How Roth earns confidence.

The method is public. Personalized fitting is not.

You can explore the Roth method, run the Free Check, review clearly labeled example customer and shop reports, and submit a founding-rider request. Public account creation, payment, personalized results, and Expert Review remain gated.

Example screens use illustrative rider information. They are not customer results, testimonials, validation evidence, or a promise of access.

The process is designed to avoid guessing.

Roth checks the phone setup, compares two side recordings made without changing the bike, adds a front view, and starts with one change you can undo. You then check that change while the bike is held indoors and across three real rides.

• No guidance when a recording fails a quality check
• Recording quality, matching views, and confidence in a change are checked separately
• Only one suggested adjustment at a time
• Your starting setup and return instructions stay with every change

Performance claims begin with real-world evidence.

Roth publishes results only after using a clear testing method, a varied group of real riders, and qualified review. The evidence program focuses on recording success, repeatability, fitter agreement, and rider-reported outcomes.

• Recording success across supported phones
• Repeatability between matching side views
• Agreement with qualified fitter review
• Rider outcomes across the complete follow-up period

The planned first-release result path is automatic.

Roth is recruiting founding-rider requests for a future controlled beta. The public website does not currently sell or deliver a personalized fit. Before that path opens, the automatic method must be compared with independently measured reference fits and qualified fitter assessments. Expert Review remains a separate planned service and is not available.

• Testing on a useful range of real phones and riders
• A written process for recording and comparing views
• Independent comparison with qualified fitters
• Follow-up through all three ride stages
• Hands-on testing of privacy, deletion, rollback, and safety controls

Show how results were reached—and correct mistakes openly.

Future results pages should explain which model, automatic policy, and method were used, whether optional expert help was requested, which riders were included or excluded from validation, when the work was done, and what its limits were. A newer method should never quietly rewrite an older rider report.