63-day
+22.6%
−14.8% drawdown
The Nifty 500, ranked by traded value and cut to the top 150. Liquidity is measured on prior data only — never on the window being tested.
Twelve months of price movement, skipping the last five days so short-term reversal cannot contaminate the ranking.
The strongest decile, equal weighted. No conviction sizing, no overrides, no opinion about which name deserves more.
Twenty-one trading days between rebalances. Low turnover keeps transaction costs below three percent of the move.
Brokerage, STT, exchange fees, GST, stamp duty and SEBI charges — applied to every trade in every backtest, not added afterwards.
Jegadeesh & Titman, 1993. Replicated on Indian equities by Sehgal & Balakrishnan and others. The edge was implemented from decades of evidence, not found in our data.
Tested across an identical date window, every lookback period beat the index. Longer lookbacks performed better — which is what the literature predicts.
Each looked promising before it was measured properly. The discipline is not in what survived — it is in what didn't.
Sixteen rebalances. The tested window covers roughly one and a half years — a single market phase. The research behind momentum spans decades precisely because that is what validating it requires.
Momentum crashes. It is documented to fail sharply during market reversals. A regime filter was tested and made results worse, so the exposure is deliberate and unhedged.
Backtests assume perfect fills. Live execution will differ by an amount that has not yet been measured.
Indian anomalies have faded. Published work on the NSE 500 finds size, value, volume and momentum effects weaker than in the 2000s.
Everything above was known before this page was written. None of it is a reason to stop — it is the reason the system runs on paper until live months accumulate.
Portfolio against Nifty 50, rebased to 100 at inception.
Positions, rankings and rebalance history come from the engine itself. Nothing on this page is illustrative.