Contents · 11 / 11
Use the AI & ML screener presets
Updated 31 October 2026 · 3 min read · 14 Contents
Plain-language meaning
The AI & ML presets are eight ready-made screens built on stored model outputs: 20-session outperformance probability, edge over the base rate, volatility-expansion probability, three-year compounding probability, unusual-activity percentile, the market regime and the quant factor composite.
Why it is useful #
They let you filter a market by model evidence in one step, with the exact thresholds visible in each preset's rule.
Where it appears and the workflow #
Open Scan Center and choose the AI & ML chip (/screener/library?category=ai), or the AI & ML section of the Screener menu, which also links the ML shortlist, Unusual Activity and Factor Lab. Presets: ML Outperformance Leaders (probability at least 60%), ML Edge Over Base Rate (at least 10 points), Regime-Aware ML Momentum (risk-on at least 60%, probability at least 55%, price above its 50-day average), ML and Quant Factor Agreement (probability at least 60%, factor composite at least 70), ML Unusual Activity (percentile at least 95), Unusual Activity on Rising Volume (percentile at least 90, relative volume at least 2, up on the day), ML Volatility Expansion Likely (at least 65%) and ML Three-Year Compounders (at least 60%). On mobile the chip is labelled Machine learning & AI.
Calculation or source #
Each field reads the latest stored model output for the listing; nothing is scored on request. The models and their validation are described in Read the Radar ML shortlist and Read Machine Learning Insights.
Prerequisites, inputs, period and unit #
Every AI & ML field requires Max. Below Max the presets show a lock and an upgrade button, and running one is refused. Probabilities and the regime are percentages, edge is in percentage points, and the composite and percentile run from 0 to 100.
Worked example #
Illustrative only: ML Edge Over Base Rate keeps a listing whose probability is 64% when the base rate for its exchange is 52%, an edge of 12 points.
What high and low mean #
A higher probability or edge means stronger modelled evidence for that outcome, not a larger expected return.
Positive, negative and zero #
Edge over the base rate can be negative, meaning the model rates the listing below the average outcome.
Limitations and common mistakes #
Model outputs are estimates validated on past data, and thresholds narrow the list without making it more reliable. Combining several model fields does not add independent evidence.
Market-specific differences #
Each exchange has its own models and benchmark; outputs are not comparable across exchanges.
Suggested next steps #
Run ML Outperformance Leaders on one exchange, open one match's ML Insights tab to read its drivers, and compare the count with ML Edge Over Base Rate.
Educational use only #
This guide explains descriptive research evidence. It is not investment advice, a price prediction, a recommendation, a claim of predictive accuracy, or an instruction to buy, sell, rebalance or place an order.