It is interesting to see the differences between the two equations.
A noteworthy point is to ensure that when collecting data in this format to ensure days with no data are still shown in the spreadsheet too – as a formula like this only uses the data that is available.
Previously I’ve used pivot tables to automatically display days and dates with no data and input a 0 if no data is present. It may be a case with a formula and set up like this, that you have to get your athletes to submit ‘rest days’ with a zero training load, to ensure these days are accounted for in the calculations. Otherwise you can end up with a 28 day average that is spread over 40 days due to no data for rest days etc.
Shout out to Sean Williams for his time and feedback on this!