i don't think i could divine or dictate these numbers any better than u could,
 but i do know how to get them.

first we don't mind overpaying a little the first few months, but if were going to pay in orange
we have to be able to control the amount.  eventually we pay just enough to keep em lining up for a MNs, but we don't just throw tons of money away for decades.

idk, so what does it cost to set up/and maintain an MN?  we want to be generous, especially in the beginning. this has to work.

if someone invests say $1000 or a couple of btc for a year, we should probable arrange at least a $300 to $500 (30-50%) profit for the first 6 months or so, but even for that period we would need to control OC amounts, unless we promise btc or USD equivalents (which would also vary OC amounts)

i think there is also a limit to the number of MN we will need at a given moment based on historical user traffic ratios,
50% more than we seem to need would be one thing, while 10x what we need would be something else.

so we want to overpay at first because this service has to be first class all the way, and we want to always have a fluent number of nodes, ie more than we need by some comfortable %, say 15-30% maybe.
 
but we need a way to control the OC payouts down the road, we simply cannot make long range payment plans in a currency we know is going up dramatically, but not how much, this would be self destructive.

yeah halo if dark needs any piece of wise-up intel, it's that 4 mil figure, which is almost 10% of their cap
so every ten years or so these simple trans stations get the whole bleeding cake

anyway we need control down the road, no getting around it, that is the mistake dark made, and we don't need to repeat it.

i think a tranpay (earlier post) method might solve all the problems with relatively little programming,
and it would only cost a fraction of the blind payment plan.  But anything that allows the market to function naturally would work.