Why a slightly larger swap can suddenly receive a much worse rate

A swap can have good average execution while it remains inside a dense range, then deteriorate faster after it crosses into a thinner region. The important threshold is how much input reaches the boundary, not a round dollar size.

Inspect the added portion

Consider hypothetical quotes from the same state: an input of 1,000 returns 998 output units, while 1,100 returns 1,090. The extra 100 input produces only 92 additional output, despite the smaller trade's average rate being 0.998.

Those numbers alone do not prove a range boundary caused the change. They identify a deterioration that deserves inspection. A route switch, fee change or timing mismatch can create similar observations.

Match the depth evidence

Active liquidity can change when a swap crosses an initialized tick; the net change can be zero when entering and leaving positions offset. If the larger swap crosses a boundary where substantial liquidity ends, the additional portion can encounter a thinner curve.

Compare quotes near the suspected threshold and retain the same starting state where possible. Inspect whether the route stays in the same pool and whether the quoted fee treatment remains unchanged.

Do not average away the warning

The 1,100-input quote has an average output rate of about 0.990909. That number is less revealing about the last 100 than its incremental 0.92 rate. Both measures are useful, but they answer different sizing questions.

A smaller trade may avoid the thin region, yet splitting and immediately repeating against the same pool does not restore the original state. A later piece can still reach that boundary. Any claimed benefit from waiting must identify the assumed replenishment or market change.

Report the threshold as state-dependent. Liquidity providers can change their ranges, and other trades can move the starting price. A boundary observed earlier is not a permanent maximum efficient trade size.