Position Sizer

Position Sizer Skill

Advanced position sizing using multiple methodologies to optimize risk-adjusted returns.

Overview

This skill provides:

• Fixed percentage method (simple risk-based sizing)
• Volatility-adjusted sizing (normalizes risk across assets)
• Kelly Criterion (optimal growth rate calculation)
• ATR-based sizing (Average True Range volatility)
• Portfolio heat management (total risk exposure)
• Dynamic position adjustments based on volatility changes

Position Sizing Methods

1. Fixed Percentage Method

Risk a fixed % of account per trade (e.g., 1-2%).

Formula: Position Size = (Account Value × Risk %) ÷ (Entry Price - Stop Price)

Best For:

• Consistent risk management
• Beginning traders
• Low volatility markets

2. Volatility-Adjusted Method

Adjusts size based on asset volatility.

Formula: Adjusted Size = Base Size × (Target Vol ÷ Asset Vol)

Best For:

• Multi-asset portfolios
• Variable volatility regimes
• Professional risk management

3. Kelly Criterion

Maximizes long-term growth rate based on edge.

Formula: Kelly % = (Win Rate × Avg Win/Loss Ratio - (1 - Win Rate)) ÷ Avg Win/Loss Ratio

Best For:

• Systems with known edge
• Experienced traders
• Always use fractional Kelly (25-50%)

4. ATR-Based Method

Uses Average True Range for volatility assessment.

Formula: Position Size = (Account × Risk %) ÷ (ATR × Multiplier)

Best For:

• Trend-following strategies
• Volatile markets
• Technical traders

Tools

1. calculate_position_size

Calculates optimal position size for a trade.

Parameters:

• symbol (required): Trading symbol
• account_value (required): Current account value
• risk_per_trade_pct (optional): Risk per trade % (default: 1.0)
• method (optional): Sizing method ("fixed_pct", "volatility_adjusted", "kelly", "atr", default: "volatility_adjusted")
• current_price (optional): Current market price
• stop_loss_price (optional): Planned stop loss price
• win_rate (optional): Historical win rate (for Kelly, default: 0.55)
• avg_win_loss_ratio (optional): Average win/loss ratio (for Kelly, default: 1.5)

Returns:

{
  "success": true,
  "symbol": "AAPL",
  "recommendations": {
    "primary_method": {
      "method": "volatility_adjusted",
      "position_size_dollars": 5420.00,
      "position_size_shares": 35,
      "rationale": "Adjusted for 18.5% annualized volatility"
    },
    "alternative_methods": {
      "fixed_percentage": {
        "position_size_dollars": 5000.00,
        "position_size_shares": 32
      },
      "kelly_criterion": {
        "position_size_dollars": 6250.00,
        "position_size_shares": 40,
        "kelly_fraction": 0.25
      },
      "atr_based": {
        "position_size_dollars": 5100.00,
        "position_size_shares": 33
      }
    }
  },
  "risk_metrics": {
    "dollar_risk": 500.00,
    "risk_pct": 1.0,
    "position_value_pct": 5.42,
    "estimated_volatility": 0.185,
    "max_loss_at_stop": 500.00
  },
  "constraints": {
    "max_position_size_dollars": 10000.00,
    "max_position_size_pct": 10.0,
    "min_position_size_dollars": 100.00,
    "constrained": false
  },
  "validation": {
    "within_risk_limits": true,
    "sufficient_buying_power": true,
    "liquidity_adequate": true
  }
}

Usage:

python scripts/position_sizer.py calculate_position_size \
    --symbol AAPL \
    --account-value 100000 \
    --risk-per-trade-pct 1.0 \
    --method volatility_adjusted \
    --current-price 155.00 \
    --stop-loss-price 150.00

2. calculate_portfolio_heat

Calculates total risk exposure across all positions.

Parameters:

• account_value (required): Current account value
• positions (required): Array of current open positions
• pending_trades (optional): Array of trades being considered

Returns:

{
  "success": true,
  "portfolio_heat": {
    "total_risk_dollars": 2500.00,
    "total_risk_pct": 2.5,
    "individual_positions": [
      {
        "symbol": "AAPL",
        "position_value": 5000.00,
        "risk_dollars": 500.00,
        "risk_pct": 0.5,
        "stop_loss": 148.50
      }
    ],
    "risk_distribution": {
      "tech_sector": 1.2,
      "finance_sector": 0.8,
      "healthcare_sector": 0.5
    },
    "capacity": {
      "max_total_risk_pct": 5.0,
      "remaining_capacity_pct": 2.5,
      "remaining_capacity_dollars": 2500.00
    }
  },
  "recommendations": {
    "can_add_position": true,
    "max_new_position_dollars": 1000.00,
    "warnings": []
  }
}

3. adjust_position_for_volatility

Adjusts existing position size based on volatility changes.

Parameters:

• symbol (required): Trading symbol
• current_position_value (required): Current position value
• target_volatility (optional): Target volatility % (default: 20.0)
• rebalance_threshold (optional): Rebalance if exceeds threshold (default: 0.15)

Returns:

{
  "success": true,
  "symbol": "AAPL",
  "analysis": {
    "current_position_value": 5000.00,
    "current_volatility": 0.28,
    "target_volatility": 0.20,
    "volatility_ratio": 1.40
  },
  "recommendation": {
    "action": "reduce",
    "target_position_value": 3571.00,
    "adjustment_amount": -1429.00,
    "adjustment_shares": -9,
    "rationale": "Current volatility 40% above target"
  },
  "execution_plan": {
    "recommended": true,
    "urgency": "medium",
    "method": "market_order"
  }
}

4. calculate_kelly_fraction

Calculates Kelly Criterion for position sizing.

Parameters:

• win_rate (required): Probability of winning (0-1)
• avg_win_loss_ratio (required): Average win ÷ average loss
• kelly_multiplier (optional): Conservative multiplier (default: 0.25)

Returns:

{
  "success": true,
  "kelly_calculation": {
    "raw_kelly_pct": 25.5,
    "adjusted_kelly_pct": 6.375,
    "kelly_multiplier": 0.25,
    "inputs": {
      "win_rate": 0.55,
      "avg_win_loss_ratio": 1.8
    },
    "formula": "(win_rate * avg_win_loss_ratio - (1 - win_rate)) / avg_win_loss_ratio"
  },
  "recommendation": {
    "position_size_pct": 6.375,
    "rationale": "Using 25% Kelly for conservative approach",
    "warnings": [
      "Full Kelly (25.5%) is aggressive - using fractional Kelly"
    ]
  }
}

Safety Constraints

Hard Limits

• Max Single Position: 10% of account value (configurable)
• Max Total Risk: 5% of account value
• Min Position Size: $100 (avoid excessive trading costs)
• Max Leverage: 2x (if using margin)

Dynamic Adjustments

• Reduce size after losing streaks
• Increase size cautiously after winning streaks
• Scale down in high volatility
• Respect circuit breakers

Integration with Risk Manager

This skill wraps and extends the existing src/core/risk_manager.py:

from claude_skills import load_skill

position_skill = load_skill("position_sizer")

# Calculate position size for new trade
position = position_skill.calculate_position_size(
    symbol="AAPL",
    account_value=100000,
    risk_per_trade_pct=1.0,
    method="volatility_adjusted",
    current_price=155.00,
    stop_loss_price=150.00
)

# Check portfolio capacity before adding
heat = position_skill.calculate_portfolio_heat(
    account_value=100000,
    positions=current_positions,
    pending_trades=[position]
)

if heat["recommendations"]["can_add_position"]:
    execute_trade(position)

Usage Example

from claude_skills import load_skill

position_skill = load_skill("position_sizer")

# Calculate position size for new trade
position = position_skill.calculate_position_size(
    symbol="AAPL",
    account_value=100000,
    risk_per_trade_pct=1.0,
    method="volatility_adjusted",
    current_price=155.00,
    stop_loss_price=150.00
)

# Check portfolio capacity
heat = position_skill.calculate_portfolio_heat(
    account_value=100000,
    positions=current_positions
)

if heat["recommendations"]["can_add_position"]:
    execute_trade(position)

CLI Usage

# Calculate position size
python scripts/position_sizer.py calculate_position_size \
    --symbol AAPL --account-value 100000 --risk-per-trade-pct 1.0

# Calculate portfolio heat
python scripts/position_sizer.py calculate_portfolio_heat \
    --account-value 100000 --positions-file positions.json

# Calculate Kelly fraction
python scripts/position_sizer.py calculate_kelly_fraction \
    --win-rate 0.55 --avg-win-loss-ratio 1.8