How RangeScout evaluates liquidity ranges

RangeScout compares concentrated liquidity ranges using pool data, historical price tests and simulated scenarios. The scanner finds pools to investigate; the analyser estimates fees, divergence loss and rebalancing costs.

Updated 10 September 2026 · RangeScout Research

Start with the right pool and the right clock

A pool is identified by its chain and address. Pools with the same pair name can have different fees, liquidity and behaviour. Solana addresses are case-sensitive. Provider timestamps and unavailable fields matter: a missing fee is not a zero fee and an old scan is not a current quote.

Up to 180 days of available price history. The actual dates, observations and data limits belong to each report. Token or underlying-asset price proxies may be needed when pool history is limited. Those proxies do not establish historical pool liquidity or historical fees. Prices for ranges are quote-token units per base token; a dollar valuation additionally assumes or measures the quote token’s dollar price.

How fees are estimated

The preferred model uses observed pool fees and a position-specific share of active liquidity. This is a snapshot, not a record of that share throughout history. When that share is unavailable, a pool-wide fees/TVL or volume × fee-tier estimate is adjusted for the range. Those fallbacks are estimates, especially in sparse bins, changing fee tiers and unusually high turnover.

Historic fee income can therefore be a replay using today's supplied fee assumptions. It must not be read as fees that the position actually earned. Check each report’s fee source and model. The scanner’s gross fee yield excludes divergence, execution costs and time spent outside the range.

Historical strategy results

Each tested strategy specifies range boundaries and a rebalance policy. The model accrues fees while in range, accounts for divergence at simulated rebalances and at the end of the remaining position, and charges modelled rebalance costs. Daily closes cannot reconstruct every intraday fill; the optional bridge adjustment estimates missed range touches.

The displayed historical model return is fees minus divergence and execution costs over the tested window. It excludes the underlying tokens’ dollar price gains or losses and is not a complete wallet return. An annualized rate is a different quantity. Divergence relative to holding the starting tokens is also different from the wallet’s dollar return. Large token-price changes can dominate both fees and divergence. Recommendations must retain the boundaries whose metrics accompany them.

Simulated scenarios are a separate experiment

Monte Carlo scenarios resample historical returns. They depend on historical data and fee assumptions and are not forecasts of achieved returns. The range Monte Carlo evaluates the full allocation across its ranges with a static-hold payoff. Its policy differs from the historical rebalancing test. Percentile bands describe that conditional simulation, not guaranteed limits or the probability that a real account will achieve the same result.

What validation and confidence mean

Market-stability checks compare parts of the price history. Candidate ranges are not re-selected and tested on untouched future data, so this is not out-of-sample strategy validation.

Confidence describes the available evidence and model assumptions. A confident estimate of a loss is still a loss. Confidence is not a safety rating. No test here proves that tomorrow’s best range has been found.

What remains outside the model

Unexpected depegs, token restrictions, smart-contract failures, issuer pauses, liquidity withdrawals, MEV, discrete-bin behaviour, missing intraday data and execution delays can produce outcomes outside simulated bands. Live results require complete wallet and position accounting, verified deposits and withdrawals, and actual transaction costs. A position subtotal alone cannot establish a whole-portfolio return.

The concentrated-liquidity position equations are described in the Uniswap v3 whitepaper, section 2.

Inspect a calculation

Our reproducible examples show both a profitable and a losing synthetic price path. Their inputs and reference calculation are downloadable. They demonstrate the mechanics; they are not backtests of real pools or validation of the range-selection engine.

Read a saved pool analysis · Analyze your own pool · Pricing