Rating Conclusions Explained


The ratings I assigned to what I determined to be the accuracy of each analyst recommendation ranged from One to Five (1-5). Five (5) is the highest rating and One (1) is the lowest. A rating of Three (3) is assigned if there was no substantial change to the share price, with a margin for error, up or down, of three percent (3%). A rating of Five (5) was given for a recommendation that substantially exceeded expectations, that is, very very good. A rating of Four (4) was given if the recommendation met or exceeded expectations, which is good, a rating of Two (2) if the recommendation fell short of expectations, which is not so good, and a rating of One (1) if the recommendation fell significantly short of expectations, which is quite obviously atrocious.

However, there is a small degree of leeway, and dare I say, discretion, in assigning ratings at the margins of those price movements. For example, if a stock increased in price by say 35%, which by almost any measure is an outstanding result had you bought the stock, which would automatically qualify it for a five (5) rating, it might be given only a Four (4) rating, if say it was recommended as a “Hold” because the result begs the question, why wasn’t a “Buy” recommendation made. Still, it would be a good result and classified as a successful recommendation.

Having said that, as an act of generosity to the analysts, if for example, a recommendation of a “Moderate Sell” is made, and the share price, to the chagrin of those who acted on the advice, increases by up to 5%, we might assign a Neutral rating, giving the analyst the benefit of the doubt in not making the investor any more money as opposed to losing the investor their money.

To arrive at the aggregated summary, we take the proportion of accurate ratings (4 or 5) compared to the inaccurate ratings (1 or 2) as a percentage of total ratings. We disregard the Neutral rating, we do not include it in calculating the total accuracy percentage. Therefore, if the final summary percentage is greater than 50% the analysts are getting it right more often than they are getting it wrong, and of course the converse applies.