The truth about October markets

There is almost nothing unique about October, at least not in an investable sense. It's a mental trap, almost irresistible precisely because it's so simple.

People in general are superstitious lab rats hardwired to see patterns where none exist. That's because the world is an enormous, frightening place. There is an impossible amount of stimulus to process, almost all of it random. If we took in everything all at once our minds would be blown. It’s impossible. As a result our brains work off instinct, taking in only a fraction of what’s before us and filling in the rest from our generally unreliable memories.

For instance, we all know October is when stocks crash! We base that on a sample size of three. October 1987, 1929 and to a lesser degree 2008 we’re all horrible. Those memories stick out and as a result October is regarded as terrifying.

The reality is that October is pretty bland. It’s the 6th best month of the year, according to Ryan Detrick’s awesome SeeitMarket.com.

How could the unofficial month of stock market catastrophe be anything else? Actually it’s just math. The only reason it’s surprising is because of how our brains work. We remember pain more readily than pleasure. We’re risk-averse animals. Our ancestors who didn’t fear risk didn’t survive long enough to produce babies.

The S&P 500 in October of 2011
The S&P 500 in October of 2011

As a result we totally forget the 2011 11% October rally for two reasons. First, because 2011 was an almost flat year. Remember August was horrendous. Government shutdown etc. The second reason is painful events stand out. As a result of evolution stuff that hurts us is highly correlated to things that kill us. "October = bad" is to investor minds what "fire = burn" is to cavemen.

October in year six of presidential cycles has been fantastic since 1950. In fact, October has been higher on average over the last ten years despite falling 17% in 2008!

So October is good! That’s awesome! It’s reassuring! It’s garbage.

On a day to day basis stock market movements are statistically random. There are about 250 trading days a year (365 - 104 weekend days - 1 holiday per month). That makes 12,500 days over the last 50 years. It's overwhelming.

So we chop them into months and get a slightly more than 600 months since 1963 but only 50 Octobers! I can deal with 50, mentally but it could still use some chipping to make it more manageable. One way to do it is by connecting Washington DC and stock market moves via Presidential cycles.

Why Presidential cycles? Because it seems more mentally rigorous than using leap years (which would have exactly the same pattern since our Presidential cycle aligns with leap years), or hemlines or Robert Downey Jr. arrests divided by Iron Man box office. They’re all equally valid but Presidential cycles seems more serious than, say, taking note of whether the Super Bowl winning team was originally in the AFC or NFC.

October in year two of Presidential cycles has been higher 75% of the time. That's almost exactly the same frequency at which stocks have been higher over the course of any January for the last 50 years (37 of 50 per moneychimp.com).

October in year six of cycles has been higher 100% of the time since 1950 but there have been only four of them.

Going back 50 years October has been higher 30 times and lower 20 times. If plucking out four of those 30 positive instances and noting they happened to coincide with year six of the Eisenhower, Reagan, Clinton and George W. Bush administrations makes you feel more confident about the next 32 days that's great. We could all use a little more confidence. Just don't confuse confidence with brains.

More from Yahoo Finance:
Forget Hong Kong and Pimco — junk is driving this market1 year later: Buy stocks booted from the Dow?
PIMCO's new CIO: Who is Dan Ivascyn?

Advertisement