LA Times Data - Interesting trends in "cases by county" data - Printable Version
+- The CardBoard (https://thecardboard.org/board)
+-- Forum: Emergency (https://thecardboard.org/board/forum-11.html)
+--- Forum: Covid-19 (https://thecardboard.org/board/forum-12.html)
+--- Thread: LA Times Data - Interesting trends in "cases by county" data (/thread-19736.html)
+- The CardBoard (https://thecardboard.org/board)
+-- Forum: Emergency (https://thecardboard.org/board/forum-11.html)
+--- Forum: Covid-19 (https://thecardboard.org/board/forum-12.html)
+--- Thread: LA Times Data - Interesting trends in "cases by county" data (/thread-19736.html)
LA Times Data - Interesting trends in "cases by county" data - Leftcoast - 05-18-2020
The LA Times by county data shows two trends we've long expected would emerge.
1. Highest growth rate (as opposed to new cases) in the inland counties - Fresno, Tulare, San Bernardino, Kings, Imperial, and, not in trend, Humboldt. We've long expected the inland counties were less serious about shelter-in-place and that CV-19 would expand more quickly there once it got a toe hold. The overall rate is not horrible but they are in the 10-17 day doubling range.
2. Alameda's curve has crossed the near flat-lined Santa Clara and will likely assume the overall Bay Area lead in total cases within the week. This was also expected since CV-19 is hitting lower income areas harder.
Also, LA county is still climbing relatively rapidly (doubling every 25 days) and all the most impacted counties are in So Cal.
https://www.latimes.com/projects/california-coronavirus-cases-tracking-outbreak/
RE: LA Times Data - Interesting trends in "cases by county" data - M T - 05-18-2020
The LA Times graphic that has doubling time is, IMO, poorly done. There are lines coming from the origin that indicate doubling per month, week, etc. The first thing I thought was that where the endpoint was for a county meant that the doubling time was between the two doubling lines surrounding it, but that is not the case.
Doubling rate may be a way to view things in the first few weeks of an outbreak, but is mostly meaningless now. Consider if you had 1M cases in Big county in the February, but now had 1,000 cases a day. All those early cases are no longer infectious so they shouldn't clog the statistics of growth rate. Consider if it had 500 cases 2 weeks ago, 1000 cases 1 week ago, 2000 cases this week. Consider Little county having 200 cases a day every day for a month. Growth rate is constant, but high value.
RE: LA Times Data - Interesting trends in "cases by county" data - Leftcoast - 05-18-2020
I find it useful. The lines coming from the origin show the slope for each doubling frequency and were incredibly useful at the beginning of the infection, less so now. The doubling rate looks to be based on a 7-day average and shows how long it would take to double at that rate.
It works for me since the chart identifies the highly impacted counties by cumulative cases (LA, San Diego, Riverside) plus the high growth rate counties by color code and slope (Fresno, Humboldt, Kings).
RE: LA Times Data - Interesting trends in "cases by county" data - Farm93 - 05-18-2020
The graphs work for me, but that is likely because I have been looking at the LA Times and Worldometers graphs and tables each day.
I think Gov Newsom will need to do a bit of work to set CA expectations as we move forward. In NY their case counts climbed quickly and they are now experiencing a quick drop in cases and deaths. Perhaps it could have been worse, but they had a few weeks in there with a lot of cases and deaths. Meanwhile CA really did a lot more to flatten the curve at the onset. Even now CA ranks #29 in deaths per capita even though CA has some of the largest and most densely populated metropolitan areas in the USA. That was a great accomplishment, but that may mean we don't see a dramatic drop in cases like we may see in other states.