Message Font: Serif | Sans-Serif

No. of Recommendations: 0

From Hidden Gem to CAPS Blog, one persisting question that bothers all Fools is whether or not CAPS succeeds in draw out diverse opinions from different people, or if it is just a place for people to hop onto a bandwaggon and perpetuate a line of thought that may or may not be right in the first place?

This is more or less my research interest in graduate school. Should we try to set up a place that attract lots of people and then simply weight their opinions by just taking a simple average of their rating on different stocks (incidentally, CAPS does not do this-- it gives preferential weight to those with more stock picks and those who achieve the "All-Star" status)? Or, should we simply check the comments for in-depth analyses? Many seem to believe that CAPS discourages in-depth analysis because it only gives you enough space for a "60-second" pitch and this creates a bandwaggon effect where people vote on a stock by simply following what other people have done.

I believe we are ignoring a basic mathematical concept-- the law of large numbers. And a corrollary to that is the inverse relationship between variance and sample size. Francis Galton, Darwin's cousin, first found that if we were just to take a numerical mean of all estimates given by different people, the average of all estimates came very close to the actual value. Why? This is because variance in estimatation is inversely related to the number of estimates sampled.

Recent research has also found that if we take estimates from people with different backgrounds and level of expertise (read: it is better to take estimates from experts AND non-experts rather than just experts alone!!), the final average will get even more precise. Again, this has a lot to do with how variance in estimation is reduced through smart sampling technique.

So, CAPS is a great tool, right? Ah, not so fast. While having more people giving estimates can give us an average that is closer to the actual value, CAPS calculates your score partly based on how many homerun stocks you can find. To see how this can bring about the bandwaggon effect, tune in tomorrow.

(Sorry to do this but I need a break)

Cheers