This is how you do a sliding scale:1.) Max return is whatever the highest investor(s) is willing to lose. Doesn't matter if it's percentage or a literal number.Edit: A lower tier could possibly crowd out a higher tier, but I don't think that's a problem. They would just get lumped together in the payout. Edit2: Removed the stupid.
Max Return = (Tier1Limit*Tier1Bankroll)+(Tier2Limit*Tier2Bankroll)... etc
or
Max return = (.0025*2000)+(0.005*2000)+(0.01*2000)
2.) If a gambler loses the bet, divide the sum by the number of "tiers" that intersect it.Example:
All investors have 400btc stake, and there are 5 investors in each tier.
35 btc max return, and 35 btc lost by the gambler.
1% tier = 17.5btc
.5% tier = 8.75btc
.25% tier = 8.75btc
Note: Tiers do not necessarily need to be percentages.
Edit: Also, the distribution could be very different due to different bankrolls in the tiers.
3.) The top tier is divided among the top investor(s) in proportion to their bankroll.
4.) Each following tier is divided equally among every investor in the tier and all tiers above, in proportion to their portion of these tier(s) bankroll.Continued Example:
1% tier would yield 3.5 bitcoins for all 1% investors. 17.5*(400/2000)
.5% tier would yield .875 bitcoins for every tier .5% investors
and tier 1% investors. 8.75*(400/4000)
.25% tier would yield .58 btc to each .25% investor and all tiers above. 8.75*(400/6000)
Totals:
.25% investor --> .58btc
.5% investor --> 1.45btc
1% investor --> 4.95btc
The losses propagate samesies. Each person gets a return/loss in proportion to the amount they "contribute" to the bet.
My example doesn't quite come out to 35, because I did some rounding for brevity, but that's the idea. While theoretically you could have an arbitrary amount of tiers, you likely would want to limit them.
Edit: Updated numbers to something more realistic.