Does the opening range contain useful information?

Does a breakout of a predefined opening range reach a one-range target before the opposite barrier or session cutoff? This is an untested hypothesis, not a claim that opening-range breakouts are profitable.

Start with one instrument and session. Write the timezone, daylight-saving convention, data frequency and sample dates explicitly.

Choose a range window, such as the first 30 minutes after a specified session open. Let its high be $H$, low be $L$, and width be $w=H-L$. Define a breakout using an executable rule: first touch, first close beyond a boundary, or a stop order with an explicit fill model.

For an illustrative long entry at $H$, a stop at $L$ risks $w$. A 1:1 target is therefore $H+w$. A stop at the midpoint risks $w/2$ and needs a different 1:1 target. These definitions must not be mixed.

State the short-side equivalent, cutoff, and whether a second attempt is allowed.

Classify eligible observations as target first, stop first, or neither by cutoff. Report the total eligible sample and all three counts. If both barriers are touched in one bar, use finer data or label the observation ambiguous; do not assume the favorable order.

$\widehat p_{target}=\frac{N_{target\ first}}{N_{eligible}}$

Keep a confidence interval with the estimate. A conditional estimate that excludes neither outcomes answers a different question and must be labelled separately.

A signal describes an event probability. A strategy adds executable entry, exit, sizing and risk rules. Include spread, commission, slippage and the actual return of cutoff exits.

$\widehat{\mathbb E}[R]=\frac{1}{n}\sum_{i=1}^{n}R_i$

For the restrictive two-outcome, equal-size payoff case before costs, $\mathbb E[R]=p-(1-p)=2p-1$. Real cutoff exits, variable fills and costs generally invalidate that shortcut.

1. Reserve a chronological holdout before selecting filters.
2. Compare the baseline and a small preregistered set of filters on the same dates.
3. Record every tested variant, including failures.
4. Inspect rolling behavior, sample size and sensitivity to costs.
5. Evaluate the final unchanged rules on the holdout.
6. Forward test execution on a demo account before making any deployment decision.

Status: proposed research only. No test has been run for this article.