Market makers offer trades through order book & make money on bid-ask spreads (supply-side fees).

dYdX is a non-custodial trading and derivatives protocol initially built on Ethereum and later expanded to their own blcokchain built on top of the Cosmos SDK. dYdX was founded by Antonio Juliano and launched in 2017.

DYDX tokenholders own and govern the protocol. dYdX Trading Inc. currently takes all of the total trading fees paid by traders (revenue). dYdX’s investors include a16z, Polychain, The Spartan Group, etc.

governance

github

twitter

discord

etherscan

website

Derivative exchanges

dYdX

# dYdX Active users (monthly) Methodology

## 1. Overview

Monthly Active Users (MAU) measures the count of unique users who interacted with the dYdX protocol within any 30-day rolling window. This metric captures both general blockchain activity (transaction senders) and trading-specific activity (makers and takers in order matches). The calculation uses HyperLogLog probabilistic counting for efficient computation of unique user counts across large datasets and provides dimensional breakdowns by chain, user type, and product version.

## 2. Data sources

- **dYdX blockchain message events**: Raw blockchain event data from the dYdX chain containing all protocol interactions including general transactions and trading matches
- **Transaction message events**: Events with `event_type = 'message'` containing transaction sender information in the `event_attributes.sender` field
- **Trading match events**: Events with `event_type = 'match'` containing maker and taker subaccount information in `event_attributes.maker_subaccount` and `event_attributes.taker_subaccount` fields
- **Incremental processing metadata**: Table modification timestamps for efficient incremental updates

## 3. Step-by-step calculation

### Component A: Daily active user identification
- **Purpose**: Identifies unique users active on each day across different user types and products
- **Logic**: Extracts user addresses from three distinct activity patterns. For general transaction activity, parses the `sender` field from message events to identify transaction senders. For trading activity, parses match events to extract both `maker_subaccount` and `taker_subaccount` addresses, treating them as separate user types. Each user address is associated with its activity date (truncated to day), chain identifier, user type classification, and product version.
- **Key fields**: `block_timestamp` for temporal grouping, `user_address` for unique user identification, `chain` for protocol identification, `user_type` for activity classification, `product_id` for version tracking

### Component B: HyperLogLog sketch creation
- **Purpose**: Converts daily active user data into probabilistic counting sketches for efficient aggregation
- **Logic**: Groups daily active users by timestamp, chain, user type, and product dimensions. Initializes HyperLogLog sketches with precision parameter 15 for each group, allowing approximate counting of unique users with controlled memory usage. The HLL sketches enable efficient merging across time windows without storing individual user addresses.
- **Key fields**: `timestamp` for temporal grouping, `hll` sketches containing compressed user count information, dimensional keys for grouping

### Component C: Rolling window construction  
- **Purpose**: Creates 30-day rolling windows for each date to capture monthly active user periods
- **Logic**: Generates a complete date spine from the earliest available data to the current date. For each date in the spine, creates join relationships with the previous 29 days (30 days total including the current date). This ensures every day has a complete 30-day lookback window even when some days have no activity data.
- **Key fields**: `timestamp` for the target date, `joinday` for each day within the 30-day window

### Component D: Window aggregation
- **Purpose**: Merges HLL sketches across each 30-day window to count unique monthly active users
- **Logic**: Joins the rolling window structure with daily HLL data, matching each window day with its corresponding daily sketch. For each target date, collects all HLL sketches from its 30-day window and merges them using HyperLogLog merge operations. Extracts the final unique count from the merged sketch, providing the monthly active user count for each date and dimension combination.
- **Key fields**: `timestamp` for the MAU calculation date, merged `hll` sketches, dimensional breakdowns

### Component E: Dimensional breakdowns
- **Purpose**: Creates separate views of MAU data across different analytical dimensions
- **Logic**: Generates four distinct breakdowns from the window-aggregated data. The project-level breakdown aggregates across all dimensions to show overall protocol MAU. The chain-level breakdown groups by chain identifier while nulling other dimensions. The user type breakdown groups by activity type (transaction sender, maker, taker) while nulling other dimensions. The product breakdown groups by product version while nulling other dimensions. Each breakdown uses HLL merge operations to properly aggregate unique user counts.
- **Key fields**: `dimensions` label for breakdown type, relevant dimensional fields, nulled unused dimensions

### Final aggregation
- **Combining components**: All dimensional breakdowns are combined using UNION ALL operations to create a comprehensive MAU dataset with multiple analytical views
- **Currency conversion**: Not applicable - this metric counts users rather than monetary values
- **Time aggregation**: Uses 30-day rolling windows calculated daily, with each date representing the MAU for the trailing 30-day period ending on that date
- **Final calculation**: The final MAU value is the unique count extracted from merged HyperLogLog sketches across the 30-day rolling window, providing probabilistic but highly accurate unique user counts with controlled memory usage and computational efficiency

*This methodology was automatically generated by Token Terminal's methodology analyzer.*

Unique revenue-generating users over a rolling 30-day window.

dYdXDYDX

Active users (monthly)

Active users (monthly)

Competitive landscape

Chain breakdown