bands.md

Bands

Config

{
    minMargin,
    avgMargin,
    maxMargin,
    minSize,
    avgSize,
    maxSize
}

Given a target price, the bands strategy ensures that net size of all orders within each band is in the range [minSize, maxSize]. Due to the binary nature of the order book, buy orders for $\text{TokenA}$ are equivalent from a liquidity perspective to sell orders for $\text{TokenB}$ with the same size and complementary price.

For each band, given a target price $\text{targetPrice}$, a $\text{minPrice}$ and $\text{maxPrice}$ are computed as follows:

Buy orders

$$ \begin{align*} \text{minPrice} &= \text{targetPrice} - \text{minMargin}, \text{ and} \\ \text{maxPrice} &= \text{targetPrice} - \text{maxMargin}. \end{align*} $$

Sell orders

$$ \begin{align*} \text{minPrice} &= \text{targetPrice} + \text{minMargin}, \text{ and} \\ \text{maxPrice} &= \text{targetPrice} + \text{maxMargin}. \end{align*} $$

There are two sets of bands: one for buy orders for $\text{TokenA}$ and sell orders for $\text{TokenB}$, and another for buy orders for $\text{TokenB}$ and sell orders for $\text{TokenA}$.

Each band considers buy orders and equivalent sell orders and

1. Cancels orders down to avgSize if open size is greater than maxSize.
2. Places orders up to avgSize if open size is less than minSize.

The strategy will always try to fill a band by placing sell orders before it attemps to place corresponding buy orders.