Overview
Tardis.dev specializes in historical cryptocurrency market data at the highest resolution — tick-by-tick order book snapshots, trades, liquidations, and funding rates. It provides normalized data across major crypto derivatives exchanges, ideal for quantitative research and backtesting. Unlike traditional OHLCV data providers, Tardis captures every market event with microsecond precision, preserving the exact sequence of order book changes and executed trades. This granularity is essential for accurate simulation of high-frequency strategies, market microstructure analysis, and understanding real-world order execution dynamics.
What It Is
Tardis.dev is a specialized data infrastructure provider for cryptocurrency derivatives markets. It reconstructs and stores complete historical market state from raw exchange feeds, then serves that data in multiple formats suited to different analysis needs. Rather than aggregating or sampling data, Tardis maintains the complete audit trail of market events — every order book update, trade execution, liquidation, and funding payment. This approach ensures your backtests reflect actual market conditions rather than interpolated or smoothed approximations.
The platform normalizes raw exchange protocols into a consistent schema, eliminating the friction of working with each exchange's proprietary API format. You get uniform column names, timestamp handling, and data types whether you're analyzing Binance Futures, Bybit, or Deribit. This consistency saves weeks of data cleaning and reduces bugs introduced by format translation.
Key Features
Tick-level data: Access L2/L3 order book snapshots at every update, not just at regular intervals. Every trade execution, every order book change, every liquidation event is timestamped and sequenced. This lets you model slippage, market impact, and fill probability accurately instead of assuming instant execution at mid-price.
Derivatives data: Complete coverage of perpetual futures, quarterly/dated futures, options, and funding rates. Open interest snapshots and liquidation data let you assess market structure and systemic risk. Funding rates are crucial for arbitrage, carry trades, and understanding long/short imbalance.
Normalized format: Instead of learning each exchange's data schema, work with consistent column names and data structures across Binance, Bybit, OKX, Deribit, BitMEX, and others. Symbol naming, timestamp precision, and decimal handling are standardized, reducing integration errors.
Data replay: The WebSocket-like streaming API lets you replay historical data as if it were live. This is critical for testing strategies that depend on timing, latency simulation, or multi-leg order logic. You can throttle replay speed, pause/resume, or jump to specific timestamps.
Machine-readable formats: Download data as CSV, JSON, or binary formats. CSV suits pandas workflows and quick analysis. JSON is flexible for nested structures like order book snapshots. Binary format compresses large datasets and speeds up bulk downloads.
Exchange coverage: Access data from Binance Futures, Bybit, OKX, Deribit, BitMEX, FTX (historical archive), Huobi, Gate.io, and growing coverage. Most exchanges have data from 2019 onwards, with daily updates for current data.
Open-source tools: The tardis-dev npm and Python packages provide local utilities for data manipulation, replay, and format conversion without API calls. Use them in your data pipeline or backtesting harness.
Who Uses Tardis.dev
Quantitative traders and researchers use Tardis to develop systematic strategies on real market data. Quants building factor models, pairs trading algorithms, or statistical arbitrage need tick-level accuracy to validate edge.
Day traders and scalpers rely on order book microstructure data to understand support/resistance, liquidity clusters, and short-term price momentum. The order book snapshots reveal how major orders stack and move, informing entries and exits.
Swing traders use Tardis for multi-day backtests that incorporate funding rates, liquidation cascades, and realized volatility. Understanding when liquidations cluster helps predict volatility spikes and breakouts.
Algorithmic execution traders develop market-making and liquidity-taking strategies on actual order book state, then backtest with realistic fill assumptions rather than best-bid/ask approximations.
Risk managers and researchers analyze historical liquidations, extreme price moves, and market structure to understand tail risk, systemic vulnerability, and correlation breakdowns during stress.
Crypto hedge funds use Tardis as their foundational data layer for strategy research, portfolio backtesting, and live execution analysis. The normalized format scales across multiple exchanges and asset classes.
Practical Trading Scenarios
Order book imbalance signals: Analyze the ratio of buy to sell volume at different price levels. When bid-side order book weight spikes, price often follows. Use tick-level snapshots to detect these imbalances before they show up in OHLCV data, giving you a 100–500ms edge.
Liquidation prediction: When liquidation cascade data shows heavy clustering at round-number price levels, expect volatility and slippage to worsen. Adjust position sizing or avoid crossing those levels during high leverage periods. Backtest strategies that tighten stops or add hedges when liquidation risk is high.
Funding rate carry trades: The exact timing of funding rate payments matters for entry/exit. Tardis data lets you simulate funding accrual accurately and test whether capturing carry is worth holding through intra-period volatility.
Slippage modeling: Rather than assuming fixed spread or percentage slippage, rebuild realistic fill assumptions using actual order book depth at trade times. A 1M USDT market sell encounters different slippage at different times of day and market conditions — Tardis captures this variance.
News/event reaction timing: Backtest whether your strategy reacts faster or slower than the market to events. With microsecond timestamps, you can identify the exact moment price momentum shifts and test if your signals would have captured or missed that move.
Multi-exchange arbitrage: Compare price and order book state across Binance, Bybit, and OKX simultaneously. Identify fleeting arbitrage windows and verify they're large enough to cover transaction costs after realistic fill assumptions.
Liquidation-triggered strategies: Some traders profit from rapid recovery after liquidation cascades. Test whether your entry logic would have triggered, whether fills would have been achievable, and what your slippage would realistically have been.
Integration with StratBase.ai Backtesting
Tardis.dev and StratBase.ai complement each other powerfully. Tardis provides the market data foundation; StratBase executes your strategy logic against that data.
Data import workflow: Download your target symbols and date ranges from Tardis in CSV or JSON format. StratBase ingests this tick-level data and replays it through your strategy, preserving timestamps and event order. You avoid the friction of custom parsing or format conversion — both platforms speak the same language.
Realistic execution simulation: StratBase's order matching engine can consume Tardis order book snapshots to simulate realistic fills. Instead of assuming you get filled at mid-price instantly, StratBase matches your orders against the actual order book state at each timestamp. This reveals true slippage, partial fills, and cases where your order would not have filled in the size you intended.
Backtest accuracy without black boxes: By combining Tardis data with StratBase's transparent strategy engine, you see exactly how orders would execute, what prices you'd receive, and how your returns change under realistic assumptions. You're not trusting a vendor's slippage model — you're testing against real market state.
Multi-timeframe and multi-exchange testing: Tardis covers perpetuals on eight major exchanges; StratBase lets you build strategies that trade across multiple venues simultaneously. Test statistical arbitrage, market-making, or execution algorithms that need coordination across exchanges.
Strategy refinement loop: Use StratBase to iterate strategy logic quickly on Tardis data. Test signal thresholds, position sizing rules, and risk limits. Once confident, export the backtest results and dive deeper into specific trades using Tardis's raw data exports for post-analysis.
API Details
| Feature | Details |
|---------|---------|
| Access | HTTP REST endpoints for data download and streaming replay via WebSocket-like API |
| Authentication | API key provided via Authorization header |
| Download formats | CSV (row-based), JSON (object-based), binary/protobuf (compressed) |
| Streaming API | Connect to live replay of historical data with customizable speed and pause/resume |
| SDK availability | Python: pip install tardis-dev / Node.js: npm install tardis-dev |
| Timestamp resolution | Microseconds (UNIX microsecond integers) |
| Rate limits | Depends on plan; Professional and Enterprise have higher concurrency |
Data Coverage
Perpetual futures: Binance Futures, Bybit (inverse and linear), OKX, Gate.io, Huobi
Quarterly/dated futures: Binance, OKX, Deribit
Options: Deribit (most complete), some coverage on Binance and OKX
Spot trading: Limited on most exchanges; focus is derivatives
Data depth: 2019+ for most exchanges; FTX historical archive through 2022
Update frequency: Daily for historical data; streaming updates for current/recent data
Order book depth: L2 snapshots (top 1–50 levels) and L3 (full order book) depending on exchange and plan
Pricing
Free tier: Limited to recent data (last 30 days) and one exchange; good for initial exploration and format testing
Starter plan: $49/month — access to one exchange with six months of historical data; suitable for traders testing single-venue strategies
Professional plan: $149/month — all covered exchanges, full historical data back to 2019 or exchange launch, priority support
Enterprise: Custom pricing — dedicated account management, SLA, custom data exports, higher API rate limits, integration support
Most traders running serious backtests use Professional or higher; the monthly cost is minimal compared to the value of accurate backtest results.
Getting Started
-
Sign up at tardis.dev and verify your email. Create an API key in your account settings.
-
Install the client library:
pip install tardis-devor
npm install tardis-dev -
List available symbols and date ranges for your target exchange to understand what data exists and when it starts.
-
Download historical data for your symbols and date range. Specify format (CSV or JSON) and data type (trades, order book snapshots, liquidations, funding rates).
-
Ingest into StratBase.ai by uploading the CSV/JSON export or connecting via API if available. StratBase will parse timestamps, symbol names, and prices automatically.
-
Run backtests within StratBase using realistic order book fills and execution logic.
-
Iterate: Refine strategy parameters, test new signals, and compare backtest results across different symbol pairs or market regimes.
For raw exploratory analysis, use the Python package directly in Jupyter notebooks to query, aggregate, and visualize tick data before formalizing it into a strategy.