[M T]

Santa Clara testing counts are climbing while positive test percentage is dropping per the county dashboard (see below).  The overall positive rate is 6.6% since the inception of the dashboard but it's much lower since the expanded testing approximately 12 days ago.  Do we have similar data for LA county?



https://www.sccgov.org/sites/covid19/Pag...board.aspx



(Relevant dashboard is the third block from the top)

LA City (not incl. Long Beach, Pasadena) dashboard.  (Click on Testing on the left.  Then Daily.)
L.A. apparently doesn't figure that the epidemic is bad enough to have people work on weekends, so their number of tests fall over the weekend. (I'm not sure if it is test givers, test processing, or the people that take the results and put them in the charts)
So their numbers jump all around.  

You can download data at "Counts by Date Table".  (I realized while composing this note that this does NOT allow you to compute %positives.)

These data do NOT show the number of positive tests, but shows the number of cases and the number of tests.    If a person tests positive on April 1, but then tests negative later, (or vice versa) maybe they're going back and changing the accounting that is a case or not.    That's appropriate for reporting the number of cases, for any particular point in time, but alters the measured positive rate of the daily tests.    (Given that, you can have fun guessing why for 5 days, they had to adjust the number of new cases by roughly 4%.) 

I took a snapshot of their data and compared it to a snapshot 2 days later (this morning).  At 12 days back, there were > 1800 tests added for that one day.  At 13 days back, there were nearly 2800 tests added for that one day.  There was not a single day over the past month that didn't change in the "new persons tested" (some are negative, but none were 0).   Same thing is true for "new cases" (ie, daily cases).  Of the past 35 days that I checked, only one day's daily new cases didn't change (17 days back).   12 to 16 days back, the new numbers reduce the daily cases by 22 to 49 per day.   At that 17 days back, the total number of positives (over all time up to 76 days ago) was up by 191, then got reduced over those 5 days by 171 (to just up by 20). 

Except for a few stragglers, recorded deaths only increased over the most recent 8 days (from the earlier snapshot; 10 days relative to now.) 

The number of tests over the last four 7-day periods: (old to new):  27,737; 40,998; 37,030; 26,551
The most recent number (26K) is under-reported at this time.  I would not be surprised if another 4K tests came in from 7 to 14 days back.  So testing may be increasing, but not much in the past 2 weeks.

(Note that SCC has "Positive Results" equal to "Total Cases" which seems unlikely.  So "Positive Results" probably isn't a raw number of outcome over the number of tests given.)

[M T]

Santa Clara testing counts are climbing while positive test percentage is dropping per the county dashboard (see below).  The overall positive rate is 6.6% since the inception of the dashboard but it's much lower since the expanded testing approximately 12 days ago.  Do we have similar data for LA county?



https://www.sccgov.org/sites/covid19/Pag...board.aspx



(Relevant dashboard is the third block from the top)

LA City (not incl. Long Beach, Pasadena) dashboard.  (Click on Testing on the left.  Then Daily.)
L.A. apparently doesn't figure that the epidemic is bad enough to have people work on weekends, so their number of tests fall over the weekend. (I'm not sure if it is test givers, test processing, or the people that take the results and put them in the charts)
So their numbers jump all around.  

You can download data at "Counts by Date Table".  (I realized while composing this note that this does NOT allow you to compute %positives.)

These data do NOT show the number of positive tests, but shows the number of cases and the number of tests.    If a person tests positive on April 1, but then tests negative later, (or vice versa) maybe they're going back and changing the accounting that is a case or not.    That's appropriate for reporting the number of cases, for any particular point in time, but alters the measured positive rate of the daily tests.    (Given that, you can have fun guessing why for 5 days, they had to adjust the number of new cases by roughly 4%.) 

I took a snapshot of their data and compared it to a snapshot 2 days later (this morning).  At 12 days back, there were > 1800 tests added for that one day.  At 13 days back, there were nearly 2800 tests added for that one day.  There was not a single day over the past month that didn't change in the "new persons tested" (some are negative, but none were 0).   Same thing is true for "new cases" (ie, daily cases).  Of the past 35 days that I checked, only one day's daily new cases didn't change (17 days back).   12 to 16 days back, the new numbers reduce the daily cases by 22 to 49 per day.   At that 17 days back, the total number of positives (over all time up to 76 days ago) was up by 191, then got reduced over those 5 days by 171 (to just up by 20). 

Except for a few stragglers, recorded deaths only increased over the most recent 8 days (from the earlier snapshot; 10 days relative to now.) 

The number of tests over the last four 7-day periods: (old to new):  27,737; 40,998; 37,030; 26,551
The most recent number (26K) is under-reported at this time.  I would not be surprised if another 4K tests came in from 7 to 14 days back.  So testing may be increasing, but not much in the past 2 weeks.

(Note that SCC has "Positive Results" equal to "Total Cases" which seems unlikely.  So "Positive Results" probably isn't a raw number of outcome over the number of tests given.)