I have finally setup a system to find the stocks that have the highest ratio of All-Star recommendations to total recommendations as defined in my first post here. It takes periodic samples of the ~3000 stocks with more than 50 ratings from the CAPS website and allows me to run the necessary calculations.

For the first week I changed the math slightly to use the difference between the ratios instead of the ratio of the ratios between All-Stars and the total population. I think this way it is a bit more reflective of what I was trying to accomplish in finding the stocks where the smart money disagrees with the average money most drastically. I ran the numbers and found the top ranked stocks then setup a separate account on CAPS to record my predictions. You can see a list of the current stocks I am predicting to be undervalued in a table lower on that page, as well as their price movement since I predicted them. Every week I will adjust the account to reflect the new top ranked stocks if there are any differences. As time progresses and I add more predictions these tests should provide pretty accurate results of my ranking mechanism; the error margin is still really high, but so far it has climbed ahead of the S&P 500 by 2% in a week!

I plan to setup my own graph to track the results using the portfolio tracker on this site, but the way I originally wrote it doesn’t scale particularly well so I need to make some modifications first. Until then, the CAPS account provides a pretty handy way to create a public record of the predictions and their performance over time.

As I get more data I also want to setup a secondary portfolio that uses the same idea but instead focuses on the rate of change of the difference in the two ratios rather than just the difference itself. This may be a better indicator of the undervalued stocks I’m looking for since recent ratings are probably more reflective of what the ’smart money’ is thinking now as opposed to what they thought cumulatively starting from a year or so ago. If it sees that a bunch of ’smart money’ recommendations are coming to a particular stock and the ‘average money’ recommendations are diminishing or not moving for that same stock, it could signify a major development that the market as a whole (and therefor the price of the stock) hasn’t yet taken into account which would be an optimal situation to take advantage of.

Just as a note, if in the rare event that this method does end up producing extraordinary market-beating returns I intend to purge all record of it so it remains effective and I can exploit it myself to become a gazillionaire!

Similar Posts:

• A Quick Update On CAPS And The Website
• A CAPS Hypothesis - Mathematical Outperformance Using Collective Intelligence
• It Wasn’t Good Enough: Landec
• “Dollar Cost Averaging” is Economic Silliness
• Hello!
• Finance Site Roundup
• How The Stock Market Really Works
• E*Trade Redefines ‘Partial’
• Insured Probabilities
• Rally Review: My Portfolio - part 2