[range]bounddocs Open the app
Methodology

Data sources and freshness

rangebound builds its numbers from on-chain data first. Market data from outside the chain is used only for what the chain can't provide: listed option prices and the underlying stock's own trading.

Sources

DataSourceHow often
Pool price, liquidity in range, fee counters, dynamic-fee stateThe pool accounts on SolanaEvery 30 seconds
Vault balances (TVL)The pool's token accounts on SolanaEvery 10 minutes
Liquidity at each tickRaydium's pool data, checked against the on-chain liquidity at the current price; for Orca, built from the positions in the poolEvery 10 minutes on Raydium, with each position scan on Orca
Open positions, ranges, owners, uncollected feesA scan of every position account in each poolHourly
Each position's historyThe position's own transactions on SolanaWhen a position is new or has changed
Swaps missed during an outageThe pool's transaction history on SolanaEvery 10 minutes, for the last 24 hours
24-hour token price changeJupiterEvery 5 minutes
Realised volatility of the underlyingPublic market data for the listed stock or ETFEvery 5 minutes
Implied volatility and option chainsPublic listed-option data for the underlyingEvery 30 minutes in US market hours
Meme tokensLaunchpad and AMM pools on Solana, and Jupiter token dataSee Memes
Tracked tokensA fixed list of issuer mints, checked against issuer lists and on-chain authoritiesUpdated when issuers list new tokens

Why fees come from counters, not from swap-by-swap indexing

Each pool keeps running totals of the protocol's share of fees. Because that share is a fixed fraction of every swap fee, the change in the totals between two readings gives the exact fees paid in between, with no swap missed. Volume follows by dividing fees by the fee rate.

This was checked against fully decoded swap histories: volume from counters matched decoded swaps to within a few percent on almost every pool. See Fees, volume and APR.

After an outage

If data collection stops, for example during a restart, the gap in the last 24 hours is rebuilt from the pool's transaction history, one reading per minute. The rebuilt hours match swap-by-swap data to within about 1% on volume. 24-hour figures therefore don't need a day to recover after a restart.

When data is stale

If the latest data is more than 5 minutes old, the app shows a banner saying so. Pages keep working on the last good data and refresh on their own when new data arrives.

Tokens and lookalikes

Only tokens minted by the real issuers are tracked. A token with the same name but a different mint address is listed under Lookalike mints excluded on the Assets page and left out of every figure.