Beeks System -- Rules

Rules for the Beeks system (as of January 2008)

Beeks is a primarily a momentum system, which looks for stocks in an uptrend...but at the same time is a shorter term countertrend system, looking for stocks that have taken a quick breather, attempting to get back in as the stock gets its second wind and takes off on that next leg up.

Entry criteria:

• Stock must be behaving well; that is, RSI needs to be in "overbought" range at some point over the last month.
• The stock then needs to have slowed, as measured by 3 to 4 down (or very small up) closes.
• Volume condition: Average volume over last month must be over 100,000 shares, but the prior day's volume needs to be not more than 3 times the average volume over the past month.
• Volatility: this system takes higher volatility stocks as measured by the stock's average true range -- the range needs to be at least 2-2.5%.

Exit criteria:

Exit at some point the day following two consecutive up closes; or a break of the 5 day SMA. Do not hold for any longer than 5 days.

Position Sizing. There are no stops in this system; all risk management is done through position sizes -- so any one position should never be over 5% of your trading assets -- and preferably would be less.

Knish System -- Rules

Rules for the Knish system (as of December, 2007)

Knish is a mean reversion system, looking to capitalize on stocks ready for a quick rebound after being "overly stretched downward".

Entry. Knish uses two methods to determine potential trade entries:
1) A streak. You may enter the day after five consecutive down days, with the condition that the difference of yesterday's close (Bar) and the previous day's close (Bar - 1) is greater than the day before yesterday's close (Bar - 1) and the close of the day preceding (Bar - 2).
2) A band violation. A buy can be be executed if the price goes outside the difference of the highest high of the last 13 days less the lowest low of the last 13 days by at least 24% of that range (the "band").

Filters. Both entry conditions have additional prerequisites, or filters:
1) Oversold. The 11 period RSI must be "oversold" (I use an RSI value of 33 or less);
2) Trend Filter. The longer term trend must be up (Close > 200 day MA). Exception: except if the streak condition has reached seven or more down days, at which point entry is still allowed, if the streak condition in 1) is still met.
3) Band/Volatility filter. In order to make the trade worthwhile, the band needs to be at least 5/8 of 1% of the limit price (since it is the profit target).

Exits. Knish trades have multiple conditions for an exit.
1) Profit target hit. Price rises by 100% of the "band" that is calculated band violation condition in the entry criteria.
2) A two day up streak. Sell at market the following open after two up closes in a row.
3) Rate of change drops below average after 5 days held. If the trade has been held for at least 5 bars, and the one day rate of change is below the average (geometric average) rate of change for the last five days, then exit the trade.
4) Stop hit. The stop is a wide stop - 5 times the "band". It's not meant to be hit as much as it is to prevent a meltdown.
5) Time stop. A trade will not last more than 10 days.

Knish will then review candidates, favoring those with lower RSI, and stocks that that are less correlated to positions already in the portfolio. Only two positions in a security can be open at one time.

My Active Trading Systems

I have done a lot of system testing over the last year. I use Wealth-Lab Pro as my testing platform. At this point I trade two systems which I have developed using ideas I have gained from trading books. I have even given these systems names!

My Active Trading Systems


Knish. This is a mean reversion system, that attempts to take weak, recently "oversold" stocks as measured by the 14 day RSI ( or relative strength index ) that have had a streak of one week's down closes, or the lowest low over a two week period and bets on a short term swing up. It takes profits via a profit target or after two up closes. It is looking to grind out small profits.

Beeks. This is also a mean reversion system; however, this system looks for stocks that are aggressively and efficiently trending higher, and that have had a pullback over the last three to four days, looking for that burst up shortly thereafter. This system is more volatile because it does not take a profit target and exits on two up closes, but is also looking to grind out lots of small profits over time.

I will write up summaries of the rules of these in later posts.

Why Consider Active Trading?

Good question. You hear about traders -- there are "rags to riches" stories; there also stories about those who mortgaged the house, only to lose it in a day trading account?

What are my beliefs about active trading, versus buy and hold?

Consider the following scenario?

• You have been given a balance of 100,000 to invest. Let's also assume you can earn a "risk-free" rate of return (U.S. treasury bills) of around 3% (let's ignore the ravages of inflation for now because that hits any investment denominated in US Dollars). If you do nothing else, you will earn this 3% per year.
  
• You have been convinced by a friend to start actively trading a small piece of your account. You've told your friend that you are willing make one trade at a time, but you are limited to investing 10% of your portfolio (using the balance at the beginning of the year for this calculation). The other 90% you are going to keep in the T-bills.



• In this friend's "system", you will have streaks of winners and losers, but on average throughout the year you will earn on average of 1% return per trade (after commissions). Each trade will on average last a week, or five trading days.
  

For simplicity's sake, let's assume there are 240 trading days in a year.

Let's calculate the rate of return on your portfolio with hypothetical system assuming our returns are "smooth" -- that is, we earn the average on every trade, have no taxes involved (what a world!) and occur exactly as planned.

• 240 days/5 days per trade = 48 trades per year.
  


• 10% of the portfolio is in active trading...assuming a rate of 0.5% average return per trade over 48 trades = 1.0% x 48 x 10% = 4.8%



• 90% continues to be in the risk free investment, so 90% times 3% = 2.7%



• 4.8% for 10% of portfolio + 2.7% for 90% = 7.5%

So in this example, risking 10% of our portfolio, our total return has gone from 3% to 7.5%!! And we were 90% in T-bills.

Now, you say, c'mon, tman, get real...you couldn't really achieve returns like you discuss in your example. Let's discuss the concerns.

These types of returns don't occur uniformly each year -- there are some major ups and downs, just as in long-term investing. This is most certainly true -- we would have to endure some major swings with that 10%. But if over long periods of time, if we could achieve these types of returns, and our drawdowns weren't any higher than they would be being "buy and hold" investors (and let's be honest, there are some big drawdowns in "buy and hold" stock investing), we could probably live with that concern. Remember in our hypothetical account we were also holding 90% T-bills - which give us some cushion for the other 10%.

What makes you think there are systems that can achieve that rate of return? Good question. The only thing is I can say is that I have seen these types of systems and have developed and tested some of them myself. And some of them have opportunities to make even more money than 1.0% per trade.

So, this might give some insight as to why I am at least interested in pursuing the benefits of active trading. I might be willing and able with a piece of my portfolio to increase my returns and reduce my risk. If I am willing to swing for the fences, I even increase the amount I put into it. Who can resist the thought? I, for one, cannot.