A different approach - that has less (but still present) risk is to research the companies and projects during their ICO phase, but don't buy in yet.

A general cycle that's apparent (and been taught to me) is that when an ICO doesn't blow up from a scam, hack, or loss, it generally dumps for the first 1-2 months.

Once it bottoms - and if you've done your research verifying it's a project you feel strongly towards - buy it then.

Sure you won't get ICO-level pricing, but likely not far from it.

Since the vast majority of the risk is at the ICO stage, the slightly higher cost post-dump from those who did buy at the ICO stage will be worth the wait.

Just my two cents, but this gives you plenty of time to research the right projects and buy at the most opportune times instead of gambling on the ICO succeeding.