QUANT RESEARCH LAB
A small quant research lab using math and statistics to design, test, and iterate systematic crypto strategies in a live environment
Live Research Portfolio
Research dashboard for an ongoing student project — not investment advice or solicitation
Strategy families
Click one of the three families below to view every alpha signal inside it, including the signal definition and rationale for each variant.
Documentation of research signals — not investment advice
Live Algorithm
2-level dynamic allocation — EWA weighting across coins and strategies
Weights evolve via EWA softmax over cumulative PnL performance. Higher-performing coins receive more capital.
Level 2 — Strategy Weights per Coin
Each coin runs 26 strategy variants independently. Weights shift toward strategies with higher risk-adjusted returns.
Strategy Research
Published research papers on systematic trading strategies
SMA Crossover System
Fast Stochastic Oscillator
Market Data
Methodology
Our research pipeline: hypothesis → backtest → robustness → live evaluation
SMA Crossover Strategy
Uses two Simple Moving Averages — a fast one (short-term trend) and a slow one (long-term trend). When the fast SMA crosses above the slow SMA, it signals a buy (golden cross). When it crosses below, it signals a sell (death cross). We explore parameter ranges and validate with out-of-sample checks to reduce overfitting risk.
Multi-SMA Consensus
Instead of relying on a single SMA pair, this strategy runs multiple pairs simultaneously. It only executes a trade when the majority of pairs agree on the direction, filtering out false signals and reducing risk. Think of it as a committee vote — no single indicator controls the decision.
Sharpe Ratio
Measures risk-adjusted return — how much return you earn per unit of risk taken. A ratio above 1.0 is good (decent return for the risk), above 2.0 is very good, and above 3.0 is exceptional. It's calculated as the average excess return divided by the standard deviation of returns.
Alpha
The excess return a strategy generates compared to simply buying and holding the asset. Positive alpha means the strategy outperformed the market. Negative alpha means you would have been better off just holding. Alpha is the ultimate measure of whether active trading adds value.
Max Drawdown
The largest peak-to-trough decline in portfolio value during the backtest period. If your portfolio went from $15,000 down to $10,000 before recovering, that's a 33% max drawdown. Lower drawdown means the strategy better protects against large losses — critical for real-money trading.
Backtesting
Simulates a trading strategy on historical price data to evaluate how it would have performed. Our engine uses real candle data from Coinbase, starting with a virtual $10,000 balance. It tracks every buy/sell, calculates returns, drawdowns, and risk metrics — letting you test before you risk real capital.
Validation
Every strategy undergoes rigorous validation before live deployment:
- Walk-forward / out-of-sample evaluation
- Parameter stability checks
- Stress tests across regimes
Execution Assumptions
Real-world constraints factored into every backtest and live execution:
- Fees + slippage
- Rebalance frequency
- Order/execution constraints
About Us
The team behind SJ Capital Quant Research Lab
Team & Collaboration
Student-led quant research project focused on designing, testing, and iterating systematic crypto strategies in a live environment.
Location
Based in the North East US and Hong Kong.
Open to Collaboration
We're always interested in connecting with like-minded researchers, engineers, and traders. If you'd like to collaborate, reach out at research@sj-capital.co.uk
Timeline
Founded: December 2025
Live capital: February 23, 2026