I have used TC2000 off and on for the past 3 years, and like most users, have tinkered with various combinations of PCF's to attempt to acheive "the Holy Grail" (i.e., a perfect stock picking system). Unfortunately, this software is only good at presenting viable candidates on an intermittent basis. You, the trader, still have to pull the trigger on which stock to trade, and live (or die) with the results. An interesting set of PCF's that I use to determine when to be in and get out of the market came out of my infatuation with moving averages and RSI (Relative Strength Index). My first step on the last day of each month is to choose the 800-1000 stocks that I will use for the next trading month. I like solid stocks, so my three criteria to select these stocks are: (1) $15 or higher stock price, (2) trading 200,000 shares per day for the previous 60 days, and (3) stocks that are 5+ years old. The number of stocks vary from month to month, but are in the 800 to 1000 range. The second step is to create watch lists for each of the 31 TC2000 sectors (Aerospace/Defense, Automotive, Banking, etc..) and add the selected stocks from the first step to each of the sector watch lists. This step takes about 10 minutes, using the mark and copy process. The third step is to utilize the 7 PCF's that I have created. These PCF's are (1) Current Price above 5 Day MA, (2) Current Price above 10 Day MA, (3) Current Price above 20 Day MA, (4) Current Price above 50 Day MA, (5) New High Today, (6) Current Price above all of the above 4 MA's, and (7) RSI at 50 or above AND RSI higher that RSI 10 trading days ago. Plotting these PCF's on 7 columns (side-by-side) gives me an instantaneous look into all of the stocks within a particular sector. I also plot the previous 4 days PCF's on other tables, so I can flip from 1 day to the next, looking for trends in stocks, industries, and sectors. The fourth step (utilizing Microsoft Excel spreadsheet) is to plot the total number for each sector's stocks, as well as the total number of "TRUE's" for each of the above 7 PCF's. I have a complicated weighted formula that I use to derive a TOTAL PCT score on a daily basis for each sector. I also tabulate all scores for all PCF's to come up with a daily TOTAL PCT for the overall MARKET. Using another Excel spreadsheet to copy the TOTAL PCT's for each sector, I can see each sectors daily score side-by-side, as well as the overall market, too. I also color code the scores, ranging from RED (7.49% or under), MAROON (7.5% to 9.99%), PURPLE (10.0 to 12.49%), YELLOW (12.5 to 14.99%) and GREEN (15.0% and above). I then take the above statistics and plot them on a daily basis using today's and the last two trading days stats to form an expotential pct for each sector each day. This helps to smooth out the day-to-day choppiness of the market. Using the colors schematic, you can see the elephant-like steps of the market as each rally occurs, and where the momentum of the ball tossed in the air starts to slow down and eventually fall at the end of each rally. This system allows me to be in cash when the market falters (such as Jan 16, 2003 to current date), and to be ready with my stock choices from the sectors that always lead the next rally. The cool thing about this system is that a lot of what William O'Neill says is true, in that the sectors and industries that lead a rally get there first and stay up longer through the end of the rally (i.e., they are the market leaders, or the first tier of any market rally). The first tier stocks are always the ones that gain anywhere from 5-20% on a rally (sometimes more, if you catch the right stock in the right wind). The second tier stocks are usually 2-3 trading days behind the first tier, and they don't stay around as long as the first tier stocks. Their gains are more modest (around 3-10% tops). The second tier stocks have their own steam under them, but also gain from the ground laid by the first tier stocks, and are pulled along for the ride. The third tier stocks come in on the last days of the rally, and the results from those stocks are barely noticible. They are, to use an O'Neillism, the laggards of their industries, and should be sniffed at and disgarded as dogs beneath contemp. My key is to wait until a sector reaches the MAROON stage to see which stocks are poised to breakout, utilizing the 4 MACD strategy esposed in the WORDEN Archives. When the sector reaches the PURPLE stage, I am ready to make a trade. As always, I seek confirmation through other technical indicators. I then trade the stock based on the S&P 500 Daytrade System, except I use a sliding stop loss which keeps me in the black on a longer basis (I try to get 5% out of each trade, but no worst that a 2-3% loss). I have been successful on 75% of my trades, which keeps my wife off my back. As choppy as this market has been for these past few years, you need to know when to play and when to hold them. Since I am adverse to SHORTING stocks (something seems un-American about it), I utilize the above system to keep me honest. Would like your input on the above system, as well as any system that you use in coordination with TC2000. Thanks, Bob
Interesting concept. I'd love to share more of your knowledge. I am developing a TCNet Trading and chat room and I think your concept might prove to be an interesting discussion. Take a moment and PM me and let's talk. As a software trainer, I am always looking into new innovative concepts and theories.
It is not very complicated once the excel spreadsheet has been set up. I just copy a generic worksheet within the same file, change the dates at the top of the screen to match the trading dates of the coming month, and plug in the number of stocks for each of the 31 TC2000 sectors. Each day around 7:30 p.m. EST, I download the data from TC2000 (3-5 minutes), run my scan for the PCF's for just the 800-1000 stocks for a watch list called February 2003 (4-5 minutes), then pull up each of the 31 watch lists that I created for this process (such as Bob's Aerospace/Defense Watch List), which currently has 9 stocks in it (based on the three criteria set up to select the stocks I watch on a monthly basis). I already have set up the tabs and the PCF's that I want side-by-side to each other for the 7 PCF's mentioned earlier. Click on each column header, brings up all the "True's" to the top, count the number for that particular PCF for that watch list, plug it into the excel spreadhseet. This process takes 20-30 minutes each day (easier when the market is in the toilet like it has been for 4 weeks, when there are less stocks meeting the criteria of the PCF's). The spreadsheet does the rest, giving me a total pct score for each sector to the right of the 7 PCF columns, and the overall market score is automatically calculated at the bottom of the spreadsheet. I then take the total pc scores for all 31 sectors and overall market, copy, then do a paste special (selecting only values) to another spreadsheet that lists all the daily total percentages for each sector and overall market. This way, I can see what each sector does day after day in one easy scan of the spreadsheet. More formulas are used to smooth out the day-to-day noise that a bouncing market can bring to the table. My concept for these formulas is to use a 30-30-40 mix (stocks-sectors-overall market), which helps to keep me out of a stock when the market is tanking, but let's me know who the 1st tier sectors, industries, and stocks are when any combination of the three entities start to heat up. In this way, the William O'Neill training comes out regarding his formula for knowing when a rally begins and when a rally starts to peter out. The final daily formula is done to take the current and previous two days percentages from the above calculations (formula based on expotenial moving average - done again to smooth out day-to-day bumps, sags, or surges). The bottom line is that I have already created the spreadsheet and the formulas prior to plugging in the daily numbers. Within 40-45 minutes, I know what I need to know about what happened in the market that day (without looking at CNBC or any other source) based on how the percentages went up or down. The best part about all of this is that it can be tweaked to individual tastes. The trick to is stay consistent. While TC2000 is tedious and cumbersome when it comes to creating and scanning PCF's for backtesting purposes, it is fairly rapid when it comes to a system like mine. I have attached the STOCK-02 file used for the final 6 months of 2002 for your perusal. In the next missive, I will attach the initial spreadsheet used to plug in the daily numbers. Thanks, Bob
Unfortunately, I could not attach the excel spreadsheet with my last reply, as it is over 1MB in size. However, if anyone would like, I can send them an email with the two spreadsheets attached for their examination. My email address is robert.lopeck@ncr.com (at work), boblopeck02@aol.com or rlopeck@yahoo.com (at home). Thanks, Bob