I wrote some python code to simulate what might happen with different coinage possibilities.
https://privatepaste.com/ae9dfdbb6cIt starts with 69917 txs staking, 75% of which are "JD" and are distributed with values based on the stats published by dooglus a couple of days ago. The remaining 25% are "Not JD" and have a similar distribution of value. It starts assuming the block count is 69917 and every tx is marked with which block it appeared, chosen at random.
The code is slow. I ran it with 5 different coin age algorithms for 1500 "ticks" of a clock. This takes about 5 minutes but it corresponds to what would be a few hours of really staking. I'll give the results below, but they really should be run for much longer to get more reliable numbers. So far it doesn't seem like there is any difference between the algorithms. In every case JD stakes about 80% and there in about 4% of cases Not JD could have staked but got orphaned.
Each coinage function assumes coins mature after 510 blocks. After that the age is either (1) constant [like clams is now], (2) depends on the log of age (i.e., how many blocks old it is), (3) depends on the square root of the age, (4) depends linearly on the age, (5) depends on the square of the age.
(1) Constant, like now:
clamsim.sim(1500,clamsim.coinage_const_mature510)
Total blocks: 270
JD would stake: 218 -- 80.7407407407%
Not JD would stake: 52 -- 19.2592592593%
Not JD would get orphaned: 13 -- 4.81481481481%
(2) Log age:
clamsim.sim(1500,clamsim.coinage_log_mature510)
Total blocks: 368
JD would stake: 295 -- 80.1630434783%
Not JD would stake: 73 -- 19.8369565217%
Not JD would get orphaned: 13 -- 3.53260869565%
(3) Square root of age:
clamsim.sim(1500,clamsim.coinage_sqrt_mature510)
Total blocks: 389
JD would stake: 310 -- 79.6915167095%
Not JD would stake: 79 -- 20.3084832905%
Not JD would get orphaned: 13 -- 3.34190231362%
(4) Linear:
clamsim.sim(1500,clamsim.coinage_linear_mature510)
Total blocks: 732
JD would stake: 594 -- 81.1475409836%
Not JD would stake: 138 -- 18.8524590164%
Not JD would get orphaned: 31 -- 4.2349726776%
(5) Square of age:
clamsim.sim(1500,clamsim.coinage_square_mature510)
Total blocks: 779
JD would stake: 633 -- 81.2580231065%
Not JD would stake: 146 -- 18.7419768935%
Not JD would get orphaned: 33 -- 4.23620025674%
If someone wants, I'll run one or two of these (or some other coinage algorithm) overnight to get better numbers.