Would you base your market bets on the Super Bowl?

This time of year, it seems, we read about two so-called patterns in the market—"As January goes, so goes the rest of the year" and the "Super Bowl Indicator" of the stock market. While these two may not seem related, they actually are.

The Super Bowl Indicator, which predicts what stocks will do based on the outcome of the game, is riding a six-year winning streak.

Since the advent of securities trading, investors and market commentators have looked for patterns in returns. In essence, investors want to take a complex system—the stock market—and simplify it. The Holy Grail of the investor is a simple trading rule that magically unlocks the complexities of the market.

Getty Images

The saying "As January goes, so goes the rest of the year" simply refers to the belief that if the stock market goes up in January, it will advance the rest of the year. Alternatively, if the market goes down in January, the rest of the year will be negative.

Some impatient observers have even shortened January to the first two trading days of the year in an effort to get an early insight into the fortunes of the market. More about this relationship later.

Meanwhile, the Super Bowl theory of the stock market was originally proposed by New York Times sportswriter Leonard Koppett in 1978. He simply stated that in 10 out of 11 years, the direction of the stock market was foretold by the outcome of the Super Bowl.

Koppett observed that if an old (pre-merger) NFL team won the Super Bowl, the market closed higher for the year, and if an old AFL team won, the market closed down for the year.

Read MoreHow pro athletes retire

When Koppett described the pattern, the only inconsistency was in 1970, when the Kansas City Chiefs won the Super Bowl and the Dow Jones Industrial Average advanced. However, even that year the Dow went up only 4.8 percent.

There was very little attention paid to the relationship when first proposed, but as the years went on, the remarkable "accuracy" of the indicator caused people to take notice. Headlines such as "Stock Indicator Says Root for Seahawks in the Super Bowl" and "Why Investors Should Root for the Giants to Win Super Bowl" have appeared in the business section of newspapers in very recent years.

As far back as 1989, the venerable Financial Analysts Journal published an article entitled "Did Joe Montana Save the Stock Market?" The piece referred to Montana's game-winning drive, which allowed the old-NFL San Francisco 49ers to snatch victory from the Cincinnati Bengals.

Vote

Vote to see results

Total Votes:

Not a Scientific Survey. Results may not total 100% due to rounding.

The indicator has even caused some market watchers to debate how to handle teams that have come into existence since the NFL-AFL merger.

For example, last year's Super Bowl featured two teams—the Baltimore Ravens and Seattle Seahawks—that didn't even exist pre-merger. For purposes of this analysis, I consider the Baltimore Ravens (Super Bowl winners in 2001) an AFC team.

One could argue that they are an old-NFL team because they became the Baltimore Ravens when Art Modell moved from Cleveland, where they were the Browns. But since the NFC added a team in Cleveland and called them the Cleveland Browns, I considered that double counting. Besides, just ask anyone in Cleveland if the Ravens are an old-NFL team.

Suggesting such a thing would be considered blasphemous and could lead to one incurring bodily harm. The Seahawks are an interesting case, as the team came into existence several years after the merger and was originally an NFC team that became an AFC team and is now back to being an NFC team. Since it started as an NFC team, I will stick with that.

Read MoreNew boon to college football

So how accurate is the Super Bowl indicator? It has been correct in 39 out of 48 years—an amazing 81 percent of the time. And not only does it do a remarkable job predicting the direction of the market, but the returns when an old NFL team wins average an 11.3 percent, whereas the average returns when an old AFL team wins are slightly negative.

So should we all sit in front of the TV on Sunday and root for the Seattle Seahawks to win the Super Bowl? Are we convinced the New England Patriots will deflate the stock market?

As ESPN football analyst Lee Corso says: "Not so fast, my friend."

The explanation is actually very simple. The stock market usually advances (the Dow was up in 35 of 48 years), and an old NFL team usually wins the Super Bowl (34 of 48 years). In fact, the number of times both of these events occurred in the same year was 29. Since 1966, there is a 70.8 percent chance that a pre-merger NFL team wins the Super Bowl, and there is a 72.9 percent chance that the Dow advances in a given year.

Given these probabilities, by simple chance the Super Bowl Indicator should be correct 59.5 percent of the time. This is estimated by calculating the statistical likelihood of the two circumstances in which the indicator will be correct:

1. NFL wins and market goes up.
2. AFL wins and market goes down.

Most people incorrectly compare this record to a pure chance of 50 percent.

Now, from a statistical standpoint, the fact that the indicator is correct 75 percent of the time is nothing to sneeze at. The probability of having the indicator correct 36 or more times out of 48 is less than 2 percent—a level that statisticians would say is indeed significant. It seems, therefore, that the Super Bowl Indicator has promise.

Before I became a college president, I was a faculty member in the Heider College of Business at Creighton University. I used the example of the Super Bowl stock market indicator as an illustration of the difference between correlation and causality.

Two series of numbers are correlated if—for whatever reason—they tend to move together. In our case, old-NFL teams generally win the Super Bowl, and the stock market generally rises.

However, simple correlation does not imply a sort of cause-and-effect relationship between the two series or that the relationship will continue into the future. When two series are correlated but there is no reason to infer causality, we can conclude that there is spurious (or simply random) correlation.

If you search long and hard enough, you can identify closely correlated variables that are obviously not causally related. In fact, the fascination with spurious correlations has grown to the point where there is a website detailing some of the more absurd relationships.

Read MoreFree furniture after football win

Two of the more curious are that the divorce rate in Maine correlates with the per-capita consumption of margarine in the U.S. and that the number of people who drowned in a swimming pool in a given year correlates with the number of films Nicolas Cage appeared during the same 12 months.

Now, intellectually we all understand that there is no cause-and-effect relationship between the number of people who drowned in a swimming pool and the number of films that Nicolas Cage appeared in. (Film critics,however, might argue there is a strong causal relationship between the number of films Cage appeared in and the number of bad films he appeared in—contending the correlation is close to one.)

Yet why does the Super Bowl theory of the stock market hold such a prominent place and isn't simply relegated to obscure correlation websites? It is because it combines two American obsessions—the stock market and the Super Bowl. It also plays into investors' desire to make the complex simple.

Now back to "As January goes, so goes the rest of the year." Over the Super Bowl era (since 1967), the returns to January (using the S&P 500) and the returns to the rest of the year have been directionally consistent in 32 of 48 years, for a success rate of 66.7 percent.

Again, it sounds as if an indicator that accurate is potentially a valuable guide. Examining this a little more closely shows that, on average, the market generally goes up in January; in fact, it advances 58.3 percent of the time.

Likewise, from February through year-end, the market advances a stellar 75 percent of the time. By simple chance, the January indicator should be correct 57.8 percent of the time. This is a level that statisticians would say is not significant or, in layman's terms, could very well happen by mere chance or luck.

Read MoreBudweiser's Super Bowl strategy

A post-mortem on the "January" indicator shows that it has not been consistent with the theory in four of the last six years. In fact, if you exited the market after the poor January in 2014, when the S&P 500 was off 2.9 percent, you would have missed out on the rest of the year, when returns were in excess of 14 percent.

The bottom line is that before you buy into some stock market theory, look behind the simple correlation to see if you believe there is an underlying rationale or causality that is plausible.

I doubt many market watchers can conceive of a cause-and- effect relationship between the outcome of a football game and the direction of the stock market.

However, the perceived patterns are perpetuated because many people accept superstitions—such as sitting in their favorite chair to increase the odds of their team winning, or not watching the game with a certain friend because their team lost the last two times they viewed the game together.

And, by the way, when Leonard Koppett first introduced the Super Bowl theory of the stock market, he did it tongue in cheek. He likely would be amazed at the acclaim his seemingly innocuous observation has achieved. Koppett passed away in 2003.

—By Bob Johnson, CFA, CAIA, president and CEO of The American College