Thursday, April 22, 2010

Is winning so good!!

Few weeks ago at the end of the IPL season III match between Mumbai Indians and Deccan Chargers, I was struck with a curious thought. Even though cricket as a sport is notorious for throwing up unexpected results, Mumbai Indians were the favorites in the match going by their track record and the little master’s current blistering form. And they won quite convincingly. They so far have played 8 matches and won 7 of them. Therefore by the very basic principle of probability, Mumbai Indians have 7/ 8 probability of winning the next match/matches provided the set of teams do not change (which is not going to change at least in this season), playing rules and conditions do not change, and no tie result happens as well. In philosophy this is also called the problem of induction- wherein we tend to predict the future on the basis of past events and outcomes. In this case, this may be also the prevalent way of predicting the result of IPL matches. If you were a bookie, chances are you will follow the inductive method.

The less obvious way of looking at the phenomenon kept bothering me for a while. If we could some way know about the winning potential of a particular team then the complete picture would change drastically. For example if we could know that Mumbai Indians have a finite capacity for winning matches, then with every win they reduce their potential for winning in the subsequent matches. This is counterintuitive and goes in direct opposition to probabilistic or inductive approaches. However, there are plenty of proof that evince against the utility of probabilistic approaches in real world. For example if probabilistic approach held true then big companies would have become bigger and bigger with the passage of time; dynasties would have remained undefeatable and so on and so forth. Clearly that is not how history unfolds. In any ecology the winner does not keep winning the games following some probabilistic design. If we were to look at the world a little empirically we have to just look around to find examples. For example take a cross section of the dominant corporations at any particular time, many of them will be out of business a few decades later, while firms nobody ever heard of will have popped into the scene and replaced the old ones.

Consider the following sobering statistics. Of the five hundred largest U.S companies in 1957, only seventy-four were still part of that select group, the Standard and Poor’s 500 (an exclusive group top 500 listed companies in the US by earnings), forty years later. Only a few had disappeared in mergers, the rest either shrank or went bust. Many of those 500 giants have simply disappeared from our collective memory without a trace. So those monsters could not gobble up the tiny upstarts along the way; most of the big guns mostly could not survive the test of time. Therefore it would be quite logical to say that an entity has a finite though not measurable capacity for winning; they cannot go on winning forever. As a corollary to this statement, we can also say that with every win you actually reduce your chance of winning the next match. Therefore for a Mumbai Indian fan it may not be a very good news that her team has been enjoying a winning streak so far. In the abstraction of the argument winning suddenly does not look so appealing.

No comments:

Post a Comment