In one sentence — Entry is only half the plan: Chapter 5 states in advance where to admit an error, how to protect a favorable move, and what new analysis to perform after the exit.
Why is an exit plan necessary?
Price can move against a hypothesis even after a carefully constructed entry. Without a boundary decided beforehand, a small loss can become an emotional decision. Without a protection rule, an open profit may be mistaken for a result already secured.
Hurst applies to exits the same graphic logic used for purchases. Levels derive from the observed component, its lows, and valid trend lines. This is a historical reconstruction of a 1970 method: it makes decisions more explicit but does not remove gaps, slippage, analytical errors, or sudden market changes.
1. State where the idea no longer holds
The cyclic cut-loss uses the shortest component’s lowest low before the upside VTL break as its reference. If price returns below that boundary, the expected cluster of lows may not have completed. Within the model, exiting means suspending an invalidated hypothesis; it does not predict the next low.
The analytical level and the exchange order are not the same thing. A triggered stop commonly becomes a market order, and the actual fill may differ from the stated level, particularly during gaps or fast moves. Rules also vary by broker.
2. Protect without chasing every fluctuation
The trailing loss level is raised after confirmed lows of a component shorter than the selected trading cycle. It does not follow every intraday low and cannot promise to preserve the maximum open profit. A valid uptrend line then supplies Hurst with a separate graphic profit-taking signal when the rising structure deteriorates.
The distinction matters. The initial stop limits exposure to a wrong idea; trailing manages a position still considered coherent; profit-taking closes it when a condition defined by the method appears.
3. Handle envelopes and the short side correctly
The non-real-time envelope redraws extremes in pseudo-time to make the geometry more regular. It may clarify the channel retrospectively, but the latest extreme needs later data to be confirmed. A clearer ex-post image is not a signal that was necessarily available at the time.
Short selling is not the automatic consequence of selling a long position. Hurst requires a fresh bearish context. Modern implementation adds securities borrowing, margin, possible recalls, and theoretically unbounded loss if price keeps rising.
How should the Gruen case be read?
The book links an entry around 7½, initial invalidation near 7⅛, trailing, and an exit around 12⅜; it then reports a decline toward 9 over the next five weeks. The sequence is valuable because it shows the rules in one complete case. It does not provide an error frequency, a benchmark comparison, or an out-of-sample test. Annualized percentages derived from that short move are not estimates of repeatable return.
Evidence boundary — Chapter 5 supplies disciplined language for invalidation and exit. It does not prove fills at the selected price, stable cycles, or that the Gruen case represents other securities and periods.
Sources
- J. M. Hurst, The Profit Magic of Stock Transaction Timing, Prentice-Hall, 1970, Ch. 5, pp. 86–96 — cut-loss, trailing loss levels, profit-taking, non-real-time envelopes, and short selling.
- U.S. Securities and Exchange Commission, Stop, Stop-Limit, and Trailing Stop Orders: Investor Bulletin, updated 2026 — triggering, execution price, and distinctions among stop orders.
- FINRA, Short Interest: What It Is, What It Is Not — borrowing, margin accounts, and loss when price rises.
Related entries
- Cyclic cut-loss — initial invalidation
- Trailing and profit-taking — protection and closure
- Non-real-time envelope — transformation and right edge
- Short selling — historical signal and operating risks