The purpose of my Trader Paul account is to generate cash for discretionary spending, such as vacations or major purchases. I think of it as "play money" rather than an investment for the long-range accumulation of capital or for regular income. As such, I manage it much less conservatively than I do the funds that comprise the bedrock foundaton of my financial security. For example, asset allocation is not a consideration in this account, whereas it is a huge factor in my long-term investment accounts.


The essence of my plan is to realize capital gains from short- to mid-term price movement. I may hold a position for as short as one day or as long as several months. I think of this as "churn and earn." For example, if I can do just four trades a month yielding at least $250 each, that provides about $1000 of discretionary funds.

Success of the strategy is measured by accumulating spendable dollars rather than achieving any specific rate of return. I do track the change in portfolio value month to month, adjusting for any cash withdrawals.

Is not. What I do is not "day trading," which has a specific technical definition. A day trade involves buying and selling (or selling and buying) the same stock during the same trading day. Making four or more day trades within a rolling five business-day period defines a pattern day trading account which is then subject to special rules and tax provisions.

In practice

Price range: Take positions in stocks priced from $5 to $15 per share. Stocks priced less than $5 are generally not marginable. Stocks above $15 have to be bought in proportionately smaller lots to avoid making a large bet on a single stock.

Lot size: Buy in lots of 500. That is enough shares to generate a nice capital gain from a reasonable price move without trading costs eating up the profit. For example, if 500 shares gain 50¢ each, that's $250 that can be banked. I used to buy in lots of 1000, but over time I found that involved taking too much risk on a single stock.

Screening: Identify candidate stocks by screening for ones reversing a downtrend (for buys) or an uptrend (for short sells).

TD Ameritrade provides a tool, Advanced Analyzer, that makes this easy. I use built-in screens:

• Potential bullish reversals (buy candidates)
Trying to reverse long downward trend: Stocks near the middle of their 52-wk trend that have moved up at least 5% over the past five days
Below 50-day MA; crossing up 10-day MA with volume: Stocks below their 50-day moving average but are crossing up through their 10-day moving average on 150% more than average volume over the past six months
• Potential bearish reversals (short candidates)
Exhaustion (New high but too tired): Stocks that gap up at least 1% but close lower than they open with MACD in bullish state. These stocks are generally in an uptrend but seem to want to reverse-at least for the short term
Topping out?: Stocks making a new 52wk high but close lower than they open, suggesting a possible trend reversal
Peaks: Stocks making a one-day move up in price of 10% or more

Watch list: Candidates emerging from the screen are added to a watch list for charting and technical analysis.

For charting, I subscribe to StockCharts Extra! service. It allows me to maintain multiple watch lists and define custom chart layouts. My default chart layout is a one-year daily chart with the 50-day moving average superimposed on the chart and several indicators plotted below the chart: volume, price oscillator (PPO), Wilder's DMI (ADX), Chaikin money flow (CMF), and relative strength (RSI).

I annotate the chart by drawing obvious trend lines and noting significant company events, especially earnings releases. For example, here is the current chart for Ariba (ARBA). The stock got a big boost from the earnings report in January, and the price is moving upward in a pretty clear channel (dashed lines).

For technical analysis, I subscribe to MarketEdge Second Opinion Daily service (through Ameritrade). It helps me interpret the chart and the recommendations are another data point. The current report for Ariba (PDF) calls it a "strong buy" in a "strong upward trend."

Taking positions: Buys should meet the following characteristics:

  • Averages — rising one-month moving average; the 50-MA should be moving up or about to turn up
  • Price momentum (PPO) — rising steadily for a least the previous four weeks
  • Trend (ADX) — strengthening, and buying pressure (+DI) is widening the gap above selling pressure (-DI)
  • Relative strength (RSI) — neutral or above and rising
  • Volume — average volume should be rising, but if not, upward price moves at least should be on noticeably heavier than normal volume
  • Money flow (CMF) — positive inflow, but if negative, the outflow should be slowing noticeably

In general, the indicators should themselves show an overall trendline (up for longs, down for shorts).

For selling short the indicators should be in the opposite condition.

Timing: The best timed buy will be just as the stock is bouncing off the uptrend line or has sufficient headroom below any resistance line for a reasonable gain. (For example, Ariba's price is about 75¢ from the upper channel line, room for a modest gain). Avoid buying just before earnings are reported. If holding a position when earnings are reported, set a very tight stop to protect gains or prevent further losses.

Closing positions: This is the hard part. In the past I've had a tendency to hang on too long or get shaken out too soon.

  • Stops. For long positions, set an initial stop-loss order to trigger if the ask price falls 8% or more below my buy price. For short positions, the stop-loss order triggers if the bid price rises 8% or more above my short-sell price. Defining the stops this way reduces the chances of getting shaken out by an outlying trade.
  • Sell/Cover. Close a position when a meaningful gain can be realized and the chart begins to deteriorate, such as:
    • Gap down with heavy volume (or gap up for short positions)
    • Buying pressure (+DI) and selling pressure (-DI) lines cross
    • Price has followed the Bollinger band for 10 days or more (upper band for longs, lower band for shorts)
    • Price breaks a trend line and fails to recover the next day