Paper Trading Documentation
A documented record of paper trading (and backtesting) is invaluable for improving your ability to swing trade stocks.
This page documents the paper trading that was done on the trading plan and strategy that illustrates this site. Below are the initial results recorded using our trading log and calculator (which you can tailor to your own needs).
Paper Trading Documentation for Strategy 1-2-3 (long only)
For twelve consecutive weeks, every possible trade generated by strategy 1-2-3 (long only) was paper traded. A total of 63 trades were made.
The total amount invested was $123,822.00. At the conclusion of the paper trading period, that amount grew to $125,475.58 after commissions, for a total profit of $1,653.58, or 1.34%.
Key performance statistics are as follows:
Retesting After Paper Trading
The statistics above represent an improvement over a previous version of the trading strategy which did not include the EMA(8) entry and exit criteria.
Without this criteria, 19 additional trades (stuck-through in the table above) were taken.
A total of $161,345.05 was needed to make all 82 trades, which netted $162,812.92 after commissions, for a total profit of $1,467.87, or 0.91%.
Comparable statistics are as follows:
Other retests run at the end of the paper trading period did not improve profitability (as measured by expectancy).
This includes: 'Set the initial stop at 0.5ATR, or just below low of entry candle, whichever is higher.' The idea was to cut losses as soon as a full 5 candle swing point low fails to materialize (i.e., on the 5th day, prices go below 4th day candle low). While this did maginally shave off losses on many losing trades, it turned enough winners into losers to lower overall profitability.
Also retested and failed was to 'Go long at the end of the day (3:30) of the fifth candle of the swing point low.'
Retesting During Paper Trading
The current version of the trading strategy is also a result of improvements made to it during paper trading.
For instance, an early version of the strategy defined a swing point low using only 3 candles, with entry to be made on the 3rd candle. After four weeks, total losses far exceeded total wins. The strategy was subsequently updated with a 5 candle definition, the data re-paper traded to reflect the change, and testing moved forward using the new criteria.
A different process was used to decide how and where to set stops. Testing initially proceeded by calculating a variety of stop levels for each trade: just below the lowest low; at 50% of the distance to the lowest low; just below the retest low; at the trend line below the entry point; and at the entry point minus various multiples of ATR (i.e., 0.25, 0.33, 0.4, 0.5, 0.75, 1.0, 1.5, 2.0). Once it was decided to use ATR, the multiple was optimized to the levels currently specified in the trading strategy.
Other criteria were similarly optimized. These include the 10 and 30 time units for moving averages, the 50% retracement level, '7 days after up cross,' and '1 day or more below trend line.'
Further retests can be run against the testing data. These include:
The length criteria for the downward trend. Is 21 candles optimal? Or would lowering this number possibly generate more trades with better overall results?
It is also suspected that by supplementing entry conditions with major bullish candlestick patterns (and exit conditions with their bearish counterparts), overall profitability could be improved. At stake is how the other criteria in the trading strategy could be made to work with this supplement so that the number of tradable trades generated would not be negatively impacted. This can be analyzed and tested against the documented record as well.