Mark Price

For the complete documentation index, see llms.txt. Markdown versions of documentation pages are available by appending .md to page URLs; this page is available as Markdown.

Mark Price

Overview

The Mark Price represents the platform's canonical fair value estimate for each perpetual contract. This price determines margin calculations, liquidation triggers, conditional order execution (take-profit/stop-loss), and unrealized profit/loss accounting.

Whereas the Index Price reflects external markets only, the mark price blends that external anchor with Igniz's own live order book and a broader perpetuals aggregate to produce a fair value that is both responsive to local trading and resistant to manipulation.

Mark Price Composition

Mark price derives from the median of three independent price sources:

1. Adjusted Reference Price

The reference price adjusted by a 180-second exponential moving average (EMA) of the basis spread between Igniz's mid-market price and the external reference price. This adjustment captures local supply-demand dynamics while anchoring to external markets.

2. Platform Order Book Median

The median value of three internal data points:

• Current best bid price

• Current best ask price

• Most recent executed trade price

This component reflects real-time trading activity on the Igniz order book.

3. External Perpetuals Aggregate

A weighted median of perpetual futures mid-market prices from major centralized exchanges:

• Binance: Weight 4

• Bybit: Weight 2

• OKX: Weight 2

• KuCoin: Weight 1

• HTX: Weight 1

Exchange selection prioritizes liquidity depth and operational reliability.

Fallback Mechanism:

When exactly two of the three primary inputs are available, the system incorporates a 40-second EMA of the platform's order book median (best bid, best ask, last trade) as an additional input to the median calculation. This ensures robust pricing even during temporary data source outages.

Update Frequency

Both reference and mark prices refresh synchronously with validator oracle publications, maintaining approximately 3.5-second update intervals. This frequency balances real-time responsiveness with computational efficiency across the distributed validator network.

Exponential Moving Average Specification

The platform employs a time-weighted exponential moving average to smooth price transitions and filter transient volatility. For each update sample at time interval Δt since the prior update:

smoothed_value = accumulator_numerator / accumulator_denominator

Update Logic:

accumulator_numerator = accumulator_numerator × e^(-Δt / 3.0 minutes) + sample_value × Δt

accumulator_denominator = accumulator_denominator × e^(-Δt / 3.0 minutes) + Δt

The decay constant of 3.0 minutes provides approximately 95% weight on data from the past 9 minutes, effectively filtering noise while remaining responsive to genuine price movements.

Manipulation Resistance

The composite architecture provides multiple layers of manipulation resistance:

1. External Reference Independence: The adjusted reference component is anchored to off-platform data (see Index Price).

2. Multi-Source Aggregation: Mark price requires consensus across independent sources

3. Median-Based Calculation: Outlier prices cannot skew the final valuation

4. Temporal Smoothing: EMA components prevent flash manipulation from affecting margins or liquidations

This framework ensures that accurate, manipulation-resistant pricing underpins all risk management and settlement operations on the platform.