LONDON, Jan 9 (Reuters) - Seasonal bellwether or just hocus pocus, the first five trading days bode well for 2013.
Given the often unfathomable global risks facing investors at the start of any new year, it's not hard to see the temptation of a few rules of thumb to parse the next 12 months.
Whether history repeats or just rhymes, seasonal trading patterns captured in well-thumbed investment almanacs are typically a trove of ideas for market timing and direction.
And so enter the "five day rule", often cited by Jim O'Neill, Chairman of Goldman Sachs Asset Management, as an uncanny portent of the year's trading.
Since 1950, the direction of the S&P500 index of U.S. blue chips in the first five trading days of the year has predicted the full year direction more than 85 percent of the time.
Tuesday marked the fifth market day of 2013 and, lo and behold, there's been a net gain of almost 2 percent - similar to last year and thanks mainly to a New Year's Eve deal in Washington to avert the tax and spending crunch known as the "fiscal cliff".
To be fair, the burst started late in 2012. The rally on New Year's Eve itself was the biggest final day gain since 1974 and December was positive overall despite a negative fourth quarter.
But the five-day rule worked sweetly as a way to navigate a topsy-turvy 2012, where the early surge led to annual gains of more than 13 percent. And that was despite the sort of heavy losses in the second and third quarters that would have made for a nervy year for tactical traders like hedge funds.
Of course, it's never that simple and you only have to look back as recently as 2011 for a hiccup. The first five days gave the green light that year too, but the market ended the year flat - not a disaster for anyone picking up the dividends but no success for the 'rule' either.
But prior to that, there were only four bum steers since 1950 - 2002, 1990, 1973 and 1966.
So, is there any more to this pattern than waiting for a Punxsutawny groundhog to predict when wintry weather ends?
Market traders love repetitive price patterns that give an otherwise often chaotic world a semblance of predictability. Chart analysis has flourished for decades in modern markets as a way for systematic traders to exploit waves and sequences in prices that take more from mathematical than behavioural rules.
But of course the logic behind those theories is less important than whether trading strategies built on them - often sophisticated computer-based models - average out in the money over time. And many are just self-fulfilling prophecies as a result as brokers and dealers follow them simply because others do. In other words, they work, well, until they don't.
Seasonal flows and trading almanacs are slightly different however in that they purport to capture observed behaviour rather than abstract concepts or predetermined patterns.
The hoary old stock market adage "Sell in May and go away and don't return 'til St Ledgers Day (September)" still packs a punch every Spring as early-year activity wanes into the thinner, more volatile summer months and after many national budget and tax years have ended in March or April.
So how reliable is the five day rule?
Deutsche Bank strategist and market historian Jim Reid prefers to broaden things out to what he calls the "January effect" that shows more often than not there are stronger asset returns in the first month of the year.
Using S&P500 monthly averages since data started in 1928 the average price return for all January months was a gain 1.73 percent - the highest monthly return of all calendar months, Reid pointed out.
"Clearly we are not taking a view here based on some historical statistics but it does tell us something about seasonal patterns in markets," he told clients.
It may well be that the accuracy of the five-day rule is rooted in the fact that over 50 years Wall St mostly clocks in positive annual gains anyway - making the first throes of typically bullish Januarys seem more prescient at first glance.
Since 1962, for example, positive years for the S&P500 outnumber negative ones three to one.
As to the fundamental reasoning? Could it really be as simple as big investors waiting for the first week of the year to place their bets with conviction and changing their minds less frequently than we assume?
That sounds far fetched in the modern era at least.
Most asset managers devise annual year ahead strategies as far back November and few will wait to the new year to execute. Even then, their behaviour is heavily conditioned by investment committees, trustees and client parameters.
That said, retail flows may well be more calendar based. And there may well be a reluctance to trust market prices in the holiday-strewn final few weeks of a year, leaving some pent up activity until January.
Yet, like chart analysis, the "flow" may just have become self-reinforcing over time as brokers and market makers second-guess seasonal flow regardless of any evidence in front of them.
So ignore the five day rule at your peril clearly, but don't seek an easy explanation - not even from its biggest proponent.
"I can't fathom it out," said O'Neill. Economically "it doesn't really make sense." (Editing by Ron Askew)