The data as designed is not going to cooperate with making a chart like this. However with a few quick fixes, we can easily make the desired output.
See Sheet 6 where the data has been exported, reimported as a new data source and then pivoted. The pivoted data allows us to utilize a single Date Dimension field and a single Hours Dimension field. With this, we can create a Sum(Hours) field that provides the sum of hours by day/hour. Next, we can create a window_avg() function call to average this value across the Dimension for each Day.
Using Measure Values will not allow for this. This is a good example of examining data design and making sure it is easy to manipulate for reporting. Similiar concepts can be reviewed in this Think Data Thursday regarding Survey data which is another data format that is typically not designed well to work with reporting such as Tableau. - Think Data Thursday - Visualizing Survey Data - YouTube
I hope this helps.