Decisions made in the moment
- Driven by news, opinions, chart "feel" and tips
- Hard to measure what is actually working
- Fear and greed creep into every decision
- Results can't be reproduced or improved systematically
Our approach
A scientific method for trading.
We treat every trading idea as a hypothesis to be tested, and most of them fail. Here is the process we teach, and the traps it is designed to avoid.
What "quant" means here
Rules you can write down, test and repeat.
Quant trading does not require a PhD or a supercomputer. It means every decision (what to trade, when to enter, when to exit, how much to risk) follows explicit rules that were tested on historical data before real money was involved.
Decisions made in advance
- Entry, exit and sizing rules defined beforehand
- Tested on years of data, including bad markets
- Removes most emotion from execution
- Every trade is measurable, so the system can improve
The six-step workflow
From idea to live system.
Hypothesis
Write the idea as a falsifiable statement, for example "stocks in the strongest sectors keep outperforming over the next 1–3 months." Note why it should work: behaviour, structure or risk premium.
Data
Use a point-in-time universe (only stocks that existed and were liquid at the time), adjusted for splits, bonuses and dividends, and including stocks that were later delisted.
Backtest
Simulate the rules day by day with no look-ahead. Include brokerage, STT, exchange charges, stamp duty, GST, taxes and realistic slippage for each stock's liquidity.
Robustness
Develop on an in-sample period, then judge on unseen out-of-sample data. Change parameters slightly. A real edge degrades gracefully, while a curve-fit one collapses.
Risk design
Decide position size, number of positions, maximum drawdown tolerance, volatility caps and when to stand aside, for example when the broad market is in a downtrend.
Paper → live
Paper-trade, compare live fills with the backtest, then start with small capital. Journal every trade and every rule you were tempted to break.
Traps we teach you to avoid
Why most backtests lie.
A backtest that looks amazing is usually a warning sign. These are the most common reasons historical results don't survive contact with real markets.
Survivorship bias
Testing only on today's index members ignores the companies that failed or were removed. The result looks better than reality.
Look-ahead bias
Using information that wasn't available at decision time, such as the day's close to trade at that same close, or restated fundamentals.
Overfitting
Tuning many parameters until the past looks perfect. The system memorises noise and fails on new data.
Ignoring costs & slippage
In small and mid-caps, the gap between the signal price and your actual fill can wipe out the entire edge.
Single-period luck
A strategy that only worked in one bull run is a bet on that regime repeating, not a system.
Underestimating drawdowns
The worst drawdown is still ahead of you. Live drawdowns are often deeper and longer than the backtest showed.
Strategy families we study
Well-researched ideas, tested on Indian data.
Momentum & relative strength
Ranking stocks or sectors by recent strength and rotating into the leaders.
Trend following
Moving-average and breakout systems that ride long trends and cut losers.
Sector & ETF rotation
Allocating across sectors or ETFs based on relative strength and cycles.
Regime filters
Rules that reduce exposure when the broad market is weak, to control drawdowns.
Important: Strategies are discussed as general concepts with historical examples only. We do not recommend specific securities, publish live signals, or suggest that any strategy will be profitable for you. Backtested results are hypothetical. See our Disclaimer.