Austin permit-sequence result survives three annual tests
A fixed plumbing + electrical → mechanical hypothesis beat the building-permit control by 35.8, 37.5 and 38.2 points in three untouched periods.
Direct answer
Yes, the Austin pattern persisted after the rule was fixed. We tested one predefined question: when both plumbing and electrical permits are present on one authoritatively linked project, and mechanical is not yet present, does a mechanical permit follow within 180 days more often than after a building permit alone?
It did in three later periods. This is an historical association for Austin, not a probability or recommendation for any individual project.
Results
| Period | Sequence | Building-only control | Difference |
|---|---|---|---|
| Rule discovery, 2019–2023 | 70.5% (17,529 / 24,856) | 30.8% (20,715 / 67,179) | +39.7 pp |
| Untouched validation, 2024 | 64.4% (1,261 / 1,959) | 28.6% (1,696 / 5,936) | +35.8 pp |
| Independent annual replication, 2025 | 69.0% (1,445 / 2,094) | 31.5% (1,981 / 6,281) | +37.5 pp |
| Later 2026 cohort observable by 19 Sep | 69.0% (591 / 857) | 30.8% (783 / 2,540) | +38.2 pp |
The control is every Austin building permit in the period, including the 42% that are never followed by any trade permit; an earlier version drew it only from projects that already had a trade permit, which knew the future and overstated the control at 50–55%. The rule, horizon, target and inclusion method were unchanged across the three later periods. A project is counted only if all 180 days after its observation point are inside the named period; this prevents newer records from being treated as failures.
What exactly was included
The City of Austin’s Build + Connect dataset supplied 483,881 relevant permits, issued from 2 January 2019 through 19 September 2026. We retained only 62,333 project groups joined by Austin’s shared permit-number base. That key was independently checked against addresses on 152,940 trade-to-building pairs, with 99.7% agreement. Inferred parcel or address joins are excluded.
The comparison is deliberately modest: a building-permit timeline, not a randomized experiment. It answers whether this sequence was more informative in this record; it does not establish why it occurred or guarantee that a specific job will need mechanical work.
Reproduce or audit it
The source manifest records the official dataset, retrieval query, row counts, period definitions and SHA-256 hashes of the raw and derived files. The public analysis cohort contains every included project’s dates and linkage method, but excludes names, addresses and contact details. The analysis code is published with this package. The City remains the primary record: Austin Build + Connect dataset.
The package is also archived with a permanent identifier: doi:10.5281/zenodo.23293324
(Zenodo; this DOI always resolves to the latest version). It adds a control-group file and a script, verify_signal.py,
that recomputes the results from the two CSV files alone, with no network access.
Why this is one market, not five
Five city labels would make a better-looking chart but weaker research. New York, Seattle, Chicago and Orlando currently have useful public records but an unmeasured inferred method for joining permits across a project. Philadelphia’s current source no longer supplies the trade and contractor fields used by its earlier analysis. We list those as replication candidates, not confirmations, until each linkage method is measured against an authoritative reference.
Direct answer
Our analysis uses the public records and sample shown below to measure a historical pattern, not to predict one future job with certainty. The answer is useful only with its jurisdiction, sample and time horizon attached; therefore each result should guide research priority and be verified against the issuing authority before commercial action.
How to use this research
These results are descriptive measurements from the stated records, market and observation window. Use them to choose what to investigate first, then verify an individual project with the issuing authority. A historical result is not a guarantee of a future permit, award or hiring decision.
Conclusion
The practical value is a narrower and more explainable research queue. The evidence is strongest when its jurisdiction, sample and horizon remain attached to the claim; that is why Projects Ahead does not turn local measurements into a single national score.