The squeeze, the bear-market rally, and the borrow that vanishes

The squeeze: the failure that compounds while it happens

A short squeeze is the market's feedback loop running against you: a bounce forces some shorts to cover, their buying lifts the price further, which forces more covering. In a crowded name the loop can move the price 30% in a session — through any mental stop, sometimes through the hard one, with fills far from the trigger.

Remember which direction the arithmetic runs: the rally is growing your position as it grows your loss. This is the one failure mode in the library where doing nothing makes the problem larger by itself — and it is why this system cannot be run without a hard stop.

The defences are all pre-trade: avoid heavily-shorted names, keep size below the long side's numbers, respect the earnings-date exclusion, and place the stop as a working order rather than an intention. During the squeeze itself there are no good decisions left — only the stop you placed or the one you meant to.

Bear-market rallies: the whipsaw built into the cover rule

Declines do not move in straight lines. Stage 4 trends are punctuated by rallies fierce enough to reclaim the 30-week average — bear-market rallies are historically among the sharpest moves in equities — and the cover rule will honour every one of them. Some of those covers will be immediately followed by the decline resuming, without you.

This is the short-side rendering of the whipsaw every slow-average system accepts, with one difference of temperature: the rallies that trigger it are faster and larger than the pullbacks that whipsaw the long side. Re-entry is the answer — the tests are re-armed the moment the rally fails back under the falling line — but each round trip pays slippage, borrow fees and nerve.

• Do not widen the cover rule to dodge whipsaws. Holding a short above a reclaimed average because 'it will fail again' is the exact reasoning that turns a contained loss into a squeeze casualty.
• Expect worse trade statistics than the long side. Faster rallies against a nearer stop mean more exits per trend; well under half of entries catching a clean leg down is the realistic texture.
• Count the carry. Borrow fees and the occasional dividend accrue across every round trip; a decline that arrives after three whipsaws has partly been spent in advance.

Failures that arrive from outside the chart

A long position's failure modes all show up on the chart eventually. A short has failure modes with no chart at all: the lender recalls the borrow and you are bought in at the market's convenience; the borrow fee reprices from 1% to 30% annualised as a name gets crowded; a special dividend is declared and the cost lands on you. None of these is a signal, and none of them cares about your entry tests.

Treat these as weather rather than injustice. They are checked at entry — borrow availability and fee, dividend calendar, earnings date — and monitored daily precisely because they change without notice. The twenty-minutes-a-day commitment is mostly this: confirming the position you hold is still the position you opened.

Being early: the Stage 3 short that keeps looking almost right

The most repeatable way to lose with this method is to run it before its conditions exist — shorting a topping stock below a still-rising average because the top 'must' be in. Stage 3 is precisely the phase where dips below the line get bought back to the highs, and each failed short there costs money and, worse, conviction that will be missing when Stage 4 actually arrives.

The rule already contains the cure, stated once more: below a falling average is Stage 4 and shortable; below a rising one is Stage 3, a reason to be out, not a reason to be short. Every expensive version of 'early' is a violation of that sentence.