Methodology & Evidence

How a diagnostic is built

1. Start from a symptom — what the driver can actually observe.
2. Branch on decisive questions — each answer must split the possible causes in a meaningful way.
3. Rank causes — most likely, possible, less likely — with the mechanism explained in plain language.
4. Resolve the safety verdict first — the worst credible state across the causes wins. This step is independent of any commercial consideration.
5. Attach evidence, checks, and costs — then a plain-language decision summary on top.

Source hierarchy

When sources conflict, this order wins:

1. Manufacturer documentation and official technical publications
2. Regulator and safety-authority records (e.g. NHTSA)
3. Established trade-technical references
4. Owner and community reports — always labeled as such, and never the basis for safety, procedure, specification, or fitment claims

Claim types & likelihood

Every piece of evidence carries a claim-type label so you can see what kind of statement it is — an observed pattern, a documented mechanism, a manufacturer procedure, or something weaker. Causes are labeled most-likely, possible, or less-likely. These are judgments about probability, not measurements, and they’re presented as judgments.

Cost methodology

Cost figures are labeled estimate ranges, built from U.S. parts-catalog spreads across major retailers plus labor-rate bands (major metro / mid-market / rural). Each range carries an uncertainty level (low, medium, high) and a check date. They are not quotes — your vehicle, location, and shop change them. Where we lack reliable data, we say so instead of inventing a number.

What we will not do

• Invent torque values, fluid specifications, electrical specs, part numbers, fitment data, recalls, codes, TSBs, or costs.
• Present a possible cause as a confirmed diagnosis.
• Let commercial interests change a safety verdict or suppress a warning.
• Rank or score sources with “trust scores” — sources are named and cited, not rated.