Premium Market Analysis, Trader Education, Software, and Trading Strategies

Trading Strategies

Equity Long-Short Performance in 2019

Performance of long-short strategies for Dow 30 stocks in daily and weekly timeframe. The strategies are based on features developed by DLPAL LS software.

The equity long-short strategies described in this article are based on features generated by DLPAL LS software. The features are the result of identification of price action anomalies in daily and weekly stock data. The strategies use two of the generated features: Pdelta and S, which stand for the directional bias and its significance, respectively.

Examples of how the features are generated by DLPAL LS can be found in this article. DLPAL LS generates various types of features for individual but also for groups of securities but the two relevant for the long-short equity strategies in this article are the directional bias Pdelta and its significance S.

Note that long-short strategies exhibit convex returns with respect to long-only strategies and may provide a hedge in case of a bear market (left tail hedge.) However, long-short strategies also trend to underperform buy and hold during bull markets.

Long-Short Strategies for Dow 30 stocks

Two types of strategies are considered in this article.

A) Strat

Strat has the following general form:

Buy/Cover if Pdelta × S > 0
Short/Sell  if Pdelta × S < 0

Maximum allowed positions can all be long or short at times. Therefore, this strategy attempts to take advantage of the market trend. The maximum number of open positions is set in advance and if there are ties the rank/score function Pdelta × S is used to resolve them. There is a requirement for a minimum of two days holding period.

Maximum open positions
Daily strategy:       6
Weekly strategy: 6

B) Rank

Securities are ranked according to Pdelta × S . There is a maximum number set for open positions and for long and short positions separately. Longs are marked top stocks with positive rank and shorts are marked bottom stocks with negative rank. There is no minimum holding period.

Maximum open positions
Daily strategy:    30,  max. long: 15, max.  short: 15
Weekly strategy: 6,  max.  long:    3, max. short:    3

Backtest details

Time-frame: Daily and weekly (adjusted data)
Universe: Dow stocks (DOW not included due to limited history and DWDP data ended 05/31/2019)
Backtest period: 01/02/2019 – 12/31/2019
History start date considered for feature calculations: 01/03/2000
Commission: $0.05/share

Performance based on backtests

Parameter Daily Strat Daily Rank Weekly Strat Weekly Rank
CAGR 23.7% 2.3% 16.9% 4.4%
Max. DD -6.0% -4.3% -6.0% -4.6%
Sharpe 2.1 0.3 0.62 0.4
Trades 202 2739 90 586
Long trades 170 1378 64 311
Short trades 32 1361 26 275
Win rate 60.4% 50.1% 55.5% 51.2%


  • Rank strategies tend to underperform most in bull markets because there is an upper limit on the number of long positions and they cannot take advantage of the uptrend. This is the main reason for 2019 low performance.
  • The performance of Strat strategies was satisfactory with 19.6% for daily and 16.9% for weekly.
  • Drawdown for all strategies is small and less than 6%. Leverage can apply.
  • Daily rank performance is affected most by commissions due to large number of trades.
  • Combination of Strat and Rank may be a better alternative.
  • Strat and Rank are basic strategies and there is room for improvement using more features and an added layer of machine learning.

More details about DLPAL LS can be found here. For more articles about DLPAL LS click here.

If you have any questions or comments, happy to connect on Twitter: @priceactionlab

Strategy performance results are hypothetical. Please read the Disclaimer and Terms and Conditions.