Real money drama — scams, layoffs, debt, windfalls and family fortunes — reported and told well.
A simulation over 237 trades tested four exit policies on an intraday momentum system. Holding to the closing square-off beat every alternative — and one popular early-exit rule didn't just underperform, it destroyed the edge entirely.
A single-session retest of an intraday system's trade tags saw order-book imbalance, gap behaviour, conviction scores and extension readings all reverse sign — while two structural patterns held exactly as before.
The strongest long pattern in our order-flow research kept its profit edge on its first out-of-sample day, earning far more per trade than the average signal. But the win rate, the one metric this project is built around, did not replicate.
A scan of six trading sessions found that longs breaking into a wall of resting sell orders win more often, especially when paired with momentum confirmers. The mirror pattern for shorts does not work.
Two out-of-sample sessions have now missed the setup's original hit rate. What replicated instead was asymmetry: the filtered trades roughly doubled the baseline payoff while winning slightly less often.
A replay of 391 intraday trades found that screening entries on how a stock's own sector was trading kept more money — the trades it declined looked like marginal winners in gross terms and were losses once real charges were paid.
A sweep of every gap day in India's derivatives-eligible stocks found continuation only in the smallest gaps, and only when the sector gapped the same way; the dependable signal turned out to be which gaps to leave alone.
Ranking every trade by how far it actually ran left one variable standing: relative volume. The catch is that the measure counts a whole session's turnover, most of which arrives long after the position is already open.
Across gap fades, block charts, oscillators and smart-money structures on Indian derivatives-eligible stocks, every entry-on-strength rule lost money net of fees — and only the deepest gaps survived a realistic slippage test.
On a bullish session that gapped up and faded, Strota's most selective long setup fired thirteen times, won 31% of them and lost money, against the 80% it showed on the data it was built from.
We re-ran a residual-momentum strategy with a positive out-of-sample record in crypto across ten years of Nifty 500 data. The same rules produced a negative Sharpe, a slower net-long rebuild cleared costs, and the best-scoring variant turned out to be harvesting an entirely different anomaly.
An audit of 430 trades across eight sessions found the whole gain sat in two of them, while 107 short trades averaged a loss — so fresh shorts were switched off and chasing already-extended stocks was capped.
Eleven rounds of testing failed to stop a momentum portfolio losing money in bad years; swapping in the twenty calmest liquid names did it, though the headline Sharpe of 1.84 is inflated by survivorship and unexecutable prices.
Six rounds of testing on a ten-year panel of Indian stocks, net of fees, evened out a lumpy strategy - but only by removing its signal, and the returns that remain rest on survivorship and prices nobody can trade at.
A decade-long test of intraday short-selling on Indian equities found the largest raw edge in the hardest-to-trade names, which is exactly where costs swallowed all of it and more.
Drawing the calmest names from the full Nifty 500 rather than the most-traded hundred turned a lumpy backtest into ten positive years and a Sharpe of 1.86 — at a level the study itself says survivorship inflates.
India VIX ran 2.5 percentage points above the movement that actually arrived, but only the defined-risk version of the trade kept its worst loss inside the limit, and the whole study rests on a proxy rather than real option prices.
Our first run said an outside momentum recipe was overstated. It was our data that was wrong: on dividend-adjusted prices the rule scored a Sharpe of 1.10 against the Nifty's 0.66 — on a universe of survivors.
Rebuilt on dividend- and split-adjusted prices, a low-volatility book kept a Sharpe of 1.93 and still managed 1.76 when restricted to large caps, while momentum's apparent strength leaned on the part of the market most likely to flatter a backtest.
A test of retail-executable fund-house algorithms on the Nifty from 2007 to 2026 found that a monthly switch between the index, gold and cash scored a Sharpe of 0.76 against 0.25 for buy-and-hold, and held up in every sub-period it was cut into.
Strota tested small fixed intraday targets five different ways. Every one lost. A fixed 0.0824% round trip, plus a 0.340% penalty on stop-triggered entries, swallowed effects worth 2 to 16 basis points.
Across 9,760 signals over six years of Nifty 500 data, the breakout the screen publishes nightly averaged -0.29% as it would actually be traded — and the friendlier long-horizon numbers answer a different question.
Cheaper instruments, delayed entries, sweep-and-reverse chart patterns and stacked indicator filters were all tested on Indian equities. None cleared costs, and three failed for reasons that had nothing to do with their signals.
Over the tested history, a long-only book buying Nifty 500 names that opened 5-7% down returned about 10-13bp per calendar day at Sharpe 1.1-1.5 — with survivorship, tail risk and a hard capacity ceiling attached.