Educational and paper-trading simulation only. Not investment advice. Past performance does not guarantee future returns. Blanket Finance is not a SEBI registered investment adviser or research analyst.
Help
How to use the Quant Lab
What it does, what every term means, and how to get the most out of it.
"Is this a good stock?" — what the Quant Lab can and can't answer
The Quant Lab does not give a yes/no verdict on a stock and it is not a price predictor. It answers a different, more useful question:
“If I had mechanically followed strategy X (e.g. a trend-following moving-average rule) on this stock over this period, how would that have behaved — the returns, the worst loss, how often it traded, and how bumpy the ride was?”
So to research a stock: go to Builder, search the symbol (e.g. RELIANCE), pick one or more strategies, and run a backtest. Then read the metrics below to judge whether a given rule historically worked well on that stock — high risk-adjusted return (Sharpe), tolerable drawdown, and a reasonable number of trades are what you are looking for. Compare several strategies and symbols before drawing any conclusion.
Everything here is a historical simulation. Past performance does not guarantee future results, and this is not investment advice.
Getting started in 5 steps
1
Your market
The Quant Lab is currently focused on the India market. Capital, currency (INR) and the benchmark (NIFTY 50) all follow this market.
2
Browse the catalog
The Strategies and Dashboard tabs show pre-run results for a set of standard strategies, ranked by risk-adjusted return. This is the fastest way to see what tends to work in each market.
3
Run your own backtest
In Builder, search a symbol (e.g. RELIANCE, TCS), choose a strategy and a date range, then run it. You’ll need to be signed in to run a live backtest on the engine.
4
Read the result
Open a backtest to see the summary metrics, the equity curve vs the benchmark, the drawdown chart, monthly returns, and the list of simulated trades. Use the glossary below to interpret each number.
5
Compare & stress-test
Use Compare to put strategies side by side, and the Risk, Monte Carlo and Walk Forward tabs (available for catalog strategies) to check how robust a result is rather than trusting a single lucky run.
Glossary — performance metrics
Total Return
The overall percentage gain or loss of the simulated portfolio over the whole period.
CAGR
Compound Annual Growth Rate — the total return expressed as a smoothed yearly rate, so you can compare runs of different lengths.
Max Drawdown
The largest peak-to-trough drop the strategy suffered. A −40% max drawdown means that at the worst point you’d have been down 40% from a prior high. Lower (closer to 0) is better; this is the main “how painful was it” number.
Sharpe Ratio
Return earned per unit of risk (volatility). It’s the headline “quality” score: roughly, <0.5 is weak, ~1 is decent, >1.5 is strong. The leaderboard ranks by this.
Round Trips
The number of completed trades (a buy followed by its matching sell). More round trips means a more active strategy.
Trade Legs
The number of individual buy/sell executions. One round trip is usually two legs.
Invested Time %
The fraction of the period the strategy was actually holding a position rather than in cash. Low values mean the strategy sits out a lot of the time.
Avg Holding Days
How long, on average, each position is held before being closed.
Trades / Month
Average trade frequency — a feel for turnover and how much attention/costs the strategy would imply.
Glossary — charts & analysis
Equity Curve
The simulated portfolio value over time. The benchmark line (NIFTY 50, where available) shows how a simple buy-and-hold of the index would have done over the same window.
Drawdown Curve
The “underwater” chart — how far below its previous peak the portfolio was at each point in time. Useful for seeing how long and how deep the bad stretches were.
Monthly Returns
A calendar heatmap of each month’s gain/loss, to spot seasonality and consistency.
Monte Carlo
Reshuffles the historical trades thousands of times to estimate a range of plausible outcomes and the probability of loss — a check that a result isn’t just one lucky path.
Walk Forward
Tests the strategy on rolling out-of-sample windows (optimise on the past, test on the next unseen slice) to gauge whether it holds up on data it didn’t “see.”
The strategies, in plain English
What each one does, when it works, and what a good result looks like
SMA Crossover (trend)
Goes long when a faster simple moving average (e.g. 20-day) crosses above a slower one (e.g. 50-day), and exits when it crosses back below. It tries to ride sustained, smooth up-trends and step aside in downtrends. Works best in: strongly trending markets. Struggles in: choppy, sideways markets (lots of false crosses). Good output: beats buy-and-hold of the benchmark with a noticeably smaller max drawdown, Sharpe near or above 1, and a modest number of trades.
EMA Crossover (trend)
Same idea as SMA Crossover but uses exponential moving averages, which react faster to recent prices. It catches turns sooner but also produces more whipsaws. Works best in: fast-moving trends. Good output:similar to SMA but you’d expect more trades; the win is a higher Sharpe without the drawdown blowing out.
20-Day Breakout (breakout)
Buys when price closes above its highest level of the last 20 days and exits on a break to recent lows, betting that a new high attracts further buying. Works best in: markets that make strong, persistent moves. Good output: a few large winning trades carrying the result, an equity curve that steps up in bursts, and a controlled drawdown between breakouts.
RSI Mean Reversion (mean-reversion)
Buys short-term oversold dips (low RSI) within an up-trend and sells once price recovers, betting that overdone drops bounce back. Works best in: range-bound or gently rising markets. Dangerous in: sharp downtrends (it “catches falling knives”). Good output:a high win rate, short average holding days, and a smooth equity curve — but always check the worst drawdown.
Bollinger Reversion (mean-reversion)
Fades moves that stretch below the lower Bollinger band (a volatility envelope around the average) and exits back at the middle band. A volatility-aware version of buying dips. Good output: consistent monthly returns, contained drawdown, and a Sharpe that holds up under Monte Carlo.
Momentum Ranking (momentum)
Rotates into whatever has shown the strongest recent performance, on the idea that winners keep winning over the short term. Works best in: trending markets with clear leadership. Good output: outperforms the benchmark over full cycles with a reasonable drawdown and steady turnover.
Cross-Sectional Momentum 3M/6M (momentum)
Ranks an entire basket of stocks by their blended 3-month and 6-month momentum each period and holds the top names, rebalancing as leadership rotates. This is a portfolio strategy, not a single-stock one. Good output: a higher Sharpe than any single-stock rule and lower drawdown thanks to diversification across the basket.
Multi-Factor Core (multi-factor)
Blends several signals at once — momentum, low-volatility and trend — with a market-“regime” overlay that dials risk down in hostile conditions. The most diversified, “all-weather” rule set here. Good output:typically the best risk-adjusted profile in the catalog — the highest Sharpe and the shallowest drawdown, even if its headline total return isn’t the largest.
What a good result looks like (quick checklist)
- Sharpe around 1 or higher — the single best “quality” signal.
- Max drawdown you could actually stomach (shallower is better), ideally smaller than the benchmark’s.
- Beats the benchmark equity curve over the full period, not just in one hot stretch.
- Enough trades to be statistically meaningful — not 3 lucky ones.
- Robust: the edge survives Monte Carlo (low probability of loss) and Walk Forward (holds up out-of-sample).
- Consistent monthly returns rather than one giant month masking many losses.
Data freshness & how often it updates
Prices
The Quant Lab uses end-of-day (EOD)prices — the official daily open/high/low/close/volume for each stock — current to the most recent trading day.
Catalog strategies
The pre-run strategies shown on the Dashboard and Strategies tabs are recomputed automatically once per day (after the daily EOD data refresh), so the leaderboard reflects the latest completed trading day.
Your custom backtests
Computed live, on demand, against the latest EOD data available at the moment you click run — so they always reflect the most recent close.
No intraday / live trading
The Quant Lab works on daily bars only and never places real orders. It is a paper simulation by design.
Tips to get the most out of it
- Judge a strategy by risk-adjusted return (Sharpe) and drawdown together — not by total return alone.
- Always compare against the benchmark. Beating buy-and-hold is harder than it looks.
- Test the same strategy across several symbols and date ranges before trusting it.
- Prefer results that survive Monte Carlo and Walk Forward over a single eye-catching backtest.
- Be suspicious of very high returns with very few trades — that’s often luck, not edge.
- Use longer date ranges so a result spans different market conditions (bull, bear, sideways).