Feb

9

A Plethora of Auguries

February 9, 2025 |

For those who don't have a statistical mind or who have not been exposed to card games or gambling, the possibility of a random explanation for many phenomena is hard to grasp. Simply put, what we observe in real life is a sample from a population. Even if the sample is random, the mean of the values observed from the sample is likely to differ from the population mean. The differences observed for many market phenomena are merely due to sampling variations resulting from the large number of samples taken and the high variability within the population.

Consider, for example, the world famous Superbowl indicator. Germany watches it like a hawk. If the winner of America's Superbowl is a team that hailed originally from the NFL, the signal is bullish. If a team originally from the AFL wins, the signal is bearish. The moves in stocks in the 12 months following the NFL and AFL wins are enumerated and summarized in table 3.3.

The results are striking. The total change in the DJIA during the 12 months following victories by teams originally in the NFL has been 4412.08. The average change is 259 Dow points. During years when AFL teams were victorious, the total DJIA change has been -7.3. Reviewing the average ranks of changes in the Dow with respect to the two origins of winners, it turns out that differences as large as these could be explained by chance only once in 300 occurrences. But how many different sporting events could be served as a benchmark predictor with equal plausibility? The World Series, the NBA Championship, the NHL Stanley Cup, the New Year's Day bowl games, or the won-lost record of the Chicago Bulls or the New York Yankees over a single season? Each of the winners might provide an independent prediction of the Dow. Reformulating the question: Since 1967, what are the chances that, in five major annual sporting events, the win by one of the 10 competitors will be associated with the DJIA's rising at least 250 points in that year? The answer, determined by simulation, turns out to be about 1/2. And if not predictive of the Dow, how do bonds match up with the winners?

The Superbowl analysis shows why the random walk or efficient markets model is often consistent with many of the effects found by technical analysts or the more dressed up retrospective anomalies of the academics.

When people ask me whether markets are random, I feel as I did when people asked me who, in my opinion, was the best squash player of all time. "Jack Barnaby, my coach," I always replied.

- The Education of a Speculator, pages 78-82.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search