How to read the NSE in four steps, in order.

The wrong answers, eliminated first.

Of the 2,254 stocks on NSE today, the four-step funnel below will eliminate roughly 2,234 of them before you ever look at a chart. The order matters — regime, sector, setup, stock — because each filter takes the next one's job seriously.

The cost of getting the order wrong shows up as a high false-positive rate on otherwise-clean charts. A clean cup-and-handle in IT Services during a regime where banking and metals are leading fails more often than it works — the 2024 H2 IT consolidation was full of base breakouts that became bull traps because the rotation had already left. The same breakout in a leading sector during an expanding regime usually holds. Same chart, different odds — determined by what you should have looked at first.

Today the funnel narrows like this: the NSE universe holds 2,254 stocks. Of those, 863 are in Stage 2. After applying a sector filter and a setup-score floor of six, the daily prop scan typically narrows to twenty-something names. That is the entire compression in one sentence; the rest of this page is how each filter is applied.

Step 1 — Read the regime.

As of 21 Aug 2026 the platform's regime label is BULL_RISING, held for 10 sessions (state-5 classification: NEUTRAL_UP). The four-row ladder below is the read that produced it.

The current state of the four rungs decides whether systematic Stage 2 buying has positive expectancy today. A full bull ladder (20.9% of the universe today sits in a full bull stack) is the cleanest backdrop. A ladder where every rung sits below 30% is the cleanest reason to do nothing.

The four rungs argue with each other in normal markets. The 20-day above the 200-day means short-term strength has yet to break the long trend. The 20-day below the 200-day means the opposite — a fading rally inside a deeper repair job. The cumulative A/D line is the underline: when it diverges from the headline index for two weeks, the index is lying about what most stocks are doing.

The platform also surfaces three leading flags built on top of the ladder: thrust (inactive today) when participation widens fast, breadth rebuild (inactive today) when the 50-day rung crosses back above 30% after a Bear stack, and watch bear exit (active today) when the rebuild has been holding for a week. Together they answer the “is this real?” question the ladder alone cannot.

If the ladder denies a constructive read, Step 2 is moot. Cash is a position. The historical record on this is documented across the breadth-signal walkthrough on the argument page.

Step 2 — Find the leading sector.

Within a constructive regime, leadership is never uniform. Capital concentrates in two or three sectors at a time. The signal is not today's Stage 2 share — it is the change in Stage 2 share over the trailing twenty days.

A sector at 40% Stage 2 share that was at 32% last month is rebuilding. A sector at 40% Stage 2 share that was at 48% last month is breaking down. Same level, opposite stories. The 20-day delta separates them.

After Step 2 you have a list of one to three sector names. Not a forecast about which sector will lead next year — a list of where capital is concentrating today. The sector tracker shows the full table; for the slow-cycle context behind the rotation, see how sector rotation works on NSE.

If today's top sector is one you do not normally trade — e.g. PSU Banks for a software-cap-only investor — the workflow does not force the trade. Substitute the highest-Δ20d sector you do trade, and accept that your filter is now stricter than the platform's. The wrong move is to chase an unfamiliar leader; the right move is to skip the day.

Industry-level cuts beat sector-level cuts on NSE. “Healthcare” at 28% Stage 2 share might hide Pharmaceuticals at 45% and Hospitals at 15% — the sector aggregate masks the concentration. Once the sector shortlist is in hand, drill one level further on industries to see which sub-themes are actually carrying the leadership.

Step 3 — Filter to qualifying setups.

Step 3 narrows from sectors to charts. The filter is the setup score, which compresses pullback depth, sector strength, trend maturity, and price stability into one zero-to-ten number.

The pipeline is mechanical. From the 863 Stage 2 stocks today, restrict to the sectors that survived Step 2. From the survivors, keep those whose setup score sits above six. The output is a candidate set that fits on one screen — usually twenty-something names site-wide.

Of today's Stage 2 cohort, 108 names carry a prime setup score (eight or above) and 277 carry a strong score (six to eight). Prime scores are scarce by design — when they are not scarce the score has stopped saying what it should.

The three setup families — pullback to support, base breakout, trend continuation — each have a different signature. A pullback rests on a rising 150-day MA at lower volume than the prior advance; a base breakout clears multi-week resistance on a volume thrust; a trend continuation extends an already-extended move with tight price action. On NSE, watch the prior-advance volume bar against the 20-day average, not the 50-day — the 50-day is contaminated by F&O expiry days when delivery volume drops 30–40% on the cash leg. The empirical hit rates differ enough that knowing which family a setup belongs to matters more than knowing its absolute score.

Across 3,99,423 historical Stage 2 entries since 2024, the prime band (setup_score ≥ 8) hit a positive T+30 return 51% of the time, vs 45% for the mid band (5–6). The score is not a prediction; it is a base rate gating which trades earn the slippage and the attention.

Several disqualifications are mechanical and live outside the structural read. Any stock on the ASM or T2T surveillance segments is skipped — the higher margin requirements and intraday restrictions break the assumptions the setup score was tuned on. Anything on the F&O ban list is skipped for the same reason. Anything with results inside the next five sessions is skipped because gap risk dominates structural risk. None of these are commentary on the underlying name — they are operational filters that protect the score from situations it was not designed to read.

Today the prop scan returned 309 qualifying names — vs a 30-day average of 300.0. A worklist materially below average is usually the workflow's politest way of saying “tomorrow.”

Step 3 is where most discretionary traders quietly cheat — they raise the score floor when they have positions and lower it when they don't. The point of writing the threshold down is that you find out, at the end of the year, what the threshold actually was. With it logged, Step 4 is mechanical.

Step 4 — Only now look at individual stocks.

With regime, sector, and setup confirmed, stock selection becomes structural rather than predictive. The trader is no longer asking “will this work?” — they are asking “which of the surviving names has the cleanest structure?”

Pick a leading sector. Inside it, sort the Stage 2 names by setup score and relative posture. Discard anything where weeks in stage is below five or above twenty-five — Stage 2 trends younger than five weeks have not earned the trust of two MAs yet; older than twenty-five risks a topping print. Discard anything on the ASM or T2T lists. The five to ten names left are today's worklist.

Step 4 applied to Real Estate.

Today the leading sector by 20-day Stage 2 delta is Real Estate at +9.9pp. Inside it, the three RS-strongest Stage 2 names are below. The workflow does not say “buy these” — it says “if you are deploying capital today, these are the names that survived the four-step filter, ranked by structure.”

The March 2020 walkthrough below shows the same workflow run backwards over an actual breadth crash. The interesting row is the last: re-entry was not on 27 March (the absolute Nifty low was twenty-two days later), nor on 15 April (the closing low), but on 20 May — when the 50-day rung crossed back above its 30% threshold. The bottom did not ring a bell; the workflow did.

What the workflow cannot tell you.

The four steps narrow the universe to a list. They do not tell you how much of each name to own, where to place the stop, or when in the session to fill the order. Those are extrinsic to the structural read.

On 24 January 2023 the workflow would have flagged ADANIENT as a fading Stage 2 — a regime still constructive, a leading sector behind it, a setup score still respectable. The Hindenburg report dropped that morning and the chart lost roughly two-thirds of its value over the next ten sessions. The 30-week MA broke before the bottom, which is what the platform would have surfaced, but the speed of the move meant that traders who had sized 8% into one concentrated name lost more on that one chart than they had made on the prior six.

The boundary of the workflow's claim is honest. It tells you which names have a backdrop that supports them. It does not tell you that no name has tail risk. A 2% position in a Hindenburg-style event costs 1.4% of capital. An 8% position costs 5.6% — same chart, four times the damage, and the only difference was the trader. SEBI's subsequent surveillance action on the Adani group stocks added margin and intraday curbs — even if a trader had wanted to add on the way down, the operational filter would have stopped them. The structural read and the operational read agreed.

Three things the workflow does not address:

1. Entry timing within the day. The setup score is end-of-day; intraday entries are a separate craft. A common-sense default: wait for the first hour's range to close, enter on the second hour's direction.
2. Position sizing. A 2% portfolio cap on any single concentrated name. Higher caps on liquid large-caps if the regime is solidly bullish. Lower caps on F&O-banned or recently-ASM names — full stop.
3. Stop placement. A common default is the rising 150-day MA, or the most recent swing low if that sits inside the MA. The platform does not place this for you; the trader does.

The workflow tells you what to consider. The trader decides how much to risk on the consideration.

The five-minute morning checklist.

Once the four steps are internalised, the daily routine compresses to five short reads. No prose, no judgement calls — read each line, answer it, act.

This week's tape produced 38 fresh 1→2 transitions and 32 new 2→3 distributions — the second number is the one to check first, because a 2→3 in your portfolio is a tighten-or-exit before the regime even argues.

The first three steps take ninety seconds combined. The fourth takes two minutes. The fifth takes one. The point of the checklist is not the time saved — it is the decisions removed.

When the workflow tells you nothing — and you should listen.

Some weeks the workflow returns no trade. Step 2 finds no leading sector; Step 3 finds no setup above six; Step 4 has nothing to rank. That is not a failure — that is the system working.

The cleanest example on NSE history is Vodafone Idea, ticker IDEA. From the post-merger peak in September 2018 through late 2023, the chart spent more than five years in Stage 4. The 150-day MA never turned up in any meaningful way. Every six months there was a rumour-fuelled bounce — a recapitalisation deal, an AGR relief, a tariff hike — and every bounce failed inside three weeks. A trader running this workflow would have skipped IDEA roughly two hundred and sixty times across that window.

The trick is not picking up the bottom; it is to wait until the stock breaks out of a base, and the moving average has turned up.

— Stan Weinstein — Secrets for Profiting in Bull and Bear Markets, 1988

The cross-references that build this discipline elsewhere: Stage 3 distribution on /stage-analysis for IRCTC's 2021 narrative top, LIC's failed Stage 1 on /setups for the IPO-as-non-base pattern. Each is a different shape of the same lesson — when the workflow has nothing to do, the cheapest position is no position.

The workflow fails for a second reason worth naming: regime changes that take place over a single weekend. A Sunday-night macro shock, a budget surprise, a global event during NSE off-hours — the breadth ladder is end-of-day, so the trader running this routine cannot adjust until the next print. The defence is the same as in Ch6: position sizing, not workflow sophistication. The framework is structural; the protection against gap risk is operational.

The honest closer: the workflow narrows the field. Whether you trade what survives is still your decision.