How the backtester works

It's a time machine for LP strategies. Before you risk real money on a range, you can ask: "If I had done this in the past, what would have actually happened?"

You pick a strategy — say, $1,000 in a ±15% range, rebalanced weekly — and a time period. The backtester replays it against the real history of the pool: actual prices, actual trading volume, actual fees, hour by hour. Not a simulation of a made-up market — a replay of the one that actually happened.

Then it hands you the honest scorecard: fees you'd have earned, what the price movement would have cost you, gas and swap costs from rebalancing, and the bottom line — would you have beaten just holding the tokens, or not?

This is how we learned things that changed how we LP ourselves. Like: rebalancing the moment you go out of range — the thing that feels most responsible — was the worststrategy we tested, because it locks in your losses at the exact extremes prices tend to bounce back from. A boring calendar schedule beat it. We'd never have known that without replaying history.

One honest caveat, always: the past doesn't predict the future. The backtester tells you what did work, so you're at least not guessing — but no one can tell you what will work.

Accounting details (fee attribution, costs, data sources) are on the methodology page. Historical simulation — past performance does not predict future results. Not financial advice.