How a return-shape rule traded STX and held patiently for weeks

STX ran more than eightfold over the window, and what stands out about this model is its patience: it traded rarely and held each position a long time, in places for the better part of a month, so the question is what that slow, choosy stance kept of a very large move. Read what follows as a study of those fixed rules replayed over STX's saved history, not a forecast or a recommendation; with twenty-three trades the live sample is still too thin to grade.

The behaviour this model looks for is the silhouette of STX's recent return distribution distorting away from its everyday outline, the moment its extremes bulge and the familiar contour warps out of shape. It bides its time while the silhouette holds its ordinary form and stirs only after the distortion reaches an unusual extreme, going long where the warp tips up and short where it tips down. The up tip carried twenty of its twenty-three reads. Hit rate across the 23 trades came in at 65.22%, with a cumulative figure of +182.86% on uniformly sized positions.

How the model is built, end to end.

What underpins the model is how far the silhouette of STX's recent return distribution has buckled from its everyday outline, a fourth-moment gauge that climbs as the extremes bulge and the contour warps and subsides when the tape returns to ordinary sessions. The model judges it against the outline this stock normally holds, so a level rates as extreme only against STX's own past. An everyday silhouette keeps the model idle; only a pronounced warp in the contour clears the way to an entry.

A warped silhouette flags a candidate, never a command. The model holds until a parallel directional verdict lines up the same way, which spares it from reacting to any lone large session. With both in agreement it submits a trade bounded by a stop and an objective, both pinned beforehand. The close then comes by one of several paths, and for STX the path that prevailed by a wide margin was the trade ending as the silhouette eased back toward its everyday outline, far in front of either objective or stop, fitting the long holds these trades ran.

The illustration sets out a single traverse of STX: a silhouette reading to open, a gate barring the way, a settled position to close. Not one stage is ornate; the governing trait is patience, a refusal to move until the distribution warps clear of its outline and the parallel direction verdict lines up, after which each trade is granted room to run. Worth flagging too is the rule set's lineage: it was sifted from a broad pool of machine-generated candidates and lasted on the strength of its backtest plus walk-forward scores, the passage from idea to deployment standing behind every number shown.

At a glance

These numbers come from a backtest of the model on STX, scored against fixed acceptance gates, not from a live track record.

Walk-forward verification

A trade walked through

The walked example is a long held about three weeks. The shape reading stretched to an upward extreme near 190 dollars, the direction read agreed, the model bought, and the position closed near 236 dollars for +24.11% once the reading eased back toward its usual form. It is the model working as designed, a patient entry on a stretched reading and a long hold carried until the move had run.

What the full trade record shows

Across its 23 STX trades the model won 15 and lost 8. The exits leaned heavily on the reading itself: 16 closed once the shape returned toward its usual form, 5 reached the profit target, and 2 were stopped out.

A book this dominated by signal-based exits is the signature of a model that holds long and closes when its reason for the trade fades rather than at a hard barrier. Stops are rare, just two of twenty-three, which fits both the solid win rate and the patient, weeks-long holds.

The biggest winner was a long that ran to its target for +27.11% over about sixteen days in April 2026. Most winners, though, closed on the settling reading rather than a fixed target, carried until the stretched shape relaxed back toward normal.

The worst trade was a long stopped out for -15.1% on a position held for weeks into April 2025; that single loss is also the equity curve's deepest drawdown, and it is the reason the drawdown gate is one the backtest missed.

No single trade carries the record, and at twenty-three trades the book is thin. With STX up more than eightfold over the window, the lesson is how much a slow, patient rule leaves behind, holding a fraction of a very large move while waiting for the shape to stretch.

How does this compare to just holding STX

Over the same window the model was tested on, simply buying STX and holding it would have done far better. Setting the two side by side is how you tell whether the rule earned its keep or merely tagged along, and here it tagged along, holding a fraction of an eightfold run while trading only a handful of times. The tiles below put numbers on the gap.

How well does the model reproduce its tape?

Walk-forward verification checks whether the saved rule path reproduces the expected signal behavior on held-out timestamps it was never fit on. It is a consistency and replay-integrity test, not proof the model will earn money live. A clean reproduction means the deployed rules act like the studied ones; it says nothing about whether STX will keep stretching its return shape the way it did across the test window.

In live trading the model has been quiet so far, with too few signals to set against the backtest. Until more live trades accumulate, the backtest is the only evidence available, and it should be read as exactly that, a study of how the rules behaved on saved history.

When this approach fails

The model's losses come from the same place as its trades, the stretched extremes. Its sharpest one is being early on a long hold: the shape stretches, the model buys, and STX grinds lower for weeks before the stop trips, which is how the deepest trade lost 15.1% on a long held into April 2025. Because it is long almost always and slow to exit, it has little defense against a sustained slide, and that single stop-out set the equity curve's worst stretch. Its quieter cost is selectivity itself, the long calm patches it sat out during an eightfold run.

Three things are worth watching if this ever trades at size. The first is drawdown, since the worst stretch in the backtest was about 15.42% and the long holds mean a losing trade can bleed for weeks. The second is the long-heavy exposure, because the model has little way to profit from or hedge a sustained decline. The third is the gap between live and backtested behavior, the first sign the shape reading is no longer marking the same moves.

Risk and honest limits

On this run the model's automated checks logged a caution rather than a clean pass. It is one more reason to read everything here as backtest evidence about the rules, not a verdict on the stock.

Lifecycle