Hey VTC. I totally understand your point and I think the CP risk is an important aspect to take into account as an investor. I'll address this again in a moment.
However, I just want to quickly point out that while divesting everyone 75% when we were at 32,000 would make the max PROFIT (not max bet) 80 just as it did by lowering max profit to 0.25% of invested funds, these are not actually mathematically equal. Remember these numbers are dynamic. For example, let's compare these two scenarios, A and B respectively if the site a player comes in and loses 1000BTC (site wins 1000 BTC):
A) Invested goes from 8,000 to 9,000. Max profit goes from 1% * 8,000 = 80 to 1% * 9,000 = 90
B) Invested goes from 32,000 to 33,000. Max profit goes from 0.25% * 32,000 = 80 to 0.25% * 33,000 = 82.5
The same goes if a player wins 1000 BTC (site loses 1000BTC):
A) Invested goes from 8,000 to 9,000. Max profit goes from 1% * 8,000 = 80 to 1% * 7,000 = 70
B) Invested goes from 32,000 to 33,000. Max profit goes from 0.25% * 32,000 = 80 to 0.25% * 31,000 = 77.5
Thus, what Dooglus actually did (reduce max profit to 0.25%) is fundamentally different-- it reduced risk, not just adjusted the max profit downward.
Now, having said that, I wasn't thrilled either about him reducing the max profit, and I still believe going 100% Kelly Criterion is the way to go. HOWEVER, I understand his motivation (trying to avoid what would essentially be a bank run), and I think his new plan is fantastic!
Would this analysis change if we assume that the marginal investor sitting on the sidelines, waiting to get in to a lucrative investment, can jump in quickly? He has incentive because each new BTC invested after the whale attack gets a bigger-than-normal piece of the pie (until it reaches equilibrium)?
If we assume the marginal investor (waiting on the sidelines) can respond instantly, then perhaps the two cases *are* mathematically the same? (Well except for the fact that with the lower max bet % comes increased counter-party and legal risk.)