Immediate definition
In plain words — Hurst does not try to study everything every day. He narrows the universe in stages, analyses a small list deeply, and defines the conditions to monitor in advance.
Chapter 7 of The Profit Magic of Stock Transaction Timing completes two of the historical system's four pillars: which stocks to observe and how to notice the signal when it arrives. The aim is not to find the universally “best” stock. It is to focus time and data on instruments that, within the book's model, appear readable, mobile, and close to a possible decision.
The four-stage funnel
Total scanning rapidly reviews charts across the universe and retains candidates with visible cyclicality and a setup near a reversal. Screening then compares volatility, size, short interest, and selected fundamental data. This produces the stable, a group of 12–20 stocks to analyse and update. Finally, tracking links each stock to predefined conditions: targeted monitoring rather than purposeless continuous watching.
Why Hurst rejects “late” selection
News, recommendations, and most-active lists may draw attention to a stock only after a conspicuous move. In Hurst's view, that often means meeting it late in its cycle and waiting for another useful point. His scan starts from the model instead and looks for two properties together: potential movement and proximity to a signal defined by the rules.
That is an operating rationale, not statistical evidence. A filter can appear effective because it is refined on examples already observed. The more criteria are tried, the greater the chance of selecting a lucky configuration. A modern version should therefore define universe, frequency, and thresholds before testing, then examine rejected candidates and out-of-sample periods.
The correct present-day translation
The mailed chart books, binders, and 1970 “Personal Ticker” have become screeners, watchlists, and alerts. Their functions remain distinct:
- the scan generates candidates, not orders;
- screening reduces workload but may introduce bias;
- the short list receives complete analysis and fresh data;
- the alert reports a condition that still requires verification and execution.
Hurst's list sizes are historical, not universal constants. Today's universe, liquidity, venues, and personal capacity may require different numbers. His goal of maximising time invested is not universal either: holding cash can be a valid outcome when no candidate clears the rules.
The actual evidence boundary
Looking only at stocks that survive the funnel hides information about false negatives. Changing criteria after every failure makes the process impossible to falsify. A serious audit must retain each scan version, every inclusion and exclusion, costs, and the data known at the time. Only then can one distinguish useful work organisation from a repeatable edge.
The source-checked label certifies fidelity to the 1970 chapter, not the
current effectiveness of the funnel.
Sources
- J. M. Hurst, The Profit Magic of Stock Transaction Timing, Prentice-Hall, 1970, ch. 7, pp. 114–122 — scanning, screening, the stable, and tracking.
- Campbell R. Harvey, Yan Liu, and Heqing Zhu, … and the Cross-Section of Expected Returns, NBER Working Paper 20592, 2014 — multiple testing and evidence thresholds; not a test of Hurst's method.
Related entries
- Total scanning and the stable — building the short list.
- Volatility and stability screening — criteria and risks.
- Price tracking — from condition to alert.
- Trading by logic — the Chapter 8 application.