Ask Colin

What do you think about the consistent patterns concerning seasonal trading (long Nov. through April) and US presidential cycles?

I worry about these studies. I may be wrong, but it seems to me that there is the old question about averages involved here. It is not every year that shows this pattern and not every cycle that shows this pattern. So, they add them all up and average them, or come up with some percentage of the time that the rule holds.

The old average problem is best explained in terms of the man who drowned in a river with an average depth of one foot. Or the man with one hand in a bucket of ice and the other in a fire - on average he is the right temperature.

Another way to put it is from my friend Garnett Znidaric who tells about getting a 97% result in his pilot's test on navigation. When he commented to the instructor that he thought 97% was a good result, the instructor said that he hoped the 3% would not kill him sometime!

So, I think there is some value in knowing these tendencies. However, their value may lie as much in detecting when they fail as when they hold.