Florida Data 4/13/2020

[czdz]

You're showing over 100,000 cases in Florida on 4/13/2020. This does not seem correct.

[MuletownBrown]

Check Okaloosa County ... 102,103?

[vk4tdi]

Looks to me like there is a comma missing for Okaloosa County (FIPS=12091). Line ends: ...69,78,80,83,91,101,102103

[Ricks-Lab]

Is this also impacting the global confirmed file?

[jkbishay]

Report shows Okaloosa should in fact be 103 cases, not 102,103: https://www.thedestinlog.com/news/20200413/coronavirus-by-countycity-statewide-deaths-at-499-okaloosa-santa-rosa-report-more-cases-other-nwf-counties-remain-unchanged

This is likely the due to the error @vk4tdi pointed out and it appears to be affecting just about all files for April 13th.

[aakash-sahai]

I can tell that you are getting this erroneous data from 1point3acres, where I first noticed the issue. They have fixed the error since. You just need to refresh the data.

[PeterDowdy]

Also reported in https://github.com/CSSEGISandData/COVID-19/issues/2166; consolidating into this one

[tylerhg]

There has not been any new commits so I don't know how the issue could be fixed. The data is still showing 682,619 total cases in the US.

[advaithasabnis]

They'll notice and fix soon since their dashboard is also showing the wrong number so I guess just wait for it.

[GerardoRTobar]

@tylerhg It hasn't been fixed yet. Actually, their online dashboard shows that error too...

I hope they fix it soon

[aakash-sahai]

Guys, their dashboard was fixed at least 15 minutes ago. Please refresh.

[aakash-sahai]

[GerardoRTobar]

mmm nope, it still shows 682,619... I just refreshed. How many cases does it show to you @aakash-sahai?

[aakash-sahai]

Hmm, that is strange. I posted the screenshot with a 6:15 PM MDT timestamp. It is 6:35 PM MDT now. Some web-cache has it wrong, if others see a different picture.

[vk4tdi]

1point3acres gets US data from https://covidtracking.com/, which references Florida's dashboard. Florida's official number is 103 for Okaloosa.

[czdz]

You guys are discussing two different dashboards. 1point3acres has been fixed. Johns Hopkins dashboard has not been fixed.

[aakash-sahai]

Cool. Things are improving. Top level US numbers look good on JHU dashboard, but FL numbers are still inflated.

[ei-JoanneT]

They'll notice and fix soon since their dashboard is also showing the wrong number so I guess just wait for it.

Do they usually fix wrong data points at night? Have to deliver something tomorrow morning so I am not sure if I should manually fix it on my end or wait

[GerardoRTobar]

The fixed it already, but the daily report. They haven't updated the time series, which is the one we use the most, I think.

[aakash-sahai]

Whole world looks at JHU dashboard. Very few look at 1point3acres. It does not bode well for you guys. Have you contacted them to fix the time-series data?

[duh931]

https://github.com/CSSEGISandData/COVID-19/pull/2170 Opened this for hot fix. Not sure how soon they will respond though.

[aaroncolesmith]

Just pulled a corrected version

[ei-JoanneT]

In the daily report, the Okaloosa confirmed case for 04.13 is fixed, yet the active case is still showing 102102

[BJReplay]

Just pulled a corrected version

Will this be included in the global time series or do you need a fix for that submitted as well?

[aaroncolesmith]

time_series_covid19_confirmed_global.csv looks to be fixed; time_series_covid19_confirmed_US.csv looks to still have incorrect data

[aaroncolesmith]

Now both time_series_covid19_confirmed_global & time_series_covid19_confirmed_US look fine

[aakash-sahai]

All else looks good but the total confirmed cases for world is still inflated, and the plot for US and world too.

[MBtech]

The daily reports file still had the wrong number of active cases. Made the pull request #2171 to fix that