Plan

One central idea of John Lanchester is this: Over the past 30 years, there was an increasing trend of financial people to use complicated mathematical models and reduce the risk of approving a loan – basically by shifting the risk to other people, who are generally not aware of the risk.

But, the catch is, these geniuses in number were probably wrong from the very beginning, and their calculation was based on wrong assumptions. In short, the risk is not a happening of which the probability is five (or fifty) standard deviations above the average, but is the sudden change in the universe so that the distribution probability curve becomes no longer Gaussian.

It sounds familiar, eh?  My friend VW recently described how MBA holders and lay people build towers of nuts and marshmallows (see http://vwswong.blogspot.hk/2012/11/marshmallow-challenge.html).

And the idea is simple. If you have some relevant background knowledge, it is very well to formulate a plan for a project. If you have no such knowledge, it is fine to go ahead and adjust your plan along the way. The worst situation is, however, to plan ahead with irrelevant information – it generates bias, defers decisions, and easily leads to disasters.