# On-chain perpetual futures, explained > How perpetual futures work without an expiry date, what funding and liquidation do, and what changes when execution moves on-chain. - Canonical page: https://lynx.finance/blog/onchain-perpetual-futures-explained/ - Published: 2026-07-30 - Updated: 2026-08-05 - Author: Lynx - Category: Learn - Reading time: 5 min - Keywords: on-chain perpetual futures, perpetual futures explained, decentralized perpetuals, crypto leverage trading Every dated futures contract eventually expires and settles. A perpetual future never does: it is a derivative with no scheduled expiry that gives a trader long or short price exposure without transferring ownership of the referenced asset. Most of the machinery in this guide, funding included, exists to manage that missing expiry. An on-chain perpetual uses blockchain-based contracts or accounting for part of that trade lifecycle. Funding, margin, pricing, execution, and liquidation still depend on the venue's specific design. ## Perpetuals, spot, and dated futures A spot purchase exchanges payment for an asset. A perpetual position references an asset's price but remains a contract governed by the venue. A dated future has an expiry or settlement date. A perpetual does not, so venues use recurring incentives and risk controls instead of relying on expiry to resolve price differences. No expiry does not mean the position remains open indefinitely. A trader can close it, and the venue can [liquidate](https://lynx.finance/glossary/liquidation/) it when the applicable risk threshold is reached. ## Why funding exists Perpetual venues use a [funding rate](https://lynx.finance/glossary/funding-rate/) to discourage persistent imbalance. The formula is not universal. Premium-based venues compare the contract with a reference [index price](https://lynx.finance/glossary/index-price/). When the perpetual trades above the index, longs commonly pay shorts; below the index, shorts commonly pay longs. Skew-based venues instead use the imbalance between long and short exposure. Lynx follows this model, according to its [fee documentation](https://lynx-finance.gitbook.io/lynx-finance/for-users/fees). Funding changes incentives but does not predict the next market move. Compare rates only after normalizing their intervals and checking the formula. The negative case is covered in [why funding rates go negative](https://lynx.finance/blog/why-funding-rates-go-negative/). ## Leverage starts with position size and collateral [Position size](https://lynx.finance/glossary/position-size-and-notional-value/) is the exposure. [Collateral](https://lynx.finance/glossary/collateral/) supports that exposure. [Leverage](https://lynx.finance/glossary/leverage/) is the ratio between them. `$5,000` of notional exposure supported by `$1,000` of collateral is `5×` leverage. A `4%` adverse move creates a simplified `$200` gross directional loss before funding, fees, and execution effects. Leverage on its own is not a complete risk score. The same multiple plays out differently depending on how volatile the market is, how long the position stays open, and how the venue prices collateral, charges fees, and triggers liquidation. The [leverage and position-size guide](https://lynx.finance/blog/leverage-margin-position-size/) turns those definitions into a sizing method. ## Margin and liquidation [Initial margin](https://lynx.finance/glossary/initial-and-maintenance-margin/) is the equity required to open or increase exposure in a common margin model. Maintenance margin is the minimum needed to keep it open. If eligible equity falls below maintenance, liquidation can begin. Other protocols define an equivalent trigger through Net PnL or a different risk formula. Some venues use a [mark price](https://lynx.finance/glossary/mark-price/) for margin checks rather than the last trade. [Hyperliquid's documentation](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/robust-price-indices) gives one concrete mark-price design. Liquidation estimates can move with funding, fees, collateral, account equity, and position changes. They are not substitutes for a planned exit. [Crypto liquidation explained](https://lynx.finance/blog/crypto-liquidation/) covers the trigger, execution, and money flow. ## What moves on-chain “On-chain perpetual” describes a family of architectures. Custody, accounting, matching, pricing, and settlement can each sit on-chain or depend on external infrastructure. Common on-chain considerations include: - **Authorization.** The trader signs transactions, messages, or delegated permissions. - **Price data.** Contracts need a source, freshness rule, and failure path for external market prices. - **Execution timing.** The market can move between a quote, signature, transaction, and settlement. - **Contract risk.** Code, privileged roles, configuration, and integrations create non-market risks. - **Network risk.** Congestion, fees, bridge status, and chain availability can affect access. Public state can make transactions and parameters inspectable. It does not guarantee that every dependency is on-chain or that the system is risk-free. For execution models, counterparties, and fee structures, see [what a perp DEX is](https://lynx.finance/blog/what-is-a-perp-dex/). ## Execution still determines the result The displayed quote is not always the final execution price. [Price impact and slippage](https://lynx.finance/glossary/slippage-and-price-impact/) can change entry or exit. [Unrealized PnL](https://lynx.finance/glossary/realized-and-unrealized-pnl/) remains an accounting value while exposure is open. Realized results include the close plus applicable trading fees, funding, borrow charges, and liquidation costs. When evaluating a venue, ask: 1. Which price sources drive the index, mark, execution, and liquidation calculations? 2. How are position size, fees, price impact, and funding shown before signing? 3. Is margin isolated per position or shared across an account? 4. What exactly occurs during partial or full liquidation? 5. Which contracts, networks, bridges, operators, and privileged roles does the system depend on? ## Where Lynx fits Lynx prices trades from an oracle and uses liquidity pools, not an order book, as the counterparty. [Isolated liquidity pools](https://lynx-finance.gitbook.io/lynx-finance/welcome-to-lynx/core-design/isolated-liquidity-pools) settle trader PnL in the selected collateral asset. Lynx currently uses Pyth for trade prices. Pyth aggregates publisher prices and confidence intervals, as described in [Pyth's architecture](https://docs.pyth.network/price-feeds/core/how-pyth-works). Position health is measured in that collateral asset, so a move in the token's external price does not push a position toward or away from liquidation. The token determines the pool and settlement asset, according to the [Lynx trader documentation](https://lynx-finance.gitbook.io/lynx-finance/for-users/traders). Lynx uses long-versus-short open-interest skew for funding. It also deducts the opening fee before applying leverage to post-fee collateral. Supported origin-chain users can submit later interactions through signed intents after funding an account. Users connected directly to the engine chain submit regular transactions, according to the [transaction-flow documentation](https://lynx-finance.gitbook.io/lynx-finance/welcome-to-lynx/cross-chain-perpetuals/transaction-flow). These details are specific to Lynx. They should not be projected onto every on-chain perpetual venue. ## Sources - [Lynx: Traders](https://lynx-finance.gitbook.io/lynx-finance/for-users/traders) — collateral, Pyth pricing, and liquidation behavior. - [Lynx: Fees](https://lynx-finance.gitbook.io/lynx-finance/for-users/fees) — position size, funding skew, borrow rate, and fee calculations. - [Lynx: Isolated Liquidity Pools](https://lynx-finance.gitbook.io/lynx-finance/welcome-to-lynx/core-design/isolated-liquidity-pools) — pool counterparty and settlement design. - [Lynx: Transaction Flow](https://lynx-finance.gitbook.io/lynx-finance/welcome-to-lynx/cross-chain-perpetuals/transaction-flow) — account funding, intents, and engine-chain transactions. - [Pyth: How Pyth Works](https://docs.pyth.network/price-feeds/core/how-pyth-works) — oracle publication, aggregation, and consumption. - [Hyperliquid: Robust Price Indices](https://hyperliquid.gitbook.io/hyperliquid-docs/trading/robust-price-indices) — one documented index and mark-price model used for comparison. Perpetual futures are complex and leveraged trading can produce rapid losses. This guide is educational and does not provide financial advice. Verify current protocol rules before trading.