Abstract

It seems equity perps have found their product market fit in a matter of months. Since HIP-3 launched only a month and a half ago, RWA pairs on Hyperliquid have accumulated over 3.5 billion dollars in volume, with many pairs averaging 50m+ in daily volume. Not to mention the volume on RWA's being accumulated on competitors. However, for the sake of simplicity, we will focus on the pricing of Hyperliquid's HIP-3 pairs.

We are likely in the early innings of a secular migration from legacy derivatives to onchain markets. Perpetuals are simply a superior vehicle for speculation than traditional futures because they consolidate liquidity into a single continuous instrument, eliminating the fragmentation and rollover costs associated with monthly expirations. Moreover, they allow process of retail speculation significantly simpler. No strike dates, no expiration dates, no time decay, just a simple way to bet on price action with leverage and no requirements for holding spot of the asset.

How do traditional perps work?

For those familiar with the fundamentals of how perpetual futures operate, feel free to skip ahead to section 3. Below is a brief introduction to those new to perps, courtesy of jez.eth:

A quick primer on how perpetual futures work, since it will be needed later. Feel free to skip. Perpetual futures are synthetic contracts that track an underlying "oracle price" through a funding mechanism. If the perpetual's mark price is above the oracle, the funding is positive, and longs pay shorts a small % of their notional every period. This incentivizes selling through closing longs or opening shorts, which pushes the mark price down towards the oracle.

Vice versa, if the perps mark is below the oracle, the funding is negative, and shorts pay, which incentivizes buying through shorts closing or new longs opening.

This filter ensures that on long enough timeframes the perpetual tracks the oracle, although in times of volatility this relationship can break down due to stronger external forces (liquidations). Note there is always a 1:1 ratio of longs to shorts (when a long is opened either a short is opened or a long closes).

Perps naturally arose from digital assets, due to the requirement of an underlying asset that trades 24/7 (or rather an oracle price that can be sourced 24/7). The oracle price, for most exchanges, is simply an index of spot prices across major exchanges, often weighted by liquidity to avoid any manipulation and irregularity from volatility on a single exchange.

HIP-3

Before we dive into the details of equity perps, it is first required to understand why they have seemed to arise now. This is due to a fundamental update from the largest onchain venue for trading perps, Hyperliquid.

Hyperliquid recently pushed a significant update - HIP-3 - allowing builder deployed markets, permissionlessly. In other words, anyone can list their own perps market on Hyperliquid, given they meet the staking requirement (500k $HYPE ~ 20 million).

Before HIP-3, only Hyperliquid themselves could add new perp pairs. After HIP-3, anyone can deploy a new market—an equity, an index, an RWA, whatever—as long as they provide the oracle feed and basic market parameters. This means external companies can create equity markets, with their own parameters for oracle and mark pricing.

Currently, there are only 3 active deployers operating HIP-3 markets. We will now compare how they formulate their oracle and mark pricing.

TradeXYZ

Formulation for oracle prices for public equity markets can be broken into two methods, across two corresponding time periods; internal and external hours. External hours refer to regular trading hours (for NYSE or equivalent), with internal hours referring to when the stock market is closed. Note these hours can change between individual pairs.

External Hours (Indices)

In contrast to standard cryptocurrency perpetuals, where the index oracle price is typically an aggregate of spot prices from various centralized exchanges, equity perpetuals for indexes utilize a distinct pricing mechanism. The oracle price is not derived directly from the underlying spot index constituents. Instead, it is reverse-engineered from liquid, externally sourced dated-futures markets (such as the CME E-mini S&P 500).

To ensure the oracle accurately reflects the underlying spot asset rather than the futures premium, the futures quote is adjusted using a standard cost-of-carry model. The relationship is defined as follows:

where:

• represents the derived Spot (Oracle) price.
• is the observed price of the externally sourced dated-futures contract.
• denotes the time to settlement (annualized).
• is the risk-free interest rate (currently calibrated to market conditions at approximately 4.4%).
• represents the continuous forward dividend yield of the index.

External Hours (Single Name Equities)

For individual equity pairs, external hours follow traditional US trading hours (8:30-4:40 ET), sourcing real time spot prices from the NYSE or equivalent. Note here that as the oracle price is not reverse engineered from futures markets, external hours are NOT 24/5.

Internal Hours

The key challenge posed is how to formulate oracle pricing when there is no active spot market to source from. In internal hours, these perpetual equity markets become the primary venue for price discovery. Such is reflected in the oracle pricing below:

To accurately model market sentiment while maintaining stability, we first quantify the directional pressure exerted by the order book. We define the Impact Price Difference (IPD) as the asymmetry between the current impact prices and the prevailing oracle price.

where:

• denotes the current oracle price.
• and represent the impact prices derived from the bid and ask sides of the order book, respectively.

Continuous-Time Update Mechanism

The oracle mechanism is designed to be robust against irregular data updates and market halts. To achieve this, it utilizes a time-dependent weighting scheme with a decay constant hour. The state update is governed by a continuous-time Exponential Moving Average (EMA) formulation:

The decay factor, , is calculated based on the elapsed time since the previous update, ensuring smooth convergence:

To prevent excessive volatility during extended periods of inactivity (e.g., market halts), the time delta is subject to a clamping mechanism:

Given the constant , the maximum single-update adjustment is bounded, such that .

Note on Basis Sampling

This formulation represents the generalized form of the continuous-time EMA. In parallel, the Mark Price component samples the basis deviation utilizing a significantly faster time constant of seconds to capture high-frequency market microstructure effects.

Transitioning Between Internal and External Hours

The transition from external to internal hours is simple; the last external oracle price simply becomes the first internal oracle price.

However, one interesting point of discussion lies around the period of change from internal to external hours. In theory, with enough volume trading equity perps in internal hours, pricing should be efficient enough to match Monday's opening price, effectively transforming the perp market into the leading venue for price discovery, where the NYSE doesn't just open, but rather 'gaps' to catch up to the reality already established on these derivative markets.

Mark Pricing

The mark price is the median of three components:

1. The oracle price.
2. The sum of the oracle price and a 150-second continuous-time exponentially weighted moving average of the difference between the perpetual's mid-price and the oracle price.
3. The median of the best bid, best ask, and last trade.

Relayer updates are clamped to ±50 bps of the current value to mitigate significant price jumps. This applies to components (1) and (2) of the mark price, which are clamped to the current mark, as well as the oracle price.

Ventuals

One of the most innovative use cases of perps is the ability for retail traders to speculate on private markets, given a pricing feed for the private company can be sourced. If you can formulate pricing data, you can turn startup valuations into tradable futures.

Currently doing this on Hyperliquid is Ventuals, offering markets for OpenAI, Anthropic and SpaceX, three of the largest private companies in the world.

Oracle Pricing

So how do you source 24/7 pricing for private companies which trade irregularly? The short answer: secondary markets, combined with a smoothing equation of the mark price. This gives a much larger weighting to natural price discovery, appropriate for these earlier, more volatile and rapidly growing companies.

Ventuals sources their oracle pricing feed externally from Notice. Notice provides 24/7 pricing for private companies like Anthropic by aggregating a continuous stream of real transaction data from a large network of brokers and investors, combining it with internal valuations and real bids/asks, and then feeding all of it through a weighting algorithm that leans on private-market data when it's fresh and shifts toward public comparables when activity slows. The result is a continuously updated "consensus price" that functions like a real-time mark for assets that normally trade in slow, opaque OTC markets.

By weighting 50% of the oracle price to an 8h EMA, Ventuals offsets the liquidity issues that these private markets pose. Notably, Notice data is only pulled once per minute, which is considerably less frequent than traditional perps, which often update by the second. This combination allows for a much more natural price discovery, as private companies should be traded with much less rigidity.

Mark Pricing

The mark pricing on Ventuals smooths short-term volatility by blending the most recent trade, the mid price, and a 30-minute exponentially-weighted moving average of those values. Specifically:

This approach balances real-time market activity with a smoothed historical view, reducing the impact of isolated trades or temporary order book imbalances. The median-based aggregation further guards against outliers, ensuring the mark price remains anchored to robust, representative data even when trading is thin.

Conclusion

Equity perpetuals represent a critical bridge between traditional finance and onchain markets. As these instruments mature, understanding their pricing mechanisms becomes essential for both traders and market makers. The methodologies employed by TradeXYZ and Ventuals demonstrate innovative approaches to solving the unique challenges posed by 24/7 onchain markets tracking assets with limited trading hours.

As adoption grows and more deployers enter the space, we expect to see further refinements in oracle design, mark price stability mechanisms, and risk management frameworks. The ability to trade equities with leverage, 24/7, without traditional settlement constraints, represents a fundamental shift in how speculation occurs.