Ray Dalio kind of tells a similar story about his experiences with a failed prediction almost bankrupting him after believing he had it all figured out. His response was to craft a culture of radical transparency where all people were questioned and checked and where dissenting voices were spotlighted and considered on merit. Its not a perfect system, but it has been overall successful for up to this point.
On Hive, I think we can benefit by hearing out some dissenting voices and giving them honest consideration based on the merit of the ideas. I think we do a decent job of this on DHF proposals, but I also think there are some opportunities to hear out other suggestions.
Mostly stochastic prediction method can't predict a single outcome, and definitely can't if the sample is outside the modeled distribution. This is just statistics.
Yes, and the more variables there are, the more difficult it is to predict. My interests are more along the lines of business operations strategies rather than market speculation.