The Problem with Rating Stocks

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts
  • billyjoe
    Senior Member
    • Nov 2003
    • 9014

    The Problem with Rating Stocks

    Have you ever bought a stock seemingly loved by all, a can't miss winner that proceeded to tank right after your purchase? It's probably happened to us all more than once. My theory is these stocks are rated on past performance. At first they fly under the radar slowly building steam. They start to get some press, still nothing outstanding that makes them a must buy. Then the onslaught begins, another Google, Amazon, Netflix, get in on the ground floor before it's too late. By the time you read the last comment it is too late....for you. Wouldn't you be better off buying #500 on the way up than #1 on the way down?

    Those of you familiar with Vectorvest know they rate stocks on a scale of 0-2.0 with 2.0 being perfect. Anything 1.0 or higher is usually a buy and better than the average stock. Their combo proprietary rating is called the VST or Value, Safety, Timing combined. Very few stocks get a rating above 1.50 and never have I seen a 1.7 VST rating. Currently of 8234 stocks they rate only 1 is above 1.50 with only 19 above 1.40.

    Narrowing down the field to a reasonable number to consider purchasing they have a category called "Midas Touch Stocks". Currently numbering around 700 stocks, these are graphed to show a 10 day MA stop (proprietary) that is above a 65 day MA stop. Also they must have a 5 day MA RT that is higher than a 40 day MA RT showing the stock isn't weakening in price trend.

    To further narrow the field I added 3 month EPS and 3 month Sales Growth trend parameters. All must be trending to new highs. As of March 28th these fields lowered the number of stocks to 59 possibilities. I check their progress each month.

    The markets weren't so hot from March 28th to April 28th with lots of volatility the S&P 500 and the DJIA were almost unchanged during that period but the NASDAQ was down about 2%. To narrow the search further I screened for only those of the 59 stocks that were > than 98% of their price March 28th. 59% of the 59 stocks passed the test. Not bad but the % can be increased.

    I noticed that stocks priced under $7 /share didn't hold up. Eliminate the under $7 stocks and the success rate improves to 66%.
    Also the stocks with the highest VST scores, seemingly the best stocks, didn't do as well as those with lower scores to begin with. Eliminate the top 50% of stocks, those rated 1.12 to 1.39. The success rate climbs to 74%.

    I'll keep tracking the original 59 stocks month by month to see if they continue behaving the same. A big upward or downward trend would change everything. Also I've got a new batch of stocks starting Monday, 33 of them, that are all over $7, and rated from 1.03 to 1.12. All the parameters on the graphs are based on 3 month time periods as to RT,MA, EPS, Sales, etc.

    Any questions feel free to ask.


    ------------------------billy
  • mrmarket
    Administrator
    • Sep 2003
    • 5971

    #2
    Originally posted by billyjoe View Post
    Have you ever bought a stock seemingly loved by all, a can't miss winner that proceeded to tank right after your purchase? It's probably happened to us all more than once. My theory is these stocks are rated on past performance. At first they fly under the radar slowly building steam. They start to get some press, still nothing outstanding that makes them a must buy. Then the onslaught begins, another Google, Amazon, Netflix, get in on the ground floor before it's too late. By the time you read the last comment it is too late....for you. Wouldn't you be better off buying #500 on the way up than #1 on the way down?

    Those of you familiar with Vectorvest know they rate stocks on a scale of 0-2.0 with 2.0 being perfect. Anything 1.0 or higher is usually a buy and better than the average stock. Their combo proprietary rating is called the VST or Value, Safety, Timing combined. Very few stocks get a rating above 1.50 and never have I seen a 1.7 VST rating. Currently of 8234 stocks they rate only 1 is above 1.50 with only 19 above 1.40.

    Narrowing down the field to a reasonable number to consider purchasing they have a category called "Midas Touch Stocks". Currently numbering around 700 stocks, these are graphed to show a 10 day MA stop (proprietary) that is above a 65 day MA stop. Also they must have a 5 day MA RT that is higher than a 40 day MA RT showing the stock isn't weakening in price trend.

    To further narrow the field I added 3 month EPS and 3 month Sales Growth trend parameters. All must be trending to new highs. As of March 28th these fields lowered the number of stocks to 59 possibilities. I check their progress each month.

    The markets weren't so hot from March 28th to April 28th with lots of volatility the S&P 500 and the DJIA were almost unchanged during that period but the NASDAQ was down about 2%. To narrow the search further I screened for only those of the 59 stocks that were > than 98% of their price March 28th. 59% of the 59 stocks passed the test. Not bad but the % can be increased.

    I noticed that stocks priced under $7 /share didn't hold up. Eliminate the under $7 stocks and the success rate improves to 66%.
    Also the stocks with the highest VST scores, seemingly the best stocks, didn't do as well as those with lower scores to begin with. Eliminate the top 50% of stocks, those rated 1.12 to 1.39. The success rate climbs to 74%.

    I'll keep tracking the original 59 stocks month by month to see if they continue behaving the same. A big upward or downward trend would change everything. Also I've got a new batch of stocks starting Monday, 33 of them, that are all over $7, and rated from 1.03 to 1.12. All the parameters on the graphs are based on 3 month time periods as to RT,MA, EPS, Sales, etc.

    Any questions feel free to ask.


    ------------------------billy

    Backtesting is a great way to enable the odds to always be in your favor when stockpicking.
    =============================

    I am HUGE! Bring me your finest meats and cheeses.

    - $$$MR. MARKET$$$

    Comment

    • mimo_100
      Senior Member
      • Sep 2003
      • 1784

      #3
      Originally posted by billyjoe View Post
      Any questions feel free to ask.


      ------------------------billy
      Billy, is RT = Relative Trend????? If so, what is the definition?

      Tim
      Tim - Retired Problem Solver

      Comment

      • Louetta
        Senior Member
        • Oct 2003
        • 2331

        #4
        Originally posted by mrmarket View Post
        Backtesting is a great way to enable the odds to always be in your favor when stockpicking.
        Truer words were never spoken, Large Person. But what you said bears no relation to what he did. Given any (reasonably large) set of data, I can find a sub-set of that data which satisfies any criteria. Especially if I get to make up the sub-set as I go along.

        Comment

        • billyjoe
          Senior Member
          • Nov 2003
          • 9014

          #5
          Originally posted by mimo_100 View Post
          Billy, is RT = Relative Trend????? If so, what is the definition?

          Tim
          Mimo, In this case it is the trend of the Relative Timing score. Also proprietary to Vectorvest and on a 0-2.0 scale.

          --------------------billy

          Comment

          • billyjoe
            Senior Member
            • Nov 2003
            • 9014

            #6
            Originally posted by Louetta View Post
            Truer words were never spoken, Large Person. But what you said bears no relation to what he did. Given any (reasonably large) set of data, I can find a sub-set of that data which satisfies any criteria. Especially if I get to make up the sub-set as I go along.
            Louetta, I could back test my theories if I were a whiz with setting up criteria, but I'm not. I'd have to be able to transfer what I'm looking for in line graphs to numbers and that's nearly impossible for me. A MACD cross parameter would be possible but upward trending sales growth and EPS at 3 month highs? Don't know if it is offered as an option. I'll keep my own records and that will serve as my backtest as months roll by.

            ----------------------billy

            Comment

            • IIC
              Senior Member
              • Nov 2003
              • 14938

              #7
              IMO backtesting is pretty much a waste of time unless you can factor in economic, market, sector and industry group conditions.

              Doug
              "Trade What Is Happening...Not What You Think Is Gonna Happen"

              Find Tomorrow's Winners At SharpTraders.com

              Follow Me On Twitter

              Comment

              • mrmarket
                Administrator
                • Sep 2003
                • 5971

                #8
                Originally posted by IIC View Post
                IMO backtesting is pretty much a waste of time unless you can factor in economic, market, sector and industry group conditions.

                Doug
                I disagree....if you ignore market timing, then you are always making a relative choice among securities. If that's the case, then the choices are a usually a function of market psychology which is based on human behavior which never really changes, except at an evolutionary pace. Having said all that, backtesting will reveal this behavior.
                =============================

                I am HUGE! Bring me your finest meats and cheeses.

                - $$$MR. MARKET$$$

                Comment

                • IIC
                  Senior Member
                  • Nov 2003
                  • 14938

                  #9
                  OK...So what are the types of things you are backtesting?

                  Where do you perform these backtests?

                  Doug
                  "Trade What Is Happening...Not What You Think Is Gonna Happen"

                  Find Tomorrow's Winners At SharpTraders.com

                  Follow Me On Twitter

                  Comment

                  • mrmarket
                    Administrator
                    • Sep 2003
                    • 5971

                    #10
                    Originally posted by IIC View Post
                    OK...So what are the types of things you are backtesting?

                    Where do you perform these backtests?

                    Doug

                    Doug...it's a little too quantitative and boring for this forum.
                    =============================

                    I am HUGE! Bring me your finest meats and cheeses.

                    - $$$MR. MARKET$$$

                    Comment

                    • billyjoe
                      Senior Member
                      • Nov 2003
                      • 9014

                      #11
                      If I had the technical skills I could probably backtest to some extent , but I'm always looking for a simple solution. Too much information might scare me away from buying or selling a stock .

                      -------------------bill

                      Comment

                      Working...
                      X