The "above-MAs + relative-strength + volume" daily scan is the workhorse first-pass filter of trend-following swing trading: a nightly automated screen that reduces a universe of several thousand listed stocks down to a manageable shortlist (typically a few dozen) by demanding three things at once — price stacked above its key moving averages (a confirmed uptrend), strong relative price performance versus the broad market (the stock is a leader, not a laggard), and recent volume that betrays institutional participation. Its core tension: the filter is deliberately trend-biased and momentum-biased, so it surfaces exactly the kind of stock that tends to keep working in bull regimes — and exactly the kind that gets whipsawed and gap-down-killed when the market environment turns. The scan is a candidate-generator, not a buy signal; it tells you where to look, not when to buy.

How it's formed (the three filter blocks)

The scan is a boolean AND of three condition groups. The most widely-copied template is Mark Minervini's Trend Template (from Trade Like a Stock Market Wizard), which formalizes the first two blocks:

Block 1 — Moving-average / trend structure (Minervini's 8 conditions): 1. Price > 50-day MA 2. Price > 150-day MA and > 200-day MA 3. 150-day MA > 200-day MA 4. 200-day MA trending up for at least 1 month (Minervini's book says preferably 4–5 months) 5. 50-day MA > both the 150-day and 200-day MA 6. Price ≥ 30% above its 52-week low (the canonical figure in Trade Like a Stock Market Wizard; many leaders are 100%+ above) 7. Price within 25% of its 52-week high (closer is better) 8. Relative Strength rating ≥ 70 (Minervini prefers 80–90+)

This MA stack ("price > 50 > 150 > 200, all rising") is the algebraic definition of a Stage-2 advance in Weinstein/O'Neil stage terminology.

Block 2 — Relative Strength (RS). Note the critical distinction: this is the IBD-style RS rating (a 1–99 cross-sectional percentile of 12-month price performance vs all stocks), not Wilder's RSI oscillator. IBD's rating weights recent quarters more heavily — a widely-replicated open-source formula (Skyte) doubles the most recent 63-day return: 2×(close/close 63d ago) + (close/126d) + (close/189d) + (close/252d), i.e. ~40% weight on the latest quarter and ~20% on each prior quarter over 252 trading days. That raw score is then converted to a 1–99 percentile rank across the whole universe, so an RS rating of 90 means the stock outperformed 90% of stocks (practitioner versions typically benchmark the universe against SPY/the S&P 500). Scans typically demand RS ≥ 70–80, with leaders at 90+.

Block 3 — Volume. Two distinct uses: (a) a liquidity floor (e.g. 50-day average dollar-volume above a minimum, and price above a minimum, to exclude untradeable thin names); and (b) accumulation evidence — recent up-days printing volume well above the 50-day average. The O'Neil-derived convention is a breakout/up-day on volume 40–50% above average as the minimum "footprint of big money," with 50–200% above average being unambiguous institutional buying.

How it's used in practice

The scan runs after the close, on adjusted EOD data, producing the next day's watchlist. The standard workflow is a funnel:

1. Quantitative pass (the scan itself): Blocks 1–3 cut ~3,000+ names to commonly 20–100 candidates. In strong markets the list swells; in weak markets it can shrink to single digits — the count itself is a breadth signal. 2. Manual chart triage: the trader eyeballs each survivor for a recognizable setup — a tight base/consolidation (VCP, flat base, cup-with-handle), proximity to a pivot/breakout point, and orderly (not climactic) prior advance. The scan finds trending leaders; the chart read finds low-risk entry points within them. 3. Trigger, not membership: appearing on the scan is necessary but never sufficient. The actual entry is a separate event the next day(s) — typically a breakout through the pivot on expanding volume (the Block-3 accumulation threshold re-applied at the entry bar), with the stop placed below the base. Exact entry/stop/target mechanics live in the swing-execution nodes, not here.

Practical refinements masters key on: requiring the broad market itself to be in a confirmed uptrend (run the same MA logic on SPY/QQQ) before acting on any candidate; sorting survivors by RS rating descending to triage the strongest first; adding an earnings-date filter to avoid buying into a report; and watching for pocket pivots (an up-day whose volume exceeds every down-day's volume of the prior 10 sessions) as an early within-base accumulation tell.

Adoption, debate & evidence

This screen is ubiquitous in growth/momentum swing-trading communities — it is essentially the codified O'Neil/CAN SLIM and Minervini/SEPA methodology, and every major retail screener (Finviz, MarketSmith, TradingView, Deepvue, ChartMill) ships a near-identical "Trend Template" preset. That popularity is double-edged: a crowded, well-known filter means many traders are watching the same names and pivots, which can sharpen breakouts but also produce more failed/"shakeout" breakouts.

On evidence, separate two claims honestly:

• The RS / momentum leg has genuine academic backing. Cross-sectional momentum — buying recent 3–12-month relative winners — is one of the most robust documented anomalies (Jegadeesh & Titman 1993, ~1%/month gross over the subsequent year; replicated internationally by Rouwenhorst and embedded in Carhart's four-factor model). The relative-strength ranking the scan uses is the practitioner expression of this factor.
• The composite swing-screen's published win rate is mostly folklore. Minervini's documented personal track record (e.g. his 1997 U.S. Investing Championship win) is real but is one discretionary trader, not a base rate for the screen. There is no robust peer-reviewed hit-rate for "this exact AND-filter as a mechanical strategy." Treat any specific advertised win-rate as unverified, and note that momentum strategies suffer severe crash risk in regime reversals (the factor's documented weakness).

Strengths & limitations

Strengths: it enforces trading with the dominant trend and concentrates attention on leaders; it is fast, objective, and reproducible nightly; the MA stack is an unambiguous, non-discretionary trend definition; and it naturally throttles activity (fewer candidates) when the market is unfriendly.

Limitations & failure modes:

• Regime dependence is the whole story. The scan is a long-only bull-market tool. In bear/choppy regimes it produces few or low-quality candidates and those that pass are most prone to failed breakouts. Always gate it on market health.
• Lagging by construction. Requiring price above all rising MAs and a high RS rating means the stock has already moved a lot — it is mid-trend, not early. You buy strength, accepting later entries.
• The #1 misuse: treating a scan hit as a buy signal. It is a candidate. Buying everything the scan returns, with no base/pivot/volume trigger and no stop, is the classic way the screen "stops working."
• Look-alike trap: confusing the IBD RS rating with the RSI oscillator. They share a name and almost nothing else; substituting RSI < 30 ("oversold") inverts the screen's intent from buying leaders to catching falling knives.
• Whole-numbers traps: the 25%/70/40% thresholds are conventions, not laws — over-tuning them invites curve-fitting.

System relevance

For Augustus (the swing/short-term setup agent), this node defines the candidate-sourcing front end. Augustus should treat scan membership as a precondition that opens evaluation — never as a setup. The hard caveat to encode: (1) require an independent broad-market-health check before promoting any candidate; (2) require a real entry trigger (base + pivot + volume expansion) from the swing-execution nodes before signaling; and (3) flag the regime — in non-trending markets, down-weight or suspend the screen. Cross-link: the Relative Strength definition node, the moving-average trend nodes, the volume/accumulation nodes, and the VCP / base-breakout setup nodes for the trigger layer.

Sources

• Mark Minervini, Trade Like a Stock Market Wizard — the 8-point Trend Template (summarized via ChartMill and Deepvue documentation of the template criteria).
• Deepvue / FinerMarketPoints — Minervini Trend Template criteria and RS ≥ 70/90 guidance.
• IBD-style RS Rating methodology — 1–99 percentile, 252-day lookback, recent-quarter weighting (journalplus.co glossary; Skyte/relative-strength GitHub reference implementation).
• William O'Neil / CAN SLIM volume convention — breakout volume 40–50%+ above average ("footprints of big money"); pocket-pivot definition (TraderLion, SharePlanner volume analysis).
• Jegadeesh & Titman (1993) and 30-years-later reviews (Springer FMPM; AlphaArchitect) — academic evidence for cross-sectional momentum/relative strength, and its crash-risk caveat.

Flagged disputes: the composite screen has no robust peer-reviewed win rate (only the momentum leg does); advertised hit-rates and the exact threshold values are conventions, not validated constants.