In our study of Nortel, we started out with 207 shares at $48.25 each and $12.25 in cash for a total investment of $10,000. All interim high/low prices are based on the closing price each day. All buy or sell prices are based on the opening price for the day after the trade is triggered by the previous day's closing price. Analysis follows the table. Here is a key to the table's columns:

• Date: Date when Interim High or Low was reached or Date on which the Buy or Sell Trigger was tripped. The actual trade is recorded at the following trading day's opening price. That date could be the following day or a few days later if the Triggering Date fell on a holiday or Friday.
• Interim High/Low: Interim High Prices are indicated in blue. Interim Lows in red.
• 10% Trigger Price: The price that is 10% above the Interim Low or 10% below the Interim High.
• Actual Trigger Price: The actual trigger price usually differs from the 10% price because stocks rarely close at a price exactly 10% below or above the interim high/low.
• Opening Shares: Number of shares owned before a trade.
• Next Day Open: Our trading price - the opening price on the trading day after our buy/sell signal is triggered.
• Closing Shares: Number of shares held after a trade.
• Value: Value of the shares held after a trade. This is a large amount when we are in shares and zero when we are in cash.
• Cash: The value of the cash held after a trade. This is a small amount when we are in shares, and a large amount when we are 100% in cash.

At the end of 2001, we were left with 330 shares of Nortel worth $3,934.28.  This compares with a buy and hold amount of  $2463.30.  While a person who had $10,000 of Nortel at the beginning of the year would have lost 75.4%, an active trader per our formula would have lost only 60.7%. Subtracting our transaction costs - $27 times 26 trades, we have a final total of  $3232.28 for a loss of 67.7%. That may not be statistically significant. While it's gratifying to know our system beat buy and hold for Nortel in 2001, somehow it's less than gratifying to know it still lost that much.

The table is fraught with anomalies - many instances of the stock trading down almost exactly 10% or up almost exactly 10% before changing direction. Some modification - some exceptions to our formula - might produce better results. Perhaps something related to trading volume, or follow through on a trend, or waiting three days for confirmation of a move before committing to a trade. It's possible that any rules we generate to fix Nortel may not have validity for other companies or situations. But I think that the best rule is to avoid stocks of companies that are losing money in the first place, even if our example of Loewen Group rendered excellent results.

Case Study # 1 - Barrick Gold
Case Study # 2 - Microsoft
Case Study # 3 - Tecsys Inc.
Case Study # 4 - Loewen Group