Let the masses think freely

I have been reading recently about the capability of groups to find solutions to certain types of problems. It came as a surprise to me that under certain conditions groups are better at solving problems than most of the individuals of the group, regardless of the individuals level of experience.

Examples of tasks at which groups are good solving them; when a class was challenged to estimate the number of balls in a jar (that actually contained 850 balls), the average of the answers was better than 94% of the individual answers.

Another good example of problem solving at which crowds perform well is betting, a bookie's job is to try to get as much betted money for a favourite than for an under dog because the margin he wins depends on this. If a lot of people bet for a favourite the booker will incentive bets for the under dog by modifying the odds, betters will see the improvement in the cost of opportunity of the under dog and will start betting for them. Let this system (bets - odds correction) run for a while and the average of the bets will give you a very accurate prediction of what is going to happen. In horse racing most of the time the bets reflect the order of arrival of the horses.

There are some conditions for the crowds to make good predictions; as the second example shows an incentive for the individuals is always good, but also it is important the independence of the individuals, and seems that more heterogeneous groups will lead to better solutions. But it is also important that there is a mechanism that takes the opinions into account and processes them (the bookie or the person doing the mean of the estimations of balls in the jar, the stock market is another aggregating mechanism, in this case it estimates the future value of companies).

It seems as if the human being has the gift to be able to take good collective decisions, but it also seems that the best collective decisions are taken if not in isolation (there is a need of a mechanism to put all the information together) at least with independence. All of this comes as a surprise to me, I have always doubted of the sensibility of the 'herd', and there are numerous examples of crowds going wrong. But I had never considered under which conditions the herd is able to make sound decisions. Maybe that is the fundamental lesson for us to learn, to be able to identify when are the conditions ideal for the group to make a decision or when the conditions have been naturally or forcefully altered so the group may not decide independently. After all isn't this what marketing is all about?