Interesting. I’m not sure if this will be workable for you, but let’s look anyway.
It looks to me that you are using the time field as the ‘label column’ - this is clever, generating one bar per day - something I’ve not thought of doing before. Trouble is, as you noted, the value of the label field gets printed as-is, in this case, as a huge number. Not good.
I wonder if you can transform the data in the query. I don’t recognize the query editor you are using - is this influxDB? Whatever, I wonder if you cant transform the time value into something printable first and then apply the group operator. I mostly use SQL and CSV data so I’m not sure what your query language supports. The trick is to get it into a table format where the contents are formatted in a way you can display. Something like YYYY-MM-DD (this would be a good choice as it sorts properly too).
As you noted, Multistat currently only displays a single quantity for each bar - this is something I’m thinking of addressing some time soon-ish. But not right away.
I have an idea that might work, if you are willing to try thinking outside the box.
see if you can write the query to return the two values in separate rows (again, I don’t know if your query editor will support this, but in SQL this might use a UNION operator.
Turn something like
date solar usage
2019-02-28 1234 2345
2019-03-01 1111 2222
2019-03-02 1212 2323
date valuetype value
2019-02-28 solar 1234
2019-02-28 usage 2345
2019-03-01 solar 1111
2019-03-01 usage 2222
2019-03-02 solar 1212
2019-03-02 usage 2323
Then use ‘date’ as the label column, ‘value’ as the value column and valuetype as the ‘group column’
This should create two bar charts, one for the solar values and one for the usage values. It’s not quite what you are looking for, but it’s a start.
Good luck and send me some data sets in the original two value column formats to play with while I think about adding multi-column support out of the box in a future release.