Case Screw and Bolt (1968)

Who this entry is for — Every technique in the book, used once, at the right moment, on a real stock: scan, screening, envelope, inverse, nest of lows, VTL, half-span for potential and risk, trailing, final blowoff. The whole method in 57 days.

Source: J. M. Hurst, The Profit Magic of Stock Transaction Timing, Prentice-Hall, 1970 — Chapter 8, Selecting the Issue → A Model Transaction (pp. 127–135, Figs. VIII-2/VIII-5).

Prerequisites

The market state of 1 November 1968: a ~one-month long window. Now for the stock.

Selection and screening (1 November)

The weekly-chart scan brings Screw and Bolt Corporation (NYSE) to attention, among some twenty others: fifteen months between 8 and 14, a late-September breakout to 18, a five-week pullback to 14–15; pronounced cyclicality; average volume 30,000 shares/week. Chapter 7's screening approves across the board: above-average %-per-inch, small capitalization (1,663,000 shares), B rating, P/E of 20, and the best item — cumulative four-quarter earnings reversing course: 0.83 · 0.66 · 0.41 · 0.43 · 0.51 · 0.72.

The analysis: envelope, inverse, nest

In plain words — One envelope on the weekly finds three nested periodicities; the inverse confirms them; the durations projected forward draw a "nest of lows" right ahead. The state table says: the signal is very near.

From the weekly chart (Fig. VIII-3): dominant component with two samples of 20 and 16 weeks → 18-week average; within it, segments of 7–12 (9.2 average) and 4–7 (5.2 average ≈ the 6.5 nominal); centre line hard up. Given the wide variation, the check with a 19-week inverse (18 would be ideal, but even → interpolation) confirms the observations. The three series' durations, projected from their last lows, produce the overlapping shaded zones of the nest of lows: from now to ~3 weeks out. The daily chart adds the 9.9-day component (8–11).

The state table: long sum hard up; 18-week at 14, down but due to bottom; 9.2 at 4, up but topping; 5.2 at 4, down but bottoming; 9.9-day at 4, up but topping. "Three up and two down", near equilibrium — but at the nest all five will be upside together: the buy is near.

Potential and risk, before entering

Card — The half-span sum (before the order)

If the nest reverses prices, the 10-week average must turn there — and for it to turn, the mean weekly price 5 weeks ahead must exceed the one 6 weeks back: critical price 17½.
Hence: from ~14½ to 17½ in 5 weeks = half the move → total to 20½.
Potential ~6 points against ~½-point risk plus commissions (first trailing level at 14): with the expected 80–90% accuracy, odds too good to pass up.

Trading-cycle choice: the 18-week (funds ≤ $100,000, limited time: one stock to follow); VTL from the 9.2-week component.

The trade, day by day

Date	Event
Fri 1 Nov	Full analysis; VTL 1 drawn into the nest
Mon 4 Nov	A shorter component's top yields VTL 2, steeper
Wed 6 Nov	Open above VTL 2 → buy at 15½ (mid-range); TLL-1 at 14⅛
26 Nov	The thrust confirms the ~25-day component's low → zone "A" for the next one
Fri 6 Dec	The 10-week turns up: extrapolation → crossing at 18⅛; move 14½→18⅛ = half → target 22⅛ ±10% → zone 21–23
19 Dec	Low in "A" confirmed → VTL 3
~23 Dec	Prices inside the zone: take profit, or ever-steeper lines? The ride is chosen
31 Dec	The 9.9-day low in "B" → VTL 4; but prices burst past the zone to 24¾: blowoff
2 Jan 1969	VTL 5 on the last two days' lows; the break fires at 23⅞ → sold

Result: +49.7% net in 57 days — 318% per year (simple). The razor-steep final lines are not standard practice: three conditions together dictated them — prices beyond the prediction zone, beyond even the extrapolated envelope, and cyclic time running out (nine weeks on the 18-week cycle). "Such a blowoff is normally followed by an equally rapid price contraction."

Warning — The comparison closing the case is the book's moral: the same stock, in the short-interval experiment, returned 3.2% net in one day = 1,168% per year against the two-month trade's 318%. But at the price of many other stocks analysed and ready. Chapter 1's arithmetic, as always.