Width instead of range
Instead of two prices you choose a width around the peg. Narrow widths quote aggressively near parity and tend to earn more while the peg holds. Wider widths keep quoting through deeper wobbles.
When pegs break
Depegs are not hypothetical: recent history runs from a weekend wobble to a permanent collapse.
- USDC, March 2023: fell as low as roughly $0.88 after Circle disclosed $3.3 billion of reserves at the collapsed Silicon Valley Bank, then recovered within days once regulators guaranteed the deposits.
- stETH, June 2022: priced 5-7% below ETH on secondary markets for weeks during the Celsius and Terra stress, while direct redemptions were not yet live.
- UST, May 2022: an algorithmic stablecoin that lost its dollar peg entirely. It fell to cents within days and never recovered.
To a pegged position these look identical at first: price drifts outside the width and the position accumulates the weaker asset. A wobble later recovers. A collapse does not. Pick pairs whose backing you understand, and size the width for the failure you can live with.
In practice
- Stable pairs: narrow width, small fee, high volume.
- LST pairs: slightly wider width to absorb rate drift.
- Adjust with market conditions: closing and reopening is cheap because tokens never move on close.