If you not familiar with the Victorian stock-market adage, investors are told to dump their shares in May. They are also warned not to step back into the market until after St. Leger Day, which is when the last of Britain's five Classic horse races of the year is run. This year, it took place on Saturday 14 September.
This year's race was won by the favourite, which, interestingly, doesn't seem at all likely in another race - the race to be the next Federal Reserve chairman.
Summer's over
Larry Summers, who was the front runner to replace Ben Bernanke as the head of the Federal Reserve, announced that he had pulled out of the contest. The news sparked a rally - a St. Leger Day rally - in global equities.
But over the last 25 years, this bizarre piece of market-timing has only worked on 12 occasions and failed miserably on 13 others. Nevertheless, some still stand by it. Mind you, if you had sold in May this year, you would, admittedly, have side-stepped a 6% fall in the Straits Times Index over the turbulent summer months.
On the face of it, the Sell in Mayeffect appears to have redeemed itself this year. But we mustn't forget that that there are trading costs to consider whenever we buy and sell shares. Additionally, if you are an income investor, you might miss out on dividends too.
Even though Sell in Maywas right this year, that still means it has only been correct half the time, which makes it no more accurate than a flip of a coin. The coin-toss school of investing is definitely not one that I subscribe to. It is not something that any of us should follow either.
Interestingly, Sell in May isn't the only market-timing strategy. There is a school of thought that claims January is the best month for buying shares.
A number of reasons have been suggested to explain the so-called January Effect. The most plausible is a rebound in share prices following year-end tax selling, which, it should be said, tends to be more prevalent in America. 
Monday, Monday
Here in Singapore, the stock market has risen on 12 separate occasions in January and fallen on 14 others over the last 26 years. That is hardly a ringing endorsement for the January Effect.
Apart from the January Effect, researchers have reported apparent irregularities in stock market returns by examining other calendar-related effects too. For example, the Monday Effect suggests that the first day of the trading week tends to be the worst day to be invested in shares.
Meanwhile, another group of researcher claims that shares tend to show higher returns on the last day and the first four days of each calendar month. This has been cutely dubbed the Turn of the Month Effect. Other anomalies grouped by calendar frequency include the Week of the Month Effect, the Holiday Effect and the Santa Rally in the run up to Christmas.
It has to be said that there is a long tradition for trawling through stock market data in search of anomalies. This is in the futile hope of finding systematic patterns that might lead to easy profits.
The Most Dangerous Month
However, blindly buying and selling shares based on calendar effects can often be a short cut to financial ruin. Additionally, any marginal gain that might be made could easily be swallowed up by extra trading costs, commission charges and loss of dividends.
Here we don't buy shares on the basis of some antediluvian saying. Instead, we look for investments that will reward us over the long term. When you find such an investment, then the best time to buy is always...now.
Mark Twain once mused: "October is one of the most dangerous months to speculate in shares. The others are July, January, September, April, November, May, March, June, December, August and February".
Consequently, if you are an investor, it is much better not to try and speculate on calendar effects but to stay invested for the long haul. That is because the best months to invest is ...every month. The best time to buy shares is...all the time.