Wednesday, 13 February 2013

Freakonomics Exhumed: Making Decisions By Tossing A Coin

As many of you who've had the pleasure will know, there is an excellent book by Steven D. Levitt and Stephen J. Dubner called Freakonomics, in which the authors apply economic theory and research to many fascinating social situations.  This is the sort of book that will show, for example, how legalisation of abortion in America went on to reduce crime rates 15-20 years henceforward in the States in which it happened, due to the fact that the future criminals were being aborted instead of being born.

I noticed on the Freakonomics webpage that they are now having fun with social science by researching how people get on when they flip a coin to decide on a course of action.  Whether it's deciding to move house, switch jobs, ask a girl out, change cities, ask for a divorce, go to a concert, or take up a sport or hobby, they want to know how it will turn out for you if you made the decision by flipping a coin instead of carefully thinking it through.

When reading Freakonomics a few years ago, I seem to recall Levitt and Dubner showing that companies that hire big reputation consultants and sports teams that hire prestigious coaches get from their investments no better results than those of a random coin toss.  I guess their current coin-toss research will eventually reach fruition by giving us some indication whether or not our carefully thought out choices work out better for us than those made by the random walk of coin tossing. 

Recording whether our strategic and so-called 'well thought out' decisions return a better than 50/50 payout could turn out to be valuable, or at the very least edifying. One note of caution though - it won't be entirely random, because although one could make a lot of life decisions using a random walk model, the decisions one chooses to submit to the random walk model mostly aren't random.   Whether the protagonist flipa coin to decide whether to shout at his neighbour, as opposed to, say, flipping a coin to decide whether to buy her some flowers or lend her a copy of Freakonomics is nothing like as random as coin tossing.  

My other issue with the experiment is that in a great many cases I don't think there is a reliable way to know what constitutes a good outcome.  Good compared to what?  In such an experiment the good is being weighed against the weight of the other alternative - but one can never know how that would have turned out.  Say Mrs Jones from London flips a coin, and on landing a tails decides to go to America for her holiday instead of Canada.  She has no way of knowing whether the results of the coin toss turned out better or worse, because she doesn't have the Canada trip with which to make the comparison.  Most of your life decisions may have turned out better or worse if you'd have picked an alternative decision, but in most cases you can't know whether the grass is greener on the other side or whether it's better the devil you know, irrespective of whether you decided by coin flip or careful consideration.

I do, however, agree with the authors that many of the results of our so-called 'well thought out' decisions are, in fact, just as random as the toss of a coin (this has been shown to be true in sports, management, consultancy and the stock exchange, to name but four examples).

Hopefully, though, you won't flip a coin to decide whether or not to carrying on reading my blog - otherwise my readership might shrink by 50%.