As the election nears, I find myself reading Nate Silver's FiveThirtyEight blog daily. For those of you unfamiliar with Nate, he is a statistician who tracks the various voter polls. In addition to calculating the consensus of who is favored, he also provides insight about the various polls.
I'm mentioning FiveThirtyEight because in a Wednesday post, Nate discussed two statistical concepts that not only apply to presidential elections, but to investing as well. Specifically, he wrote about the difference between bias and accuracy.
Nate posted, "Bias, in a statistical sense, means missing consistently in one direction-for example, overrating the Republican's performance across a number of different examples, or the Democrat's. It is to be distinguished from the term accuracy, which refers to how close you come to the outcome in either direction. If our forecasts miss high on Mr. Obama's vote share by 10 percentage points in Nevada, but miss low on it by 10 percentage points in Iowa, our forecasts won't have been very accurate, but they also won't have been biased since the misses were in opposite directions (they'll just have been bad)."
When it comes to investing, a strong argument could be made for analysts' recommendations and earnings estimates being far more biased than inaccurate, especially when it comes to short-term forecasts. According to information from Zacks.com, the number of stocks recommended as a "buy" or better outnumber those recommended as a "sell" or worse by a margin of nearly 30-to-1. Thomson Reuters says 63% of companies have topped third-quarter earnings expectations. (The number is based on the 272 S&P 500 companies that have reported through last Friday; a more current number has not been released yet.) This earnings "beat rate" is higher than the long-term average of 62%, but lower than the average over the past four quarters of 67%.
Note the skew of the numbers. If the analysts were simply inaccurate, we would see more sell ratings and a larger number of negative earnings surprises (only 25.4% of reported companies have missed profit forecasts). Instead, there are a disproportionate number of buy ratings and a disproportionate number of earnings beats-a clear sign of bias.
It's easy to attack the analysts, and I'm certainly not the first person-nor will I be the last person-to point out their biases. But if you know a bias exists, you can look for ways to work around it. For instance, when we rank the performance of the AAII stock screens by risk-adjusted performance, two of the five best strategies require upward revisions to earnings estimates (Estimate Revisions: Up 5%, which Whirlpool (NYSE:WHR) passes, and P/E Relative Screen, which Monsanto (NYSE:MON) passes). On the other hand, two of the five screens with the worst risk-adjusted returns look for negative earnings estimates revisions (Estimate Revisions: Down 5% and Estimate Revisions: Lowest 30 Down). In other words, even though analysts are too conservative with their ratings and shorter-term profit forecasts, there is a benefit to paying attention to whether and how the profit forecasts are being revised.
Thus, while there will always be lies, darn lies and statistics, there are still ways to benefit from them if you are able to determine if, and how, the numbers are biased or if they are simply inaccurate.
As far as Nate Silver's election forecast, he blogged, "It is a close race. [President Obama's] chances of holding onto his Electoral College lead and converting it into another term are equivalent to the chances of an NFL team winning when it leads by a field goal with three minutes left to play in the fourth quarter. There are plenty of things that could go wrong, and sometimes they will. But it turns out that an NFL team that leads by a field goal with three minutes left to go winds up winning the game 79% of the time."
Charles Rotblut, CFA is a Vice President with the American Association of Individual Investors and editor of the AAII Journal.