Dataset downloads
Quick answer: The full normalized dataset (7,628 records, 3,023 lines, 214 brands) is free to download below. Source data is NHTSA's, pulled 2026-10-25; please cite NHTSA and this site.
| File | Format | Contents |
|---|---|---|
| nhtsa-utqg-records.csv | text/csv | Normalized record-level table (one row per NHTSA record) |
| nhtsa-utqg-records.json | application/json | Same as CSV, JSON |
| nhtsa-utqg-lines.json | application/json | One object per tire line with computed median, percentile and brand rank |
| nhtsa-utqg-sizes.json | application/json | Per-size rows of lines with an explicit size record |
| summary.json | application/json | Counts and pull metadata |
CSV columns
| Column | Meaning |
|---|---|
| id | NHTSA record id |
| brand, brand_slug, brand_raw | Normalized brand, URL slug, and the brand string exactly as NHTSA filed it |
| line, line_slug | Normalized tire line name and slug |
| size_raw | NHTSA's size text, unchanged |
| size_key, size_label, size_type, size_suffix | Parsed size (e.g. 225-65r17, 225/65R17, P/LT, load/speed text); blank if NHTSA gave no specific size |
| treadwear, traction, temperature | UTQG grades; blank when missing or invalid |
| quality_flags | Notes on invalid or missing source values |
License note: U.S. federal government works are generally not subject to U.S. copyright (17 U.S.C. §105); see usa.gov/government-works. Our added fields (slugs, parsed sizes, ranks) are provided as-is, with no warranty.