WhitmanTrading

How to Short a Stock

To short a stock, borrow the shares through a margin account and sell them, with the intention of buying them back later at a lower price. The borrow costs a daily fee, can be recalled at any time, and the potential loss is not capped the way a long position's is.

Shorting means borrowing shares, selling them, and buying them back later. If the price falls you keep the difference. Everything difficult about it comes from the borrowing: it costs money daily, it can be withdrawn, and the position it creates has no natural limit on its losses.

Before you start

A margin account, because shorting requires borrowing the shares. A standard cash account cannot do it, and the account terms govern what happens when things go wrong.

The borrow cost, which is charged daily and varies enormously. On a widely held name it is negligible; on a heavily shorted one it can exceed any plausible gain.

An acceptance that the loss is not bounded the way a long position’s is. A share bought at 40 can fall to zero. A share sold short at 40 can rise to 200.

The steps

1. Check the borrow is available and what it costs

A range-bound stretch of price with an ongoing charge.
The borrow fee accrues every day you hold. Illustrative chart - not real market data.

Your broker shows availability and rate. A name that is expensive to borrow is expensive because everybody wants to short it, which is information about the trade.

2. Size for an unbounded loss

A slice of price data with an open-ended move.
There is no upper bound on a share price. Illustrative chart - not real market data.

Take the stop distance and size from it as usual, then take a smaller position than that arithmetic suggests. The gap between the two is the premium for an outcome with no ceiling.

3. Place the stop above a structural level

A long-horizon price series with an invalidation level.
Above the high that would prove you wrong. Illustrative chart - not real market data.

On this site’s shared series the ninetieth percentile bar range is 1.101, so a stop inside that band is noise. Above the swing high that would invalidate the idea.

4. Account for the daily cost in your horizon

A slow-moving stretch of price with an accumulating charge.
A slow decline can be eaten by the borrow. Illustrative chart - not real market data.

A high borrow rate held for months can exceed the fall you were expecting. The cost turns a slow correct call into a losing trade, and that arithmetic belongs in the plan.

5. Know that the shares can be recalled

The first half of a price series with a forced exit.
The lender's decision, not yours. Illustrative chart - not real market data.

If the lender wants them back and no replacement can be found, your position is closed at the market price on their schedule. There is no appeal and no warning.

6. Remember dividends run against you

A section of a price series with a periodic deduction.
You pay the dividend while short. Illustrative chart - not real market data.

The shares are borrowed, so any dividend paid is owed by you to the lender. On a high-yielding name held across several payments that is a substantial recurring cost.

7. Cover on the plan, not on the panic

The first half of a price series with a defined exit.
The exit was decided before the position existed. Illustrative chart - not real market data.

Both exits written before entry: the target and the invalidation. A short position moving against you generates more pressure to improvise than any other trade type.

How to tell it worked

The borrow rate was checked before entry, in every case.

Position size was smaller than the stop-distance arithmetic alone would give.

The stop sat above a structural level, more than 1 average bar range clear of noise.

And 0 trades were held past their written invalidation.

Why the asymmetry matters so much

A candlestick chart annotated with the round-trip cost of a switch.
The borrow is charged on top of the round trip. Illustrative chart - not real market data.

A losing short grows. As price rises, the position’s value against you increases, so the exposure is largest exactly when it is going worst. A losing long shrinks as it falls.

A section of a price series drawn without volume context.
And a thin name is the hardest to cover quickly. Illustrative chart - not real market data.

Which means the position size you started with is not the position size you have. That is the mechanical reason short positions require more conservative sizing rather than merely more conviction.

The base rate runs against you

On this site’s shared series 54% of 566 ten-bar windows finished higher and 52% of 571 single bars did. A short position starts slightly behind before anything else is considered.

And the borrow cost compounds that. A position that needs price to fall, in a market that drifts up, while paying a daily fee, has three things to overcome rather than one.

None of which makes shorting unreasonable. It makes the edge required larger — the setup has to be better than the equivalent long, not merely as good.

Alternatives that cap the loss

A put option risks only the premium. It expires, which a short position does not, and it costs money up front rather than earning it — but the maximum loss is known at the moment of entry.

An inverse fund moves opposite an index without borrowing. It carries its own problems, chiefly that daily rebalancing makes multi-day returns diverge from the index’s inverse, but nothing about it can lose more than the amount invested.

Neither is a straight substitute. The put has a deadline the short does not, and the inverse fund tracks an index rather than a company.

They are worth knowing about because the bounded loss is the whole point. On this site’s shared series a 2x leveraged position produced a 7.45% drawdown against 3.76% for the base — exposure that compounds against you is exactly what an unbounded short can do, and both of these alternatives put a floor under it.

The original data

Of the 24,971 unique videos in research/search-study-corpus.jsonl, 6 mention short selling in the title, at a median of 58,304 views across 6 channels, and 67% of those titles are instruction-shaped. Short squeezes and margin appear in similarly small numbers. The counts come from site/corpus_count.py.

A candlestick series with several gaps, the largest of them marked.
A gap up against a short is the outcome with no ceiling. Illustrative chart - not real market data.

6 videos at a 58,304 median. Almost no coverage and one of the largest audiences per video in the corpus — a subject a great many people want explained and almost nobody teaches, which is the shape this site exists to fill.

A stretch of price bars cut short at a decision point.
The short is 20% against you and the thesis is intact. Add? Illustrative chart - not real market data.

The answer to the question on that chart is that adding increases an exposure that is already larger than you chose. A losing short grows on its own — the position is bigger than when you opened it, and adding to it compounds the one property that makes shorting different.

When it fails

The failure is adding to a losing short, and it is the mechanism behind most of the large losses in this trade type. Price rises, the thesis still looks right, and the higher price makes the short better value. So more goes on. Price rises again. The position is now several times the original size, in a trade whose exposure was already growing by itself, in an instrument that other shorts are being forced to cover — which is what makes the move accelerate rather than stop.

The second failure is ignoring the borrow cost. It can exceed the fall.

A third is sizing by analogy to a long. The loss profile is not symmetrical.

A fourth is forgetting the dividend. You pay it while short.

A fifth is assuming the position is yours to close. It can be recalled.

And a sixth is shorting a hard-to-borrow name. The cost is telling you something.

Short selling covers the mechanism in detail. Short squeeze is the specific failure mode. And margin account is what the whole position runs inside.

What I actually do

The asymmetry is the part to sit with before doing it once. A long position can lose what I put in. A short position can lose several times that, because there is no upper bound on a share price — and the size has to be chosen with that in mind rather than by analogy to a long.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.