Does Opening Range Breakout Still Work? Evidence From 0DTE and 5-Minute Tests

Repository reference: cutebacktests

Abstract

The question "does ORB still work?" sounds simple, but in options research it is too broad to answer honestly. A same-day ORB using fragile options fills is not the same object as a 5-7DTE ORB with a stricter option surface and a narrower entry design. In this repository, the evidence points to a conditional answer rather than a slogan.

After the realism work summarized in ORB Framework Audit, broad ORB did not survive as a family-level claim. What remained defensible was a specific pocket: directional ORB, 5-7DTE, 5 minute opening range, range-stop geometry, and a narrower trigger design. That means the answer to the title is yes for a very specific slice and no for broad generic ORB enthusiasm.

Question

The useful question is not whether ORB has ever worked. It is whether ORB still works once the backtest stops handing it easy assumptions.

That distinction matters because ORB can look durable when its tests are permissive. If the strategy can mix wide DTE ranges, use optimistic same-bar or near-same-bar entry assumptions, and treat the option expression casually, the family will almost always produce survivors somewhere. Once those choices are tightened, the question becomes narrower and much more interesting.

Method: Does ORB Still Work Under Specific DTE and Timing Assumptions?

The ORB audit in this repo effectively answered the title by decomposing the family rather than by judging it in one lump. It separated broad 0DTE, 1DTE, 2-3DTE, and 5-7DTE lanes. It looked at range duration and stop geometry. It examined parity quality and whether the surviving branch remained interpretable after those decisions were isolated.

This is the right method because ORB's strongest failure mode in public discussion is over-generalization. A researcher finds one working configuration and then speaks as if the family itself has been validated. The repo moved the other way. It kept cutting the family apart until the surviving object was specific enough to defend.

Evidence / Results

The evidence now looks like this:

• broad ORB did not survive well
• a narrow ORB pocket did survive
• the surviving pocket was directional ORB with 5-7DTE, 5 minute opening range, and range-stop geometry
• broad 0DTE, 1DTE, and 2-3DTE lanes did not survive as general-purpose claims

That result appears in both ORB Framework Audit and Episode 4. The audit also reported improving parity quality:

• bar_open: 13/18 exact matches
• live_poll_c30_l10: 15/18 exact matches

Those parity figures matter because they show the repo was not "saving" the narrow survivor by lowering standards. The standards were becoming more credible while the family itself was being narrowed.

What Worked

What worked was the decision to stop treating ORB as a monolith. Once the family was decomposed by DTE and trigger design, it became possible to say something useful. A slower-DTE, narrower ORB lane still had life. That is a stronger result than a vague family win because it is easier to falsify and easier to compare against future evidence.

The repo also benefited from better parity. A good ORB answer needs more than a backtest chart. It needs some evidence that the simulated object and the live object are moving closer together. The parity improvements reported by the audit support that.

What Failed

What failed was the broad claim. If someone asks whether ORB still works in the generic sense, this repo does not support a confident yes. Broad 0DTE and other short-DTE lanes did not hold up as general recipes. Longer opening-range durations did not act as a universal repair. The family survived only after getting smaller.

This is an important negative result because generic ORB enthusiasm is still common in public trading content. The repo's evidence suggests that the more honest answer is conditional and narrow. A lot of the edge people think they are seeing comes from bundling together configurations that should be judged separately.

Takeaway

Does ORB still work? In this repository, yes for a specific 5-7DTE, 5 minute directional lane, and no for broad family-level claims. That answer is much less exciting than a universal slogan, but it is much more useful.

If you want the fuller audit framing, Opening Range Breakout Backtest Results: What Survived After Realism Fixes is the companion piece. If you want the option-expression failure surface, Why Most Opening Range Breakout Strategies Fail Under Realistic Options Fills goes there next. Join the research log to get the next backtest and failure report.