In plain words — Short selling means selling borrowed securities and buying them back later: a decline can produce a gain, but loss can keep growing if price rises.
It is not the automatic continuation of a long trade
At the end of Chapter 5 Hurst extends to the downside the graphic principles already used for buying and selling. He makes an essential distinction: closing a long position does not mean automatically selling the same security short. The exit may reflect a brief pause inside a higher structure that is still rising.
To seek a short, the method requires fresh selection. Longer components must produce an underlying sum judged to be falling, while market averages provide directional context. Symmetry applies to trend lines, envelopes, and signals, not to the economic risk of the position.
Hurst's historical sequence
The process starts with a security whose sum of higher components is assessed as falling. Near a forecast high zone, a valid uptrend line may supply the bearish counterpart of the VTL used for buying. Envelopes and shorter components help define target, invalidation, and trailing protection.
To close the short, the method seeks upside confirmation consistent with the selected component or applies its planned protection level. Operationally this closure is a buy to cover: the security is repurchased and returned to the lender.
The book presents this construction as a logical extension of the model. The chapter does not, however, publish a sample of completed shorts with a distribution of outcomes, borrow costs, recalls, or squeeze periods. The entry should therefore be read as a historical description, not evidence of efficacy.
Why is the risk not symmetrical?
In an unlevered purchase, price cannot fall below zero: the theoretical maximum security loss is the capital invested. In a short, price can rise above the sale price without a preset upper bound. The theoretical loss is therefore unbounded.
The operation normally takes place in a margin account. The broker must locate securities available to borrow; availability, fees, and rates can change. A lender can recall shares, and the broker can require covering. If price rises quickly, margin demands and forced purchases may amplify the move.
Initial sale proceeds are neither available profit nor a cap on the loss. Dividends or other distributions owed to the lender and every broker-specific rule must also be considered.
How should signal and feasibility be separated?
A signal test must use only information available at the time and include every case. A separate operating assessment must determine whether the security was borrowable, at what cost, with what margin, and at what prices the short and cover could have been executed.
This distinction prevents a seemingly correct graphic signal from being turned retrospectively into an impossible trade or one far more profitable than it really was. Even perfect directional analysis can produce a different outcome because of timing, a squeeze, or execution.
Evidence boundary — The short side of Hurst’s model is a historical analytical convention. It does not remove borrowing, margin, recall, squeeze, or theoretically unbounded loss; selling a long position is not by itself a short signal.
Sources
- J. M. Hurst, The Profit Magic of Stock Transaction Timing, Prentice-Hall, 1970, Ch. 5, pp. 94–96 — extension of signals to the short side, directional selection, and the role of market averages.
- FINRA, Short Interest: What It Is, What It Is Not — selling borrowed securities, margin accounts, and losses when price rises.
- FINRA, Risk — relationship between expected return, risk, and possible loss.
Related entries
- Profit and exit management — chapter framework
- Trailing and profit-taking — exit logic
- Short selling — general operation