So par v udi was postponed until 8th April 2015, week 29 (row 290)

If you extend your calculations in columns Z and to the right, to row 290 or more, then change the 29 in cell D290 to 24, it goes a long way to putting things right.
Before you make that change, you'll notice also, that 11 fixtures were played for week 29; that's going to snarl things up.



This goes back to my msg#8 about what is a 'round' and how is it defined? At the moment, you've said it was league and week. Should it be something different?

However:
In your first message in this thread you said you wanted "to figure home and away teams' league position ahead of their match"

I see that in your expected results you seem to have included the ranks of the missing udi and par calculated from an earlier week/round's values in columns AN:AO. However, the values for those teams are for fewer matches played, so I think they're not comparable… but you may say different.

I think it could be possible to look at the most recent values (regardless of week) for all the teams playing in a given league and use those scores in columns AN:AO to rank them. This would make the formula even more complicated, and with your version of Excel, longer, to the point that it might be easier to have a user defined function (macro function) which would need vba code and for macros to be enabled and the file would become a .xlsm file. Acceptable?

A Power Query solution might also be possible (built in to Excel since version 2016) and would look neater.

I'll await comments on all the above before doing anything.