Please Normalize Time Series data if switching to state level

[catmba]

First, I want to give you all a huge thank you for the work you are doing on this front. It's very helpful!

While we would all appreciate data at the most granular level, I understand if the most granular level you can sustain in the U.S. is at the state level.

If that is the case, can you please roll-up the city/county counts to the new state line-item for the previous dates to 3/10? Example, Washington state obviously didn't have 0 cases until March 10th. I also think the counts on 3/10 are duplicated where you have county and a state row.

This time series data is very valuable and would have to lose its integrity.

If you all decide you need to some support for this cause, I would be happy to donate to help the data remain clean/most-granular, as others have mentioned.

Thank you.