model allocation where % allocated depends on multiple variables

I have separate from this a way to determine if I should buy or sell certain stocks. Based on if I get a buy or sell signal, I want to make allocations. However, depending on different combinations of buys or sells, I would have different allocations and this is based on a tiering system. I will attempt to explain and diagram. I have attached a simplified version of it with explanations of what I have built and why in the attached spreadsheet. Any and all help is welcome. Thank you in advance.

Exampleeach 3 letter is meant to be a stock symbol)

I have 4 stocks in tier one:
AAA, BBB, CCC, DDD if both AAA and BBB are BUYS, then they would each get a 25% allocation. If BBB signals a BUY it would get 25. Same thing for CCC, they are not dependent on each other.
If AAA and BBB are both not a BUY, then 50% becomes available to the next tier. If CCC is not a BUY 25% becomes available to the next tier and if DDD is not a BUY 25% becomes available to the next tier. It is possible for 100% to be available for the next tier. (for this problem, I have a solution in the spreadsheet).

Here is the tricky part for me.
Tier two has 3 stocks: EEE, FFF, GGG
In order to have any allocation, EEE has to signal BUY, same for FFF, same for GGG. Each one of these securities can be as low as zero or as high as 100% allocation. It depends on if there is a % left over from tier one - because if all of tier one is a BUY- then nothing left over for tier two even if they signal BUY. If there is a % left to be allocated to- it needs a prorata share with any other tier two buy signals. (this is where I get stuck)

Tier 3 is easy. Anything left over automatically gets allocated there.

I hope between this and the sample spreadsheet I put together that it is not too confusing. Thank you again.

In D6, then copy down

=IF(C6=1,(100-SUM($D$2:$D$5))/COUNTIF($C$6:$C$8,1),0)

Jason,

I was close. You solved it. Thank you!