Sizing API¶

Position sizing strategies.

Fixed Sizing¶

fixed_quantity¶

fixed_quantity ¶

fixed_quantity(qty: Quantity) -> Quantity

fixed_notional¶

fixed_notional ¶

fixed_notional(notional: Money, price: Price) -> Quantity

Size by notional value at a given price.

Percentage Sizing¶

percent_of_equity¶

percent_of_equity ¶

percent_of_equity(
    equity: Money, percent: Decimal, price: Price
) -> Quantity

Calculate position size as a percentage of equity.

A simple sizing method that allocates a fixed percentage of account equity to each position.

Parameters:

Name	Type	Description	Default
`equity`	`Money`	Total account equity	required
`percent`	`Decimal`	Percentage to allocate (e.g., 0.02 for 2%)	required
`price`	`Price`	Current asset price	required

Returns:

Type	Description
`Quantity`	Quantity to buy/sell

Raises:

Type	Description
`ZeroDivisionError`	If price is zero

Example

from decimal import Decimal

Allocate 2% of $10,000 equity at $100/share¶

qty = percent_of_equity(Decimal("10000"), Decimal("0.02"), Decimal("100")) float(qty) 2.0

Allocate 5% of $50,000 equity at $250/share¶

qty = percent_of_equity(Decimal("50000"), Decimal("0.05"), Decimal("250")) float(qty) 10.0

Risk-Based Sizing¶

risk_per_trade¶

risk_per_trade ¶

risk_per_trade(
    equity: Money,
    risk_fraction: Decimal,
    stop_distance: Price,
) -> Quantity

Size so that loss to stop_distance equals a fraction of equity.

Volatility Sizing¶

atr_position_size¶

atr_position_size ¶

atr_position_size(
    equity: Money,
    atr: Price,
    risk_multiple: Decimal = Decimal("1"),
) -> Quantity

Size inversely proportional to ATR (higher volatility -> smaller size).

Kelly Criterion¶

kelly_fraction¶

kelly_fraction ¶

kelly_fraction(
    win_rate: Decimal, win_loss_ratio: Decimal
) -> Decimal

Calculate the Kelly fraction of bankroll to risk per trade.

The Kelly criterion determines the optimal bet size to maximize long-term growth rate. The result can be negative (don't bet), or > 1 (use leverage).

Parameters:

Name	Type	Description	Default
`win_rate`	`Decimal`	Probability of winning (0 to 1)	required
`win_loss_ratio`	`Decimal`	Average win / average loss ratio	required

Returns:

Type	Description
`Decimal`	Fraction of bankroll to risk (can be negative or > 1)

Raises:

Type	Description
`ValueError`	If win_loss_ratio is not positive

Example

from decimal import Decimal

60% win rate, 1.5:1 reward/risk¶

kelly = kelly_fraction(Decimal("0.6"), Decimal("1.5")) float(kelly) # ~33% of bankroll 0.33...

50% win rate, 2:1 reward/risk¶

kelly = kelly_fraction(Decimal("0.5"), Decimal("2")) float(kelly) # 25% of bankroll 0.25

Use half-Kelly for reduced volatility¶

half_kelly = kelly / 2

Constraints¶

apply_position_limits¶

apply_position_limits ¶

apply_position_limits(
    desired_qty: Quantity,
    max_position: Quantity | None = None,
    max_notional: Decimal | None = None,
    price: Decimal | None = None,
) -> Quantity