Comment by sdrinf

Ask HN: Predictions for 2009 -> http://news.ycombinator.com/item?id=416530

Evaluating the results of that set concludes the following predictions:

-All predictions in the form of "X is going to die" have a >~90% probability of being false. The market tends towards diversification, fellows!

-~90% percent of "X is going to increase / decrease" are non-quantifiable (esp. technologies, and tool usage), and thus have a predictive power of exactly 0.

-Interestingly, there were a single quantifiable stock prediction ("I predict Apple stock goes up a lot."), and it also had merit (a 100% growth in 1 year, how's that fellows?); alas 1 datapoint does not make a trend. (while we're at it, anyone else noticing the 200% growth G produced over 2009?! )

-The rest of the predictions are either non-quantifiable, or have a massively sub-optimal prediction result.

In conclusion, strictly from a predictive perspective, these threads have exactly 0 merit. Have fun synchronizing your agendas :)