Method

TradeOrSkip reads the NSE cash-equity market after each close and decides how much new long risk the conditions permit. It then ranks sectors, screens stocks, and — when a setup survives every gate — states an entry, the level that invalidates it, and a position size derived from capital you declare.

It is decision support, not investment advice. It does not know your circumstances, does not recommend that you buy anything, and is not a research analyst’s recommendation. It is long-only, end-of-day, NSE cash equity: no intraday, no derivatives, no shorting.

It refuses more often than it acts. On most sessions the answer is no new trade, and that is the intended behaviour rather than a shortage of ideas.

Source

NSE’s own published end-of-day bhavcopy, index closes and India VIX. Public files, not a broker feed — the pipeline runs unattended and does not depend on a session that expires.

Universe

NSE cash equity, EQ series. 4,207 instruments are currently active and tradable; the screened count on any given session is those with a bar for that session.

History held

9,065,156 daily bars, from 2011-01-03 onward, adjusted for splits and bonuses using exact rational factors rather than floating-point ratios.

Runs

Ingest at 17:45 IST, engine at 19:00 IST, weekdays. A session is sealed only when its data passes the quality gate.

Latest sealed session

2026-09-07

One score from 0 to 100, computed from four inputs with fixed weights. The weights are configuration, not code, and are content-addressed into every sealed run so a past reading can be reproduced against the exact numbers that produced it.

Trend · 40%

Nifty 50 and Nifty 500 against their 50- and 200-day averages, and whether the 200-day is rising.

Breadth · 30%

The share of the liquid universe above its own 200-day average. A market where the index rises on narrowing participation scores worse than the index alone suggests.

Momentum · 20%

Rate of change over the medium term.

Volatility · 10%

India VIX as a percentile of its own history, inverted — a calm tape scores higher than a violent one.

The score maps to a band, and the band sets a ceiling on total long exposure: below 30 permits 25%, 30–45 permits 50%, 45–60 permits 60%, 60–75 permits 90%, and above 75 permits 100%. A band change needs the score to clear the boundary by a margin, so a reading does not flicker between two states on noise.

The regime is available from 2012-02-21, which is the first session NSE publishes the index files it needs. The volatility component begins 2014-05-14, because India VIX does not exist before then. Neither is estimated backwards.

Liquidity floor

₹50,00,000 average daily traded value over the last 20 sessions, and no session with zero volume. Below that, a retail position cannot enter or exit without moving the price against itself.

Trend template

Price above its 50-, 150- and 200-day averages; the 150-day above the 200-day; the 200-day rising over the last 21 sessions; at least 30% above the 52-week low; within 25% of the 52-week high; and a relative-strength rating of at least 70.

Relative strength

Measured over 126 trading days against the Nifty 50.

Caps

At most 15 candidates from any one sector and 60 in total, so a single hot sector cannot fill the shortlist.

These gates run in that order, and the counts at each stage are published with every reading. The market and sector gates run last, on setups that already exist — a stock is not excluded for its sector before anyone has looked at the stock.

Size comes from risk, not from conviction. Quantity is the largest whole number of shares for which

(entry − stop) × quantity ≤ capital × risk-per-trade

where capital and risk-per-trade are yours. Risk per trade is capped at 1% and the interface does not offer more. The result is then reduced, never increased, by the regime’s exposure ceiling, by your portfolio heat limit, and by a drawdown throttle that shrinks size after losses.

Portfolio heat is the total you would lose if every open position hit its stop on the same day. It is the number that decides whether you can afford one more trade.

A sealed reading is not edited. Its inputs, its configuration and its output are hashed, and the archive at /today shows what was said on the day.

Two things legitimately change underneath it. A corporate action — a split or a bonus — retroactively rescales historical prices, so a chart drawn today will not match one drawn last year; the adjustment factors are exact rationals and are applied as new epochs rather than by overwriting. And a data correction from NSE can cause a session to be recomputed, in which case the run is retired and replaced rather than amended in place.

No performance record

This system has published for months, most of it in one regime. Any win rate or return drawn from that is noise with a decimal point, so none is shown.

The tradable universe is reconstructed

NSE publishes no historical index membership, so the liquid universe is rebuilt from observed liquidity and is labelled as such. It is never presented as NIFTY 500 constituents.

Sectors are today’s classification, backdated

A company’s sector is stable, so this is defensible — but delisted companies carry no sector, which biases historical sector membership toward survivors.

End-of-day only

Nothing here observes an intraday move. Two strategies that need a live price feed are skipped for accounts without one, rather than approximated from yesterday’s close.

A simulation is not a fill

No backtest can tell you about slippage, or whether you would actually have placed the order at 09:15.

Material changes to the rules on this page are recorded in the repository’s history and summarised here as they happen. Nothing on this page has changed since it was first published.