A Correlation Does Not Equal Cause and Effect

There is a lot of prognostication going around which attempts to predict what the stock market will do based on past trends. You see this sort of thing all the time—sometimes it is correct, but often it’s simply randomness which would equal out in the long run. Correlations do not necessarily show ‘cause and effect’, although the news media often presents them that way. Remember the study of eating a good breakfast with eggs—it led to better performance at school. Did it really, or was it a correlation based on the fact that getting a good breakfast was also correlated with having an involved parent who stayed at home, or perhaps being from a wealthy family that could afford good breakfasts, or even perhaps a maid to cook them? The reprinted article below gives a classic example of what looks to be a predictor, but obviously isn’t: There is a strong correlation between who wins the Super Bowl and how the stock market performs. Makes you wonder about financial advisors vs. flipping a coin.

 

Schaeffer’s Investment Research

 

What the Super Bowl Indicator and randomness means for your portfolio...

My Senior Quantitative Analyst shares his insight on the market

Dear Jon,

Over the weekend Rocky White, Senior Quantitative Analyst here at Schaeffer's Investment Research, wrote an article about the Super Bowl Indicator and what we can learn from it as part our free newsletter, Monday Morning Outlook.

I didn't want you to miss out on Rocky's insight about how patterns, as well as randomness, can affect the market, so I've included his article below.

Indicator of the Week: The Super Bowl and Randomness
By Rocky White, Senior Quantitative Analyst

Foreword: I figure it's a good time to bring up the Super Bowl Indicator (it's a real thing) and what we should learn from it. It's fun stuff to know, but I don't think I'll hear many people argue that the winner of the Super Bowl gives any indication of where the market is heading. However, there is a very valuable lesson to be learned from it.

Super Bowl Indicator: The Super Bowl Indicator points out the fact that whenever a team from the NFC has won the big game, then the stock market has done very well for the rest of the year. When a team from the AFC wins, then the market has done mediocre. Below are the actual Dow returns for the rest of the year after the Super Bowl depending on which conference won. When an NFC team has won, then the Dow has been positive 88% of the time for the rest of the year -- averaging a gain of almost 11%. When an AFC team wins, then it averages a gain of just 3%.

Schaeffer’s Investment Research

You can't really argue with the numbers. Those are facts. However, I'm sure there's hardly anyone who believes an NFC team winning the Super Bowl creates a tendency for stock prices to climb. In other words, correlation does not mean causation.

More Randomness: Here is another example to show my point. I looked at yearly stock returns for 2012 and 2013 for all optionable stocks meeting certain liquidity criteria. The tables below show their performance based on the second letter of their ticker symbol. Why the second letter? Because some time ago, I heard an honest theory that when money enters the market, stocks with a letter early in the alphabet may outperform since investors often see lists of stocks in alphabetical order and buy the first good one they see. Judging performance by the second letter doesn't allow for that ridiculous theory.

Look at what I found. Stocks whose second letter in their symbol is P, for example Gap (GPS) or JPMorgan Chase (JPM), have the highest average yearly return of 16%, and stocks with the letter G second in their symbol have the worst at just 2.3%.

What does that tell us? "Absolutely nothing," is the correct response to that question.

Schaeffer’s Investment Research

The Lesson: The lesson of all of this is to buy shares of United Parcel Service (UPS) if the Seahawks win the Super Bowl. No, that's a joke. It's a lesson in randomness. The patterns that were formed from the data above (an outperforming market when the NFC wins the Super Bowl or stocks whose second letter is "P" doing better than other stocks) are obviously entirely due to randomness. It's obvious because the analysis is so silly. However, it's not always that clear. Randomness exists in all stock-market data, and it would be wise to always consider this. In fact, no one can even tell you if the randomness is the main reason for an observed pattern, or if there's actually something driving the apparent correlation.

For example, below is data from an article I wrote just three weeks ago. It shows that over the past 40 years when the S&P 500 Index (SPX) gained in January, then the rest of the year has also done well, and when the index declines then the rest of the year has struggled. Now, that pattern is easy for us to grasp and we can come up with a perfectly plausible explanation. We can say something to the effect that January returns create some sort of buying momentum that lasts the rest of the year, so when January is good it just flows to all the following months. It sounds reasonable and maybe, in fact, it is perfectly true. Maybe there is some sort of dynamic in which January returns affect -- or even predict -- the rest of the year. Or maybe (some people would say likely) the pattern in the data below is simply due to randomness.

Schaeffer’s Investment Research

The fact is that all we know to be true is the actual data. We know that when the NFC has won the Super Bowl, then stocks have done well. In that circumstance, we can be pretty certain it's because of randomness. It happened like that by accident. We also know the fact that when stocks have done well in January, then they have also done well for next 11 months. Is it simply due to randomness? Or is there something driving that correlation? Maybe there is a reason for that pattern, but it's not as dramatic as the data suggests and only part of the difference is due to randomness. It's actually impossible for us to know the answer to that.

I was a bit hesitant to write this article because almost every week I'm pointing out a pattern and making comments on it, and I don't want to lose credibility (or more importantly, readers, haha). So let me end it by saying that I think it's beneficial to study data and know patterns, like January has been a good predictor for the rest of the year or that the Dow has tended to go higher in the first several months after breaking below its 50-day moving average. In that second case, it's good to know the data just to refute those who say it's a signal of a market breaking down. However, whatever the pattern is that we're looking at, do not dismiss the possibility that randomness is the dominant reason for it.

To get more information...

I hope you found Rocky's column as interesting as I did. Here at Schaeffer's we are constantly analyzing the market and researching information that we feel is relevant for you as a trader.

For more information about Monday Morning Outlook or any of the topics that Rocky discussed in his column, you can call 1-800-448-2080 Ext. 1251 Monday through Friday 8:30 a.m. to 5:30 p.m. ET and we'd be happy to help you.

All the best,

Bernie Schaeffer, Chairman & CEO

Bernie Schaeffer
Chairman & CEO
Schaeffer's Investment Research, Inc
5151 Pfeiffer Rd, Ste 250
Cincinnati, OH 45242
1-800-448-2080 Ext. 1251
International 1-513-589-3800 Ext. 1251





Schaeffer’s Investment Research

If you have any questions about SchaeffersResearch.com, Bernie Schaeffer or any of our products or services, send an e-mail message to This email address is being protected from spambots. You need JavaScript enabled to view it. '; document.write(''); document.write(addy_text20126); document.write('<\/a>'); //-->\n This email address is being protected from spambots. You need JavaScript enabled to view it. or please call us at 1-800-448-2080.

Although there is significant profit potential associated with buying options, there is also the risk of losing the amount invested in any individual trade. In any option buying approach, it is expected that losing trades will be more numerous than winning trades. The goal is for the average gain to be significantly greater than the average loss so that the bottom line is profitable.

If you wish to be excluded from future notices, do not reply to this message. Instead, please go to the following address:
http://ads.schaeffersresearch.com/schaeffers-unsubscribe-page/?o=115135&e= This email address is being protected from spambots. You need JavaScript enabled to view it.

Your subscription fee may be 100% TAX DEDUCTIBLE if used for business/investment purposes. This offer is subject to the terms and conditions located at: http://www.schaeffersresearch.com/AS