The thing that's happening with ICOs today is that we're skipping steps, over-valuing things, and making impossible promises. The largest ICOs also tend to handwave away some very substantial, long-term research problems that the general audience they are pitching to has no awareness of.
Teams get put in a position where they raise 5 years of runway with rapid growth all at once, pre-product, pre-traction. That's not a problem with the ICO model, that's just a function of the space being extremely overheated.
But personally I would put my money on the good team over the bad regardless of the idea
If people wanted a structure that had some sort of accountability for non-delivery of results they'd do a share offering.