The idea is not exactly the same, but IMO it is pretty similar to something I remember from a book by John Brunner. (I don’t remember the title.)
The idea, as best I can remember, went like this:
If you do a survey, and ask people a factual question regarding when some historical event took place, most replies will not be accurate – but they will be close to accurate. You can plot a graph, and the right answer will be close to the middle of the range of answers.
So, if you want to know aprox. when some likely future event will happen, do a survey. Ask people, when will there be a manned expedition to Mars, when will the stock market hit (whatever number), when will marijuana be legalized in the U.S., when will Osama bin Laden be caught or killed? For each question you’ll get a range of answers. The correct answer will be somewhere near the middle of the range.
“Shockwave Rider” could be it. It was about a guy on the run who kept moving to a new location and changing his name / apearence / career. Part of the background of the story, I think, was that there had been a major earthquake in California. Any major upheaval was referred to as a bayquake.
Well, the proximate inspiration was clearly not John Brunner, but rather a bunch of experiments over the last decade or so using futures markets to predict non-financial things such as presidential elections.
The most famous of these is the Iowa Presidential futures market, which is now part of the larger Iowa Electronic Markets:
<scratches head> I’d never heard of of the Delphi technique before, but after reading the link, it surely has nothing to do with a futures market.
(A futures market is also not like a survey. The crucial difference being that in an ordinary survey, a person has no positive incentive to reveal what they actually think is going to happen. For example, if someone asks you, “Who do you think will win the election, A or B?” you may respond “A,” because you want A to win, not because you think A will win.
On the other hand, if someone asks you, “Who do you think will win? I’ll give you $10 if you guess right.” you now have a reason to reveal what you think will happen instead of what you want to happen.)