Correlation Analysis Between Intraday Setups - Strategy 9: The Dangers of High Correlation in Intraday Trading

1. Setup Definition and Market Context

This article explores the important concept of correlation analysis between intraday trading setups. The primary goal is to identify diversification benefits by combining setups with low or negative correlation, thereby mitigating the risk of concentrated exposure to a single type of market edge. For instance, a portfolio of intraday strategies might include both a trend-following setup on the 5-minute chart and a mean-reversion setup. If these setups are negatively correlated, a loss in one might be offset by a gain in the other, leading to a smoother equity curve. This analysis is particularly important in the context of SPY, where market dynamics can shift rapidly.

2. Entry Rules

Entry rules must be specific and objective. For a hypothetical trend-following setup:

• Timeframe: 5-minute
• Indicator: 20-period Exponential Moving Average (EMA).
• Trigger: Price must close above the 20 EMA for a long entry, or below for a short entry.
• Confirmation: The 50-period Simple Moving Average (SMA) must be sloping in the direction of the trade.
• Volume: Trading volume must be at least 1.5x the 20-period average volume.

For a mean-reversion setup:

• Timeframe: 5-minute
• Indicator: Bollinger Bands (20, 2).
• Trigger: Price must close outside the upper Bollinger Band for a short entry, or outside the lower band for a long entry.
• Confirmation: The Relative Strength Index (RSI, 14) must show divergence.

3. Exit Rules

Exit rules are defined for both winning and losing scenarios:

• Winning Exit: The trade is closed when the profit target is reached (see section 4).
• Losing Exit: The trade is closed when the stop loss is hit (see section 5).
• Time-Based Exit: If the trade is open for more than 3 hours without hitting either the profit target or stop loss, it is closed at the market.

4. Profit Target Placement

Profit targets are determined using a combination of methods:

• R-Multiples: The primary profit target is set at a 1.5:1 reward-to-risk ratio. If the risk per trade is $100, the profit target would be $150.0.
• ATR-Based: An alternative profit target can be set at 4.5 times the 10-period Average True Range (ATR) from the entry price.
• Key Levels: Profit targets can also be placed at significant support or resistance levels, identified on a higher timeframe chart.

5. Stop Loss Placement

Stop loss placement is important for risk management:

• ATR-Based: The stop loss is placed at 2.3 times the 10-period ATR from the entry price. This adapts to market volatility.
• Structure-Based: For a long trade, the stop loss can be placed just below a recent swing low. For a short trade, it can be placed just above a recent swing high.
• Percentage-Based: A maximum stop loss of 2.2% of the instrument's price can be used as a failsafe.

6. Risk Control

Strict risk control measures are essential:

• Max Risk Per Trade: No single trade will risk more than 1.52% of the trading account.
• Daily Loss Limit: If the total net loss for the day reaches 4.4% of the account balance, all trading ceases for the day.
• Position Sizing: The position size is calculated to ensure that if the stop loss is hit, the loss will not exceed the predefined max risk per trade.

7. Money Management

Sophisticated money management techniques are employed:

• Fixed Fractional: A fixed percentage of the account (1.52%) is risked on each trade.
• Scaling In/Out: For winning trades, positions can be scaled into at predefined intervals. Partial profits can be taken at the first profit target, with the remainder of the position left to run with a trailing stop.

8. Edge Definition

The edge of this correlated setup approach is defined by:

• Statistical Advantage: By combining low-correlation setups, the overall portfolio volatility is reduced, leading to a higher Sharpe ratio.
• Win Rate Expectations: The expected win rate for the combined portfolio is approximately 54%.
• R:R Ratio: The average reward-to-risk ratio for the portfolio is targeted to be at least 1.5:1.

9. Common Mistakes and How to Avoid Them

• Over-Optimization: Avoid curve-fitting the correlation parameters to historical data. Use out-of-sample testing to validate the strategy.
• Ignoring Regime Shifts: Correlation is not static. Regularly re-evaluate the correlation matrix to adapt to changing market conditions.
• Poor Diversification: Simply adding more setups does not guarantee diversification. Ensure that the added setups have a low correlation with the existing ones.

10. Real-World Example

Let's walk through a hypothetical trade on SPY using the trend-following setup on the 5-minute chart.

• Account Size: $100,000
• Max Risk Per Trade: 1.52% ($ 1520.0)
• Entry Signal: SPY closes above the 20 EMA at a price of 4334.
• ATR (10): 12
• Stop Loss: 4830.8 (Entry Price - 2.3 * ATR)
• Profit Target: 5031.5 (Entry Price + 4.5 * ATR)
• Position Size: The position size would be calculated to ensure that a move to the stop loss results in a loss of no more than $1520.0.

This structured approach to trading, grounded in correlation analysis, provides a robust framework for navigating the complexities of the intraday markets.