[source image]
Dow Jones NewsSep 11, 1:16 PM UTC
MW Gold keeps setting records - but its September trading pattern could be a roadblock
By Mark Hulbert
Seasonal strength doesn't show up year after year
Gold's surge to yet another all-time high is bringing attention to a seasonal trade that had fallen out of favor. I'm referring to the notion that gold (GC00) is an especially strong performer in the month of September.
As far as I can tell, this idea goes back to a study from the January 2013 issue of "Research in International Business and Finance." Entitled "The autumn effect of gold," the study found that, from 1980 to 2010, gold's average September performance was significantly better than in other months.
Unfortunately, this September pattern stopped soon after the study was completed. Of the 14 Septembers that have occurred since 2010, gold has declined in 11 - a 79% failure rate. Moreover, gold bullion lost an average of 2.2% over that time, versus an average gain of 0.8% for the other 11 months. This reversal of gold's September returns was strong enough to weaken the statistical strength of the study's conclusion significantly.
This history provides context for gold's impressive rally so far this month. In the six trading days since the end of August, gold is up about 5%. Is this month's rally strong enough to justify believing that gold's September strength has returned?
The answer depends on whether there is a plausible explanation for why gold should perform better in September than in other months. Without a strong theoretical foundation, there is too high a probability that the pattern is a statistical fluke.
I am not aware of any theoretical foundation. When I ask analysts why gold would be particularly strong in September, the most frequent answer is that gold should benefit from the same seasonal pattern where stocks perform especially poorly in the month.
But this theory doesn't hold water. Since 2010, the Dow Jones Industrial Average DJIA has lost an average of 1.0% in September (versus an average gain of 1.0% for the other 11 months). If gold were benefiting from money being withdrawn from the stock market, gold should have performed well in September since 2010.
Investing in trends and tendencies
I draw different investment implications from the reversal of gold's September tendencies. One is that betting on seasonal patterns is not for short-term traders, and instead requires consistency and discipline over many years.
Consider gold's September record over the 30 Septembers included in the study. Gold rose in 21 of them - 70% of the time. Even if each September after 2010 had been as good for gold as those before, you still would have had a 30% chance of losing money by betting on gold in any one September. The only way of translating that percentage into a reliably profitable strategy would have been to bet on gold in each September for many years in a row. So it's difficult, if not impossible, to determine whether gold's return in a particular September is anything more than luck.
If you start following a strategy because of its statistically significant track record, you'll need to stick it out. Abandon it only if, with the addition of new data, its long-term record stops being statistically significant. That might mean you need to follow the strategy despite a string of losses.
For example, gold experienced especially large losses in September 2013 (5.4%), 2015 (7.0%) and 2016 (7.4%). Those losses were larger than in any previous September since 1980, and no doubt by 2016, a trader betting on gold's September strength would have been more than ready to throw in the towel. And yet, over the entire period from 1980 to 2016, gold's September advantage over other months was still statistically significant at the 93% confidence level - remarkably close to the 95% level that statisticians often use when determining if a pattern is genuine.
So believers in gold's September seasonal strength shouldn't get too excited about gold's rally so far this month. It might be nothing more than a random fluke. If it instead means that the seasonal pattern has returned after a long absence, we won't know for many years.
Mark Hulbert is a regular contributor to MarketWatch. His Hulbert Ratings tracks investment newsletters that pay a flat fee to be audited. He can be reached at mark@hulbertratings.com
More: Investors are exuberant about stocks, bonds and gold. And that's not necessarily good.
Plus: These are the stock-market trades to make going into the end of the year, according to Goldman Sachs
-Mark Hulbert
This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.
(END) Dow Jones Newswires
09-11-25 0916ET
Copyright © 2025 Dow Jones & Company, Inc.
