How a jumpiness rule traded a tripling INTC and came up just short

INTC more than tripled across the window, and the interesting question is what a disciplined jumpiness rule did with a stock that ran that hard: it captured a meaningful share of the move while sitting out the long, quiet stretches in between. What follows runs those fixed rules forward over INTC's saved history as a study of the rules, not a forecast or a recommendation, and the live sample is still too small to grade.

What this model gauges is the peakedness of INTC's return curve, the way the distribution grows sharply spiked at its center with heavier extremes rather than spreading evenly. It abstains while the curve stays rounded and engages only when that spike reaches an unusual height, going long when the sharpened profile a peak skewed to the upside opens a long and one skewed down opens a short. The upward read held nearly throughout, twenty-eight of the thirty positions. Of the 30 trades, 70% finished in profit, and the cumulative tally reached +261.10% at a fixed trade size.

How the model is built, end to end.

The quantity the model is built around is how sharply peaked INTC's recent return curve has grown, a fourth-moment index that lifts when the profile narrows to a spike with fatter edges and softens when moves stay broad and even. That gauge is set against the level this stock typically shows, so a reading earns the label extreme only versus INTC's own record. A moderate profile keeps the model flat; only a pronounced sharpening of the peak lets a position form.

A sharply peaked reading is grounds to look, not to enter. The model goes ahead only where an accompanying direction read concurs, holding it back from acting on any one outsized session. Where both align it registers a trade with stop and target fixed beforehand. The close can land in several places, and for INTC the one that dominated by some distance was the trade unwinding as the peak rounded back toward normal, well clear of either the target or the stop.

Take the illustration end to end over INTC: the peakedness reading at the head, the gate set in its way, the position yielded at the close. No part is clever in isolation; the discipline rests in holding out until the return curve genuinely sharpens and the accompanying direction read concurs before any capital is staked. The rule set's lineage belongs in the account as well, since it came up as one contender among a machine-built field and endured by passing every backtest and walk-forward test put to it, the route from idea to deployment beneath each number reported here.

At a glance

These numbers come from a backtest of the model on INTC, 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 two weeks. The jumpiness reading pushed to an upward extreme near 19 dollars, the direction read agreed, the model bought, and the position closed near 25 dollars for +29.77% once the reading eased. It is the model working as intended, a patient entry while the tape was jump-prone and an exit booked as the effect faded rather than forced to a hard barrier.

What the full trade record shows

Across its 30 INTC trades the model won 21 and lost 9. The exits leaned heavily on the reading itself: 16 closed once the signal settled back, 8 reached the profit target, 5 were stopped out, and 1 ended on a raw signal flip.

A book this tilted toward signal-based exits is the signature of a model that mostly closes when the jumpiness drains away rather than at a hard barrier. Targets do real work here, more than a quarter of the exits, while stops stay rare, which fits the strong win rate.

The biggest winner was a long that ran to its target for +36.49% over about two weeks in April 2026. The patient winners tended to be carried until the reading relaxed, taking what the jump-prone stretch offered before it cooled.

The worst trade was a long stopped out for -12.73% over roughly four days in May 2026 after entering above 125 dollars; that single loss is also the equity curve's deepest drawdown, and it is the only acceptance gate the run missed.

No single trade carries the record. With INTC more than tripling over the window, the book reads as a disciplined rule capturing a solid majority of the climb, clearing four gates while a lone deep stop-out kept it from a clean sweep.

How does this compare to just holding INTC

Over the same window the model was tested on, simply buying INTC and holding it would have done better. Setting the two side by side is how you judge whether the rule earned its keep or merely tagged along, and here it captured a meaningful share of a tripling stock while staying in cash for stretches of it. The tiles below put numbers on the distance between the two.

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 make money live. A clean reproduction means the deployed rules act like the studied ones; it says nothing about whether INTC will keep turning jump-prone 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 jumps. Its sharpest one is being early at a high price: the tape turns jump-prone, the model buys near a peak, and INTC reverses, which is how the deepest trade lost 12.73% on a long stopped out in May 2026 after entering above 125 dollars. Because it is long almost always, it has little defense against a fast reversal, and that single stop-out at depth is what produced the drawdown the gates flagged. Its quieter cost is selectivity, the calm stretches it sat out while the stock climbed.

Three things are worth watching if this ever trades at size. The first is drawdown, since the worst stretch in the backtest reached about 12.89% and is the one acceptance bar the run failed. 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 hint the jumpiness reading is no longer marking the same moves.

Risk and honest limits

On this run the model's automated checks cleared most of the bar but logged one caution. It is one more reason to read everything here as backtest evidence about the rules, not a verdict on the stock.

Lifecycle