See whether higher-rated stocks and stronger signals outperform the market.
ShowPreliminary — tracking since Jul 2026
Performance at a Glance
Are scored stocks beating the VN30 Index?
vs VN30 Index
—
Beat VN30 Index
—
Avg. Stock Return
—
VN30 Index Return
—
What Has Worked So Far?
Best and worst patterns on completed exits, corrected results first.
Recipe Explorer
Every recipe with its sample size and outcome. Thin patterns are hidden by default.
Details · how recipes are built
Recipes pair one condition with an entry zone, a priority label or a condition from a different source: VS, TV, R, FF, S · Aggregate and, on the corrected (v2) data, individual S profiles. Two S profiles are never paired with each other.
Original (v1) columns count overlapping daily observations; patterns under 5 observations are hidden and under 10 are not ranked. Corrected columns count non-overlapping outcomes. Recipe identities are condition-based keys (for example sp:four_m|tv:buy) and never change with display names.
Checking snapshot history…
Current Stocks Matching Winning Recipes
Stocks matching a qualifying recipe today. Not a recommendation.
Details · qualifying criteria
A recipe qualifies on the corrected criteria: at least 8 non-overlapping outcomes across 3 or more stocks, a positive median raw return, a Wilson lower bound of at least 30% on the objective, no single outcome driving the gains, and the corporate-action screen available. The objective is a ≥10% raw return, with >0% as the fallback. Buy-only, mixed-signal and Sell-only recipes qualify on the same terms.
Each stock shows its single best qualifying recipe. Matching several recipes or several S profiles is not independent confirmation, and their sample counts are never added together.
Current Stocks Matching Risk Patterns
Stocks matching a corrected risk pattern today.
Details · risk criteria
A risk pattern has at least 8 non-overlapping outcomes, a negative median raw return and a loss-rate Wilson lower bound of at least 30%. Risk patterns and effective Sell warnings are reported separately from positive-return patterns.
Do Higher Ratings Perform Better?
Outperformance vs VN30 Index by rating band.
What Drives Performance?
Results by entry, quality and evidence score.
Signal Performance
Original (observation-level) results by signal. S profiles are explained in the S profile guide.
What each S profile screens for, and what is known about it.
Open the S profile guide
Does Confirmation Help?
Agreement and conflict between sources.
By Number of Positive Sources (agreement, not coverage)
Useful Combinations
Foreign Flow FF
Standalone signal, not part of the score.
Details · foreign flow
Net foreign buying or selling measured against each ticker's own baseline. Shown alone, as confirmation of another positive signal, and in conflict with one.
Foreign Flow Alone
Confirmations — FF buying alongside another positive signal
Conflicts — a positive signal while foreign money is leaving
Performance by Market Condition
Results in rising, flat and falling markets.
Performance Over Time
Average outperformance vs VN30 Index by exit week.
Recent Results
Latest completed observations.
Top Performers
Highest return vs VN30 Index among completed observations.
Stocks Beating VN30 Index
Every scored stock whose return exceeded the VN30 Index, with its full signal set.
Coverage Gaps
The biggest completed moves, and whether the model had data on them at entry.
Details · how to read coverage gaps
This describes what was missed and why. It is not a prediction, and not a claim that any pattern here repeats.
Each row is one independent price run, not one scoring day.
Methodology & Definitions
Applies to every view: how to read the numbers, and what these results do and don't prove.
How to Read This Page
Beat VN30 Index %
% of stocks in the group whose return exceeded the VN30 Index return over the same period. 100% = every stock beat the index; 50% = half did.
Why "Beat VN30 Index %" and the "Stocks Beating VN30 Index" table don't match
The glance card counts every scoring observation — the same ticker can be scored (and get a completed return) on many different days, and each of those days counts separately toward the %. It's a hit rate across all scoring events, not a count of stocks. The "Stocks Beating VN30 Index" table instead shows one row per ticker — only its most recent scoring day — and only lists a ticker if that latest observation beat the index. A stock that beat VN30 on an older scoring day but is behind it now (or vice versa) won't line up between the two.
vs VN30 Index
Each stock's return minus the VN30 Index return over the exact same dates. A stock down −5% can still show positive vs VN30 Index if the index fell −8%.
Avg. Return
Raw average price return, not market-adjusted. High return in a rising market is less meaningful than the same return when VN30 was flat.
Score Ratings
Composite score (0–100) from entry quality, fundamentals, and signal evidence. Higher ratings indicate stronger alignment across scoring components.
Signals
VS Vietstock analyst rating ·
TV TradingView analyst rating ·
S Screener technical signal ·
R Research recommendation ·
FF Foreign fund flow.
A stock can carry multiple signals simultaneously.
Foreign Flow (FF)
Net foreign buying or selling, measured against that ticker's own normal activity rather than an absolute VND amount — so unusual interest in a mid-cap registers the same as in a large-cap. A flow must clear 1bn VND before anything fires, then it is graded at 1.5× and 3× the ticker's 20-day baseline (computed over active-flow days only). Consecutive same-direction days (3+) are noted separately as a streak. See the Foreign Flow section for standalone, confirmation, and conflict performance.
Foreign Flow & scoring
Foreign flow is not part of the score (three_score_v1 is unchanged). It's tracked as a standalone daily signal so its predictive value — alone, as confirmation, or as conflict — can be measured on this page first. Whether it earns scoring weight is a decision for later, based on what the Foreign Flow section shows.
Exit Date
The date the return is measured to. Counted as exactly 21 (1m), 63 (3m), or 126 (6m) VN30 trading sessions after the score date — skipping weekends and Vietnamese public holidays. All stocks scored on the same day share the same exit date, so returns are always measured over an identical window.
Horizons
1m = 21 sessions · 3m = 63 sessions · 6m = 126 sessions from score date. Trading sessions keep comparisons consistent across weekends and holidays.
Sample size
Results with n < 10 are not statistically reliable. Model live since July 2026 — 1m exits are most populated; 3m and 6m accumulate through late 2026.
About These Results
Tracking since
July 2026
Benchmark
VN30 Index, same entry and exit dates as each stock
Horizons
1, 3 and 6-month performance is measured after approximately 21, 63 and 126 trading sessions respectively. This keeps comparisons consistent across weekends and market holidays.
Completed results
—
Results are preliminary while history accumulates. Outperformance is the stock's return
minus the VN30 Index return over the exact same dates — a negative return can still outperform
if the whole market fell further. Past performance does not guarantee future results.