I'd be curious in seeing like 30 hamsters doing this, and comparing the statistics on their portfolios, and comparing to a uniformly-weighted (according to same

I'd be curious in seeing like 30 hamsters doing this, and comparing the statistics on their portfolios, and comparing to a uniformly-weighted (according to same probabilities) fixed (not actively traded) portfolios of the coins. With just one hamster it can just be random fluke. And also pick a few different start and stop times (cause looking at the graph, sometimes bitcoin came out ahead).