Does refreshing a quote prove that routing improved?

A refreshed quote with more output proves that the displayed estimate changed. It does not, by itself, prove that the routing engine improved. The pool state, fee conditions, available liquidity or requested inputs may have changed between observations.

Identify the competing explanations

Suppose a hypothetical first quote returns 500 tokens and a refresh returns 503 for the same input. Possible explanations include a better route at the same state, a favorable market move on the original route, new liquidity, or a changed fee assumption.

A screen that changes the input amount, output asset or slippage display can create an even less comparable result. Confirm the request before interpreting the difference.

What would support an algorithmic claim?

The stronger comparison evaluates the old and new routing choices against the same market state, size, source set and fee treatment. Even then, compare net value after execution costs rather than raw output alone.

AMM execution prices depend on reserve state. A new estimate made against different reserves cannot isolate the routing algorithm as the cause of improvement.

Use language that matches the evidence

With ordinary interface observations, “the refreshed quote showed three more tokens” is defensible. “The router found a three-token improvement” requires evidence that other relevant inputs did not explain the change.

For a user deciding whether to execute, the current valid quote is usually more relevant than proving the cause of its improvement. Preserve its expiration, minimum output and separate charges. For a product evaluation, retain the request and state records needed to distinguish routing gains from market movement.

Repeated refreshing also does not guarantee a progressively better result. Quotes can worsen, expire or change route. Treat each as a fresh observation with its own conditions rather than a cumulative saving locked in by clicking refresh.