I Improved TJR’s Trading Strategy. Does It Beat the Market?

Analyzing Trading Strategies: From Theory to Practice

Introduction to Trading Strategies

• YouTube trading strategies often appear flawless on charts, showcasing perfect examples that seem unbeatable. However, the reality may differ when analyzing extensive data over 10 years.

• Previous videos involved coding and testing various versions of TJR's strategy across different forex pairs and time frames, revealing mixed results.

Performance Insights from Backtesting

• The initial backtest indicated that the simpler version of the strategy (V1) outperformed a more complex version (V2), which included additional elements like fair value gaps and order blocks.

• A key finding was that V1 performed better in directional markets with strong trends rather than in choppy or range-bound conditions. This aligns with TJR's focus on indices like S&P 500 and NASDAQ futures.

Testing Different Assets

• Data for gold, silver, Bitcoin, S&P 500, and NASDAQ was downloaded to assess performance under varying conditions. The decision was made whether to apply session filters based on asset characteristics (e.g., Bitcoin trades 24/7).

• Gold showed promising results on a five-minute time frame without session filters due to its volatile nature; however, silver consistently underperformed across all tests and was subsequently dropped from consideration.

Time Frame Selection for Indices

• For Bitcoin, an hourly time frame was selected as it provided optimal results while minimizing the impact of spreads and commissions compared to lower time frames. Indices benefited from session filters due to concentrated trading volume during main sessions.

• Ultimately, a 15-minute time frame with session filters was chosen for both S&P 500 and NASDAQ based on stability across tests despite higher returns seen in shorter intervals.

Evaluating Risk vs Reward Metrics

• The analysis revealed that while raw returns might suggest one asset performed better than another (e.g., pound dollar vs Bitcoin), drawdown metrics were crucial for understanding true risk-adjusted performance. Thus, return divided by drawdown became a key metric for evaluation.

Refining Strategy Through Exit Structures

Adjusting Take-Profit Mechanisms

• A significant discovery from earlier tests indicated low hit rates for take profit levels; thus, exploring alternative exit structures became essential for improving overall strategy performance. Trailing stops emerged as the most effective method after testing multiple exit strategies across assets.

Comparison of Exit Structures

• Results showed improvements in both raw returns and risk-adjusted returns when using trailing stops instead of fixed take-profit targets; this highlighted the importance of exit strategies alongside entry points in trading success. Most assets improved except pound dollar which required a different approach due to its non-trending behavior.

Exploring Advanced Techniques: Machine Learning Integration

Implementing Machine Learning Models

• To enhance decision-making within the existing strategy framework without creating new setups, machine learning techniques were introduced to evaluate trade quality based on historical data patterns rather than relying solely on human intuition or assumptions about market behavior.

Decision Trees & Random Forest Approach

• Utilizing decision trees allowed the model to analyze various factors influencing trade outcomes through yes/no questions related to volatility, market trends, etc., leading towards informed decisions about potential trades based on past successes or failures observed in training data sets.

Final Strategy Evaluation: Long Only Approach

Long Setups Focused Strategy

• After thorough testing with random forests revealed long setups significantly outperformed shorts particularly within indices like S&P 500 and NASDAQ; adjustments were made accordingly leading towards enhanced profitability metrics while reducing overall trade frequency needed.

Conclusion: Building a Robust Trading Strategy

Comprehensive Performance Review

• Over ten years analyzed yielded impressive compounded annual growth rates despite accounting for trading costs such as spreads/commissions resulting ultimately into sustainable profits achievable through disciplined execution.

• Each asset contributed positively post-cost deductions indicating balanced diversification within portfolio management practices ensuring resilience against market fluctuations over extended periods.