasn't gotten hit a 2nd time (yet). It seems to find places that haven't been impacted much before. A possible exception could be the state of Louisiana - but from what I understand, it was the New Orleans area that got crushed early on while it is the rest of the state that got crushed in July/August (while Orleans hasn't been nearly as bad). Cases just aren't a very good metric because we are only diagnosing a small portion of the true infections. That was particularly true early on in NY/NJ. FWIW, Infection Fatality Rates (IFR) do appear to be going down. Part of this is because a greater portion of the infections in is the younger population. It turns out that forcing nursing homes to take covid positive patients wasn't a very good strategy. It does appear we are also getting better at treating the disease....
|
|
|
Post by oldnewbie on Aug 21, 2020 11:07:19 GMT -5
But you also need to factor in the precise time frame here for each state you listed above, when the cases started spiking, when it decrease and which geographical regions had the most and least cases. It looks much similar to the US at large. If we look at the states that make up the South (7 day rolling totals). The peak was on 7/19 with 197,918 cases. Yesterday it was down to 106,460 cases for a 46% decrease. The 7 day rolling cases was roughly equal to 6/26. Pretty much a very even bell curve from 6/26 to 8/20 - with a peak on 7/19. If we look at Florida. The peak was 7/17 with 83,090 cases (7 day rolling). Yesterday, it was 31,465 - down 62% since the peak. Yesterday's rolling average is roughly equal to 6/26 with a near perfect looking bell curve. You cherry picked 6/26 because it was well above the 7 day average. 6/26 was also the single highest day to that point (even though it was over 3 months into the quarantine), so hardly an impressive number to get back to. If we more accurately use the 7 day moving average, it is not as nice of a bell curve, and we are still 10,000 cases a day higher than the 6/26 level (approx 45,000 to 35,000 daily cases). Yes, 8/17 was 31,465, but then 8/18 was 40,458, 8/19 was 45,103 and 8/20 was 43,245. The 8/17 date you cherry-picked looks to be a full 12,000 below the 7 day average. So you cherry picked a starting point 10,000 above the 7 day average and a finish point 12,000 below the 7 day average. covidtracking.com/data/charts/us-daily-positiveIn addition, the daily numbers are very cyclical on a weekly basis. Mondays almost always show a huge dip while there is often a large spike on Friday. 6/26 was a Friday while 8/17 was a Monday.
|
|
bluepenquin
Hall of Fame
4-Time VolleyTalk Poster of the Year (2019, 2018, 2017, 2016), All-VolleyTalk 1st Team (2021, 2020, 2019, 2018, 2017, 2016)
Posts: 12,447
|
Post by bluepenquin on Aug 21, 2020 13:01:55 GMT -5
It looks much similar to the US at large. If we look at the states that make up the South (7 day rolling totals). The peak was on 7/19 with 197,918 cases. Yesterday it was down to 106,460 cases for a 46% decrease. The 7 day rolling cases was roughly equal to 6/26. Pretty much a very even bell curve from 6/26 to 8/20 - with a peak on 7/19. If we look at Florida. The peak was 7/17 with 83,090 cases (7 day rolling). Yesterday, it was 31,465 - down 62% since the peak. Yesterday's rolling average is roughly equal to 6/26 with a near perfect looking bell curve. You cherry picked 6/26 because it was well above the 7 day average. 6/26 was also the single highest day to that point (even though it was over 3 months into the quarantine), so hardly an impressive number to get back to. If we more accurately use the 7 day moving average, it is not as nice of a bell curve, and we are still 10,000 cases a day higher than the 6/26 level (approx 45,000 to 35,000 daily cases). Yes, 8/17 was 31,465, but then 8/18 was 40,458, 8/19 was 45,103 and 8/20 was 43,245. The 8/17 date you cherry-picked looks to be a full 12,000 below the 7 day average. So you cherry picked a starting point 10,000 above the 7 day average and a finish point 12,000 below the 7 day average. covidtracking.com/data/charts/us-daily-positiveIn addition, the daily numbers are very cyclical on a weekly basis. Mondays almost always show a huge dip while there is often a large spike on Friday. 6/26 was a Friday while 8/17 was a Monday. Numbers I gave were 7 day rolls - because daily numbers are very cyclical. I didn't attend to 'cherry pick'. I was trying to point out that the peak that we went through in the South and Southwest was the middle of July. People seemed to have missed this - thinking we are currently going through, or sustaining a peak. That cases have been dropping since that peak 1 month ago and we are currently about where we were 1 month before the peak (late June). Wasn't really commenting on whether this is acceptable or that late June was acceptable. That said - if the trajectory would hold, there is considerable hope as to where we may be one month from now when College Football is set to begin. Below is the 7 day rolling total for Florida backing up to the beginning of June. It is clear that Florida's average cases per day is much lower today than one month ago and similar to where it was 2 months ago. Which is about 2X more than the month prior to that. 06/01 - 5,084 06/02 - 5,192 06/03 - 6,130 06/04 - 6,898 06/05 - 6,991 06/06 - 7,334 06/07 - 7,775 06/08 - 8,074 06/09 - 8,553 06/10 - 8,607 06/11 - 8,886 06/12 - 9,483 06/13 - 10,794 06/14 - 11,630 06/15 - 12,422 06/16 - 14,109 06/17 - 15,348 06/18 - 16,857 06/19 - 18,777 06/20 - 20,245 06/21 - 21,723 06/22 - 22,891 06/23 - 23,394 06/24 - 26,295 06/25 - 28,092 06/26 - 33,212 06/27 - 38,748 06/28 - 43,784 06/29 - 46,124 06/30 - 48,931 07/01 - 49,983 07/02 - 55,088 07/03 - 55,634 07/04 - 57,507 07/05 - 59,036 07/06 - 60,106 07/07 - 61,360 07/08 - 64,786 07/09 - 63,612 07/10 - 65,557 07/11 - 64,459 07/12 - 69,700 07/13 - 75,988 07/14 - 77,835 07/15 - 78,027 07/16 - 83,057 07/17 - 83,090 07/18 - 83,058 07/19 - 80,236 07/20 - 77,959 07/21 - 78,205 07/22 - 77,809 07/23 - 74,093 07/24 - 75,071 07/25 - 76,942 07/26 - 73,808 07/27 - 72,353 07/28 - 72,143 07/29 - 71,804 07/30 - 71,511 07/31 - 68,074 08/01 - 65,517 08/02 - 63,277 08/03 - 59,137 08/04 - 55,353 08/05 - 51,316 08/06 - 49,010 08/07 - 47,689 08/08 - 46,549 08/09 - 45,674 08/10 - 45,077 08/11 - 45,462 08/12 - 48,162 08/13 - 46,748 08/14 - 45,210 08/15 - 43,060 08/16 - 40,610 08/17 - 39,133 08/18 - 37,140 08/19 - 33,146 08/20 - 31,465
|
|
|
Post by bigjohn043 on Aug 21, 2020 13:32:09 GMT -5
|
|
|
Post by oldnewbie on Aug 21, 2020 13:39:07 GMT -5
You cherry picked 6/26 because it was well above the 7 day average. 6/26 was also the single highest day to that point (even though it was over 3 months into the quarantine), so hardly an impressive number to get back to. If we more accurately use the 7 day moving average, it is not as nice of a bell curve, and we are still 10,000 cases a day higher than the 6/26 level (approx 45,000 to 35,000 daily cases). Yes, 8/17 was 31,465, but then 8/18 was 40,458, 8/19 was 45,103 and 8/20 was 43,245. The 8/17 date you cherry-picked looks to be a full 12,000 below the 7 day average. So you cherry picked a starting point 10,000 above the 7 day average and a finish point 12,000 below the 7 day average. covidtracking.com/data/charts/us-daily-positiveIn addition, the daily numbers are very cyclical on a weekly basis. Mondays almost always show a huge dip while there is often a large spike on Friday. 6/26 was a Friday while 8/17 was a Monday. Numbers I gave were 7 day rolls - because daily numbers are very cyclical. I didn't attend to 'cherry pick'. I was trying to point out that the peak that we went through in the South and Southwest was the middle of July. People seemed to have missed this - thinking we are currently going through, or sustaining a peak. That cases have been dropping since that peak 1 month ago and we are currently about where we were 1 month before the peak (late June). Wasn't really commenting on whether this is acceptable or that late June was acceptable. That said - if the trajectory would hold, there is considerable hope as to where we may be one month from now when College Football is set to begin. Below is the 7 day rolling total for Florida backing up to the beginning of June. It is clear that Florida's average cases per day is much lower today than one month ago and similar to where it was 2 months ago. Which is about 2X more than the month prior to that. 06/01 - 5,084 06/02 - 5,192 06/03 - 6,130 06/04 - 6,898 06/05 - 6,991 06/06 - 7,334 06/07 - 7,775 06/08 - 8,074 06/09 - 8,553 06/10 - 8,607 06/11 - 8,886 06/12 - 9,483 06/13 - 10,794 06/14 - 11,630 06/15 - 12,422 06/16 - 14,109 06/17 - 15,348 06/18 - 16,857 06/19 - 18,777 06/20 - 20,245 06/21 - 21,723 06/22 - 22,891 06/23 - 23,394 06/24 - 26,295 06/25 - 28,092 06/26 - 33,212 06/27 - 38,748 06/28 - 43,784 06/29 - 46,124 06/30 - 48,931 07/01 - 49,983 07/02 - 55,088 07/03 - 55,634 07/04 - 57,507 07/05 - 59,036 07/06 - 60,106 07/07 - 61,360 07/08 - 64,786 07/09 - 63,612 07/10 - 65,557 07/11 - 64,459 07/12 - 69,700 07/13 - 75,988 07/14 - 77,835 07/15 - 78,027 07/16 - 83,057 07/17 - 83,090 07/18 - 83,058 07/19 - 80,236 07/20 - 77,959 07/21 - 78,205 07/22 - 77,809 07/23 - 74,093 07/24 - 75,071 07/25 - 76,942 07/26 - 73,808 07/27 - 72,353 07/28 - 72,143 07/29 - 71,804 07/30 - 71,511 07/31 - 68,074 08/01 - 65,517 08/02 - 63,277 08/03 - 59,137 08/04 - 55,353 08/05 - 51,316 08/06 - 49,010 08/07 - 47,689 08/08 - 46,549 08/09 - 45,674 08/10 - 45,077 08/11 - 45,462 08/12 - 48,162 08/13 - 46,748 08/14 - 45,210 08/15 - 43,060 08/16 - 40,610 08/17 - 39,133 08/18 - 37,140 08/19 - 33,146 08/20 - 31,465 The biggest issue I have is that you are looking at 2 months of a 5 month chart and calling it a bell curve, when the whole chart is nothing like a bell curve. The NCAA was projecting a bell curve that continued down from Early June. Instead we had a spike that went up by a factor of around 13 (eyeballing from your numbers) and is still 5 times higher. If you didn't know about the huge spike in between and just knew the first week of June and now, the interpretation would be of a huge failure and no way kids should be in school now. The fact that we actually had a massive spike in between doesn't somehow make that better, it makes it much worse. It doesn't make it any better knowing that the countries who are way ahead of us in getting their numbers down, are now shutting down again as a second wave begins (Korea, Germany, France, etc).
|
|
bluepenquin
Hall of Fame
4-Time VolleyTalk Poster of the Year (2019, 2018, 2017, 2016), All-VolleyTalk 1st Team (2021, 2020, 2019, 2018, 2017, 2016)
Posts: 12,447
|
Post by bluepenquin on Aug 21, 2020 14:17:13 GMT -5
The biggest issue I have is that you are looking at 2 months of a 5 month chart and calling it a bell curve, when the whole chart is nothing like a bell curve. The NCAA was projecting a bell curve that continued down from Early June. Instead we had a spike that went up by a factor of around 13 (eyeballing from your numbers) and is still 5 times higher. If you didn't know about the huge spike in between and just knew the first week of June and now, the interpretation would be of a huge failure and no way kids should be in school now. The fact that we actually had a massive spike in between doesn't somehow make that better, it makes it much worse. It doesn't make it any better knowing that the countries who are way ahead of us in getting their numbers down, are now shutting down again as a second wave begins (Korea, Germany, France, etc). This could be supporting the impact of herd immunity? I don't know about France - they haven't been much different than the US in the aggregate so don't know where the new cases are coming from? Korea and Germany never got hit very hard so it may be much more difficult for them to prevent outbreaks vs. say NYC or Florida at this point. If you look at the US in the aggregate - it looks like 2 big waves. However, it could also be viewed as single waves hitting different areas of the US. Where a place in the US got hit hard - it hasn't returned (yet). Does herd immunity play into this? I think this is currently being debated among epidemiologist.
|
|
|
Post by oldnewbie on Aug 21, 2020 15:01:46 GMT -5
The biggest issue I have is that you are looking at 2 months of a 5 month chart and calling it a bell curve, when the whole chart is nothing like a bell curve. The NCAA was projecting a bell curve that continued down from Early June. Instead we had a spike that went up by a factor of around 13 (eyeballing from your numbers) and is still 5 times higher. If you didn't know about the huge spike in between and just knew the first week of June and now, the interpretation would be of a huge failure and no way kids should be in school now. The fact that we actually had a massive spike in between doesn't somehow make that better, it makes it much worse. It doesn't make it any better knowing that the countries who are way ahead of us in getting their numbers down, are now shutting down again as a second wave begins (Korea, Germany, France, etc). This could be supporting the impact of herd immunity? I don't know about France - they haven't been much different than the US in the aggregate so don't know where the new cases are coming from? Korea and Germany never got hit very hard so it may be much more difficult for them to prevent outbreaks vs. say NYC or Florida at this point. If you look at the US in the aggregate - it looks like 2 big waves. However, it could also be viewed as single waves hitting different areas of the US. Where a place in the US got hit hard - it hasn't returned (yet). Does herd immunity play into this? I think this is currently being debated among epidemiologist. Herd immunity takes 60% to 80% saturation through either illness or vaccine. Highest estimated now (which are likely off by as much as a factor of 2 or 3) is 60 million. At 60% saturation (198 million) and assuming 60 million infections, your absolute best case is to have 3 times the current, numbers, so your dream scenario is to have another 350,000 deaths. Consider if it takes 80% and we have closer to 20 million actual cases, then you are talking about 13 times the current numbers for your herd immunity without a vaccine, or a total of 2.3 million deaths. Those numbers are not acceptable. I'd rather wait 5 or 6 months to start mass vaccinations.
|
|
bluepenquin
Hall of Fame
4-Time VolleyTalk Poster of the Year (2019, 2018, 2017, 2016), All-VolleyTalk 1st Team (2021, 2020, 2019, 2018, 2017, 2016)
Posts: 12,447
|
Post by bluepenquin on Aug 21, 2020 15:24:33 GMT -5
This could be supporting the impact of herd immunity? I don't know about France - they haven't been much different than the US in the aggregate so don't know where the new cases are coming from? Korea and Germany never got hit very hard so it may be much more difficult for them to prevent outbreaks vs. say NYC or Florida at this point. If you look at the US in the aggregate - it looks like 2 big waves. However, it could also be viewed as single waves hitting different areas of the US. Where a place in the US got hit hard - it hasn't returned (yet). Does herd immunity play into this? I think this is currently being debated among epidemiologist. Herd immunity takes 60% to 80% saturation through either illness or vaccine. Highest estimated now (which are likely off by as much as a factor of 2 or 3) is 60 million. At 60% saturation (198 million) and assuming 60 million infections, your absolute best case is to have 3 times the current, numbers, so your dream scenario is to have another 350,000 deaths. Consider if it takes 80% and we have closer to 20 million actual cases, then you are talking about 13 times the current numbers for your herd immunity without a vaccine, or a total of 2.3 million deaths. Those numbers are not acceptable. I'd rather wait 5 or 6 months to start mass vaccinations. www.nytimes.com/2020/08/17/health/coronavirus-herd-immunity.htmlDebatable. Sounds like the 'experts' used to believe 60-80%. Now, a significant number believe it is closer to 50% and some believe it may be as low as 10%. Not sure where the consensus or average is - possibly below 50%. Of course this doesn't address how long immunity would last - so far so good, but that isn't a guarantee. In any case - it is hard to believe that NYC's current low infection rate isn't in large part related to some form of herd immunity.
|
|
|
Post by bigjohn043 on Aug 21, 2020 15:44:11 GMT -5
The biggest issue I have is that you are looking at 2 months of a 5 month chart and calling it a bell curve, when the whole chart is nothing like a bell curve. The NCAA was projecting a bell curve that continued down from Early June. Instead we had a spike that went up by a factor of around 13 (eyeballing from your numbers) and is still 5 times higher. If you didn't know about the huge spike in between and just knew the first week of June and now, the interpretation would be of a huge failure and no way kids should be in school now. The fact that we actually had a massive spike in between doesn't somehow make that better, it makes it much worse. It doesn't make it any better knowing that the countries who are way ahead of us in getting their numbers down, are now shutting down again as a second wave begins (Korea, Germany, France, etc). Not sure if you looked at the chart. Yes cases did spike and now they are down by over 50% and falling quickly. There are some delays in diagnosing and then reporting cases so true infections are likely much lower right now. And I don't think you understand how the epidemiology math works. You can get lower new infections based on both what mitigation steps you are taking and how many have been infected and are immune. You can suppress an epidemic through mitigation steps but when you ease up on the mitigation the disease will come back. So whatever mitigation you take you need to hold until the end or it just comes back. Interestingly, we are not seeing a second wave in any of the places hit hard by the virus. In the US, NY/NJ have very low rates while the second wave happened in states not hit the first time. The same is true on the state level. Parishes in LA that were hit in the first wave and not seeing case increases. Those are all happening in areas not hit the first time. Same thing is true in ID were one county got hit hard in the first wave and is now not seeing increases while the rest of the state does. This is also happening in Spain / Italy where Madrid / Milan that got hit the first time are not getting hit the second time while the rest of the country is. And of course Sweden is having very few cases now despite taking very few steps. The best hypothesis for this pattern is some level of herd immunity in those places already hit. The post above is a NYT article on the idea. If you assume Sweden as a model we are about 10% away from where they are. There are lots of difference in the two countries that might drive us above them but the idea that deaths are going to triple just isn't very likely. And if you go through the numbers the whole key is actually who catches the disease. If the young and healthy catch it we build up a herd immunity with very low loss of life. The whole key to mitigation while actually moving on is to really lock down the elderly with everyone else taking reasonable precautions.....
|
|
|
Post by cindra on Aug 21, 2020 15:53:30 GMT -5
The biggest issue I have is that you are looking at 2 months of a 5 month chart and calling it a bell curve, when the whole chart is nothing like a bell curve. The NCAA was projecting a bell curve that continued down from Early June. Instead we had a spike that went up by a factor of around 13 (eyeballing from your numbers) and is still 5 times higher. If you didn't know about the huge spike in between and just knew the first week of June and now, the interpretation would be of a huge failure and no way kids should be in school now. The fact that we actually had a massive spike in between doesn't somehow make that better, it makes it much worse. It doesn't make it any better knowing that the countries who are way ahead of us in getting their numbers down, are now shutting down again as a second wave begins (Korea, Germany, France, etc). Not sure if you looked at the chart. Yes cases did spike and now they are down by over 50% and falling quickly. There are some delays in diagnosing and then reporting cases so true infections are likely much lower right now. And I don't think you understand how the epidemiology math works. You can get lower new infections based on both what mitigation steps you are taking and how many have been infected and are immune. You can suppress an epidemic through mitigation steps but when you ease up on the mitigation the disease will come back. So whatever mitigation you take you need to hold until the end or it just comes back. Interestingly, we are not seeing a second wave in any of the places hit hard by the virus. In the US, NY/NJ have very low rates while the second wave happened in states not hit the first time. The same is true on the state level. Parishes in LA that were hit in the first wave and not seeing case increases. Those are all happening in areas not hit the first time. Same thing is true in ID were one county got hit hard in the first wave and is now not seeing increases while the rest of the state does. This is also happening in Spain / Italy where Madrid / Milan that got hit the first time are not getting hit the second time while the rest of the country is. And of course Sweden is having very few cases now despite taking very few steps. The best hypothesis for this pattern is some level of herd immunity in those places already hit. The post above is a NYT article on the idea. If you assume Sweden as a model we are about 10% away from where they are. There are lots of difference in the two countries that might drive us above them but the idea that deaths are going to triple just isn't very likely. And if you go through the numbers the whole key is actually who catches the disease. If the young and healthy catch it we build up a herd immunity with very low loss of life. The whole key to mitigation while actually moving on is to really lock down the elderly with everyone else taking reasonable precautions..... Stop doing the sweden thing lmao. They're still at much higher daily cases per capita compared to norway and denmark, 8th globally in deaths per capita. They clearly messed up. I'd also note that another (more plausible) explanation for NY/NJ/etc current success is that they actually implemented effective mitigation after being the first and therefore hardest hit, especially since antibody tests indicate they probably don't have herd immunity.
|
|
|
Post by n00b on Aug 21, 2020 16:15:40 GMT -5
The fallout has begun...
|
|