lev-os Analyzing Equity Market Breadth
When To Use
- Assessing whether a rally or selloff is broad-based or narrow (e.g., concentrated in mega-cap tech)
- Identifying breadth divergences that precede trend reversals (index making new highs while fewer stocks participate)
- Evaluating sector rotation patterns and capital flow shifts across market segments
- Screening for deteriorating internals beneath a stable headline index level
- Supporting risk-on / risk-off allocation decisions with participation data
Inputs To Gather
- Advance-decline data: Daily advancers vs. decliners for the target index or exchange (NYSE, NASDAQ, or composite) [VERIFY exchange-specific data source and reporting cutoff time]
- Cumulative A/D line: Running sum of net advances to establish trend direction
- New 52-week highs and lows: Daily counts, ideally broken out by exchange
- Sector-level performance: Returns and A/D ratios for each GICS sector (or equivalent classification)
- Volume breadth: Up-volume vs. down-volume ratios to weight participation by capital commitment
- Index composition details: Number of constituents, weighting scheme (cap-weighted, equal-weighted), top-N concentration
- Time horizon: Intraday, daily, weekly, or multi-week lookback depending on the analysis purpose
Workflow
Establish baseline context
- Identify the index or universe under analysis (S&P 500, Russell 2000, NASDAQ Composite, etc.)
- Note the weighting methodology — cap-weighted indices can mask breadth weakness when a handful of large names drive returns
- Record the analysis date range and comparison periods
Compute advance-decline metrics
- Calculate daily net advances (advancers minus decliners) and the cumulative A/D line
- Compare A/D line trend to the price trend of the headline index — flag divergences where the index rises but the A/D line flattens or declines
- Compute the A/D ratio (advancers / decliners) and note readings above 2:1 (strong breadth) or below 0.5:1 (weak breadth)
Analyze new high / new low data
- Track the daily count of new 52-week highs minus new 52-week lows (NH-NL differential)
- A rising index with a shrinking NH-NL differential signals narrowing leadership
- Sustained negative NH-NL readings during an index advance are a classic bearish divergence signal
Assess sector participation
- Determine how many of the 11 GICS sectors are trading above their 50-day and 200-day moving averages
- Identify which sectors are leading vs. lagging — defensive sector leadership (utilities, staples, healthcare) during a rally often signals fragile breadth
- Flag any single sector contributing a disproportionate share of index returns
Evaluate volume breadth
- Compare up-volume to down-volume; readings above 90% up-volume ("breadth thrust") are historically bullish confirmation signals [VERIFY specific threshold definitions per the indicator variant used]
- Persistent low up-volume ratios during price advances suggest lack of conviction
Synthesize and classify breadth regime
- Classify the current environment: broad participation, narrowing breadth, divergence, or breadth thrust
- Identify the most actionable signal (e.g., "A/D line divergence persisting for 3+ weeks while index is within 1% of highs")
- Note any historical analogs if relevant
Output
- Breadth Summary Table: Date range, index, A/D ratio, cumulative A/D line trend, NH-NL differential, sectors above 50-DMA / 200-DMA, up-volume ratio
- Divergence Flags: Explicit callouts where breadth metrics conflict with headline index direction, with severity rating (early, developing, confirmed)
- Sector Participation Map: Heatmap or ranked list showing sector-level breadth contribution
- Signal Classification: Current breadth regime label with supporting evidence
- Actionable Implications: What the breadth picture suggests for position sizing, hedging, or sector rotation — framed for trading and execution desks
Quality Checks
- Confirm data covers the full trading session (not partial or pre-market only) [VERIFY data source timestamp conventions]
- Cross-check A/D data against at least one independent source to catch reporting errors
- Ensure new high/low counts use a consistent lookback window (52-week standard; some sources use shorter periods)
- Verify that sector breakdowns use the same index universe — mixing S&P 500 sector data with Russell 2000 index-level data produces misleading conclusions
- Flag any days with unusual market structure events (half-days, index rebalances, options expiration) that distort breadth readings
- Do not present breadth signals as predictive with certainty — note historical base rates and false-positive frequency where available