| Concept | Hyperliquid | Binance | Join Key |
|---|---|---|---|
| Price | px | price | Direct comparison |
| Size | sz | qty | Same base unit |
| Timestamp | time (ms epoch) | time (μs epoch) | Normalize to ms |
| Side | side (B/A) | is_buyer_maker (bool) | B=!maker, A=maker |
| Coin | coin (e.g. BTC) | Symbol (e.g. BTCUSDT) | Strip USDT suffix |
| Notional | px × sz | quote_qty | Both in USDT |
| Field | Type | Research Value |
|---|---|---|
wallet | address | Track individual trader PnL, identify smart money |
closedPnl | USDC | Realized profit/loss per fill — aggregate by wallet for trader skill |
startPosition | float | Position size before fill — reveals conviction, scaling behavior |
dir | enum | Open/Close/Flip — distinguishes entry from exit, long from short |
crossed | bool | Aggressive (taker) vs passive (maker) — execution quality signal |
fee | USDC | Negative = rebate (maker), positive = fee (taker) |
| HL Coin | Binance Symbol | Binance Spot | Note |
|---|---|---|---|
| BTC | BTCUSDT | Since 2017 | Full history |
| ETH | ETHUSDT | Since 2017 | Full history |
| SOL | SOLUSDT | Since 2020 | |
| XRP | XRPUSDT | Since 2018 | |
| DOGE | DOGEUSDT | Since 2019 | |
| BNB | BNBUSDT | Since 2017 | |
| AVAX | AVAXUSDT | Since 2020 | |
| SUI | SUIUSDT | Since 2023 | |
| AAVE | AAVEUSDT | Since 2020 | |
| WLD | WLDUSDT | Since 2023 | |
| WIF | WIFUSDT | Since 2024 | |
| ENA | ENAUSDT | Since 2024 | |
| TRUMP | TRUMPUSDT | Since 2025 | |
| VIRTUAL | VIRTUALUSDT | Since 2024 | |
| ONDO | ONDOUSDT | Since 2024 | Gaps after Jun 29 |
| PENGU | PENGUUSDT | Since 2024 | |
| INIT | INITUSDT | Since 2025 | |
| BERA | BERAUSDT | Since 2025 | |
| HYPE | — | Not listed | HL-native token |
| FARTCOIN | — | Not listed | HL-native / Solana |
| MOODENG | — | Not listed | HL-native / Solana |
| POPCAT | — | Not listed | HL-native / Solana |
| GRASS | — | Not listed | HL-native / Solana |
Compare HL fill prices to Binance trades at the same millisecond. Measure price impact, latency premium, and venue selection alpha. HL has wallet-level attribution that Binance lacks.
alpha_research/src/alpha/
HL exposes wallet addresses on every fill. Rank wallets by cumulative
closedPnl, then forward-test: do top wallets predict direction?
Cluster by trading style using dir, crossed, position sizing.
ts_embed/skills/embedding/Binance 5-year history covers multiple regimes: 2020 COVID crash, 2021 bull run, 2022 bear market, 2023-24 recovery. Train HMM on volatility and volume to label regimes, then condition alpha signals.
hmmlearn + arch (GARCH)
Compute VPIN (volume-synchronized probability of informed trading) from
tick data. HL's crossed flag directly labels aggressor vs passive.
Compare maker/taker composition between DEX (HL) and CEX (Binance).
vpin repo + svpy/pyprod/
HL fills include fee (negative = maker rebate). Compare effective
cost (fee + slippage) between HL and Binance for the same coin.
Build venue selection model: when is HL cheaper than Binance?
svstrat/src/pyexecution/Aggregate tick data to bars (1m, 5m, 1h). Run ARIMA/GARCH for short-term vol forecasting. Use tsfresh for automated feature extraction. Ensemble with LightGBM, CatBoost for cross-sectional prediction.
tsprojection: pmdarima, tsfresh, optuna| Layer | Tool | Repo | Use in This Project |
|---|---|---|---|
| Data Ingest | boto3 (S3), urllib (CDN) | hyperliquid | HL fill download (requester-pays S3), Binance public archive |
| Data Ingest | yfinance, tradingview-ta | etfdata | Reference prices, technical indicators for cross-validation |
| Backtest | Custom engine (Sharpe, PnL) | alpha_research | Walk-forward backtest on HL+Binance alpha signals |
| Alpha Library | 26 signal implementations | alpha_research | Momentum, mean reversion, carry, HMM regime, seasonality |
| Portfolio | cvxpy, cvxopt | svstrat, cyfopt | Optimal position sizing, risk parity, min-variance |
| Time Series | pmdarima, arch | tsprojection | ARIMA/GARCH forecasting on aggregated bars |
| Feature Eng | tsfresh | tsprojection | Automated feature extraction from trade sequences |
| ML Models | LightGBM, CatBoost, XGBoost | svpy, nova | Cross-sectional return prediction, regime classification |
| Regime | hmmlearn | alpha_research | Hidden Markov Model for volatility regime detection |
| Embeddings | PyTorch, sentence-transformers | ts_embed | Embed trade sequences, semantic pattern matching |
| Vector Search | Qdrant, LanceDB | ts_embed, docvec | Find similar trading patterns across time |
| Risk | Custom metrics | svrisk | Drawdown, VaR, portfolio risk decomposition |
| Execution | Execution algorithms | svstrat | Venue selection, cost analysis, optimal routing |
| Visualization | Plotly, matplotlib | svpy, alpha_research | Interactive PnL curves, heatmaps, signal plots |
| Experiment | MLflow, wandb | nova, tsprojection | Track model versions, hyperparameter sweeps |
| Compute | Vast.ai (GPU), Dask | gpu-workers | Distributed feature engineering, model training |
| Optimization | optuna | tsprojection | Bayesian hyperparameter tuning for alpha parameters |
| Database | PostgreSQL, Redis | svpy, nova | Persistent trade storage, cached feature sets |
hl_trade_dl_standalone.pyHL S3 downloaderbinance_dl.pyBinance public archive downloader