> ## Documentation Index > Fetch the complete documentation index at: https://docs.defiloops.com/llms.txt > Use this file to discover all available pages before exploring further. # Backtesting a position > One strategy over one pool: what it actually did, including impermanent loss and the comparison against simply holding **Tool:** `run_lp_backtest` Where [choosing a range](/skills) answers "what would this band have collected", this answers what happens when the position is allowed to **move**. Name a pool as `chain|PAIR|feePpm` — `ethereum|USDC/WETH|500` — a strategy from [list\_lp\_strategies](/tools/list-lp-strategies), a window and a size. ## In range comes first, and it decides everything else ``` Recentre on exit on ethereum|USDC/WETH|500 — $100,000 over 365 days in range 96.7% of days fee APR 57.9% net APR 57.1% after gas and swap costs impermanent -62.8%/yr kept -5.7%/yr fees after impermanent loss fees collected $57,910 impermanent $-62,828 total -6.4% against -11.5% for simply holding (+5.0 points) ``` The in-range figure leads because **every rate under it is computed across the whole window regardless**. A band that held the price 19% of the time earned nothing for the other 81%, and its fee APR is still quoted over the full period, because that is what an APR is. Read that example again. The fee APR is **57.9%** and the position still finished **down 6.4%**. Fees alone are the number people read as the outcome, and they are half a ledger. ## Three numbers, never one Fees, impermanent loss, and the total against **simply holding the same two tokens**. The third is the one that answers "was this worth doing" — in the run above the answer is yes, narrowly, because holding did worse. ## It tells you when its own inputs were poor Two warnings come back and both change how much the figures are worth: * **The indexer priced none of this pool's volume.** Fees were rebuilt from the stable leg rather than read. An estimate, and called one. * **The venue's protocol fee could not be read.** The cut taken out of every fee is assumed. On a pool where the cut is 25%, assuming none overstates by a third. The service also warns about the strategy itself. `recentre` reopens the band the day it goes out of range, so it is in range almost always *by construction* — and the answer says so, along with what a single day of delay costs. ## Size is an input, not a scale `capital` changes the answer and not merely its magnitude: fees split by liquidity, so a large position earns a share it could not earn twice. Past roughly a tenth of the pool, the answer describes a pool the deposit itself would have changed — and the tool says which fraction it is. ## It is history No forecast, nothing signed, and gas is not modelled. ## Related topics - [What we hold](/data/streams.md) - [What the agent knows](/skills.md) - [Every tool](/tools/reference.md) - [The catalogue](/adapters/catalogue.md) - [What it can do](/adapters.md)