CYCLEPEDIAThe trading encyclopedia
IT
James Marsden Hurst 1924—2005

Chapter 7.2 — How to Select and Track

Total scanning and the stable (Hurst)

The weekend scan across every NYSE+AMEX issue: two visual tests, a volume floor, 80–140 candidates — and the 'stable' of 12–20 issues that keeps capital always ready.

On this page

Who this entry is for — You do not wait for a stock to "arrive" from the news: you sweep the entire board, every week, with two visual tests applied as fast as you can turn pages. Out comes the stable — and capital never sits idle.

Source: J. M. Hurst, The Profit Magic of Stock Transaction Timing, Prentice-Hall, 1970 — Chapter 7, The Total Scanning Concept, When You Should Use Alternative Scanning Methods, Take Advantage of the "Stable" Concept (pp. 115–121).

Prerequisites

Selection and tracking — why "traditional" selection is not enough for optimized profit.

The scan: two tests, at page-turning speed

In plain words — Saturday evening, the stack of weekly charts of every issue on the two major exchanges. Two questions per chart: is the cyclicality visible at a glance? has the underlying trend just turned? Two noes → next page.

More than 2,000 issues trade on the two major exchanges — far more than needed, with data readily available: you limit yourself to those. The book's ritual is weekly: the era's chart services (Mansfield for weeklies, Geiergraph for dailies) arrived by the weekend, and selection for Monday morning took "little more than half a day". The scan's two tests:

  1. dramatic visual evidence of dominant cyclicality;
  2. an overall trend that has just rounded a bottom (for longs; just bent over a top, for short candidates).

Whatever passes deserves the closer look: average volume ≥ 10,000 shares a week (~2,000 a day), and a dominant cycle whose edge or mid-band signal seems imminent → the name goes on the list. A finished scan typically yields 80–140 issues.

HURST 1970 · CH. 7 The selection funnel From the whole board to the stable: total scanning, screening, stable CYCLEPEDIA DIAGRAM — EMICICLO 2,000+ — every NYSE + AMEX issue the weekend scan, “as fast as you can turn pages” 80–140 — scan candidates volume ≥ 10,000 shares/week · edge or mid-band signal imminent ~20 — after screening volatility (the queen) + stability factors 12–20 — the stable analyses kept current, issues ready to go TIME REQUIRED ~half a day a week STABLE OUTPUT 4–5 signals a day Capital should never wait for an idea: the stable already has one saddled.
Chapter 7's funnel: total scanning, screening, stable.
Tap the four stages

The alternative scans

In plain words — List too short, or not convincing? Re-scan with other criteria — as long as they are grounded in the model. The book's two favourites: triangles in formation and the "nine-year sub-chart".

  • Triangles in formation (daily or weekly): you will rarely find more than a few, "but these few are excellent candidates" — Chapter 3 showed how well the model reads them.
  • The nine-year sub-chart: on the chart's multi-year inset you look for a steady, large decline (the more years the better) followed by no more than one year of rising prices, current price still low against the old highs; then, on the main chart, a 9–24 month horizontal range inside a flat envelope — and a break of its upper bound in the last week puts the stock on the list. The meaning: "the large smooth sweep of fundamental motivation has recently turned up", and cyclic action rides it "like frosting on a cake". The short-side analog is obvious.

Any scan variation works, on two verbatim conditions: cyclically readable and volatile stocks, criteria soundly based on price-motion model concepts.

The stable

Card — The stable's threefold purpose (12–20 issues)

  • 1. Centre your full attention on a specific group, until the next scan (or until signals pass by because you are fully invested elsewhere).
  • 2. Force yourself to keep charts and analyses current on your chosen issues.
  • 3. Assure ready-to-go issues at all times: idle time between trades cut to the minimum.

Properly handled, a 12–20 issue stable yields "a minimum of four or five action signals per day" — more than any one account can ride. Want fewer? Shrink the stable, and the work with it.

1970 and today — Mansfield and Geiergraph by air mail, plastic folders on a pegboard, charts redrawn by hand from ISL quarterly volumes: the logistics belong in a museum, the method does not. Today the visual scan across 2,000 charts is a screener loaded with your criteria, and the stable is a watchlist with current analyses. What does not change is the substance: model-based criteria, weekly review, complete analysis on a few issues instead of opinions on many.

Summary card

Element Rule
Universe Every issue on the two major exchanges (>2,000)
Scan tests Evident cyclicality + underlying trend just turned
Immediate filter Volume ≥ 10,000 shares/week; signal imminent
Typical yield 80–140 candidates → screening → stable of 12–20
Rhythm Weekly, ~half a day
Alternatives Forming triangles; the nine-year sub-chart
Suggest an edit My proposals