Sadly, I have to agree with you on this "issue". There's barely any ICO's left that already have a working product, or atleast a MVP. I tend to invest in ICO's that already have something to show for it. Like Friendz recently, working product, active platform and generating revenue.
Although I can't deny that I never invest in "ideas" myself. Some projects look very interesting, and to be honest if they live up to their whitepaper it will be BIG. But in the end I'm just investing large sums of money into an "idea". Which is actually quite shocking.
RE: Nearly Half Of All ICOs Last Year Have Already Failed... And That's A Good Thing