PULSCAR ANALYSIS OF NHTSA PUBLIC DATA - OIL USE BETWEEN OIL CHANGES PulsCar analysis of NHTSA public data, September 2026. License: CC BY 4.0. Files in this folder may be reused with attribution to PulsCar (pulscar.io). -------------------------------------------------------------------------- 1. SOURCE FILES -------------------------------------------------------------------------- Complaints FLAT_CMPL.txt 777,040 unique complaints (ODINO), received 1 January 2015 - 8 September 2026. Field definitions: https://static.nhtsa.gov/odi/ffdd/cmpl/CMPL.txt Communications FLAT_TSBS.txt manufacturer communications; counted DISTINCT on the NHTSA document id. Complaints are OWNER REPORTS, not verified repairs and not failure rates. There is no denominator of vehicles on the road in this data, so a make with more complaints may simply sell more cars or have more organized owners. Nothing in these files ranks reliability. -------------------------------------------------------------------------- 2. COMPLAINT QUERY (the 7,734) -------------------------------------------------------------------------- Step 1 - full-text search (SQLite FTS5) of complaint narratives for any of: "oil consumption", "burning oil", "burns oil", "consuming oil", "consumes oil", "losing oil", "loses oil" -> 7,761 complaints. Step 2 - regex confirmation on the narrative: (oil consumption | burn\w* (through )?oil | consum\w+ (a lot of|excessive| too much )?oil | los\w+ oil | add\w* (a )?(quart|qt) | quart\w* of oil (every|per|each)) -> 7,734 complaints. Date range 2 January 2015 - 8 September 2026. Step 3 - SAMPLE READING. 30 random matches were read in full. 30 of 30 were genuinely about an engine using or losing oil between changes. Residual false-positive rate in the read sample: 0 of 30. No further tightening was applied, because none was needed - "oil consumption" is a phrase owners use literally. Mileage statistics use the 3,099 records with a usable failure mileage (blanks, zeros and anything at or above 400,000 are dropped): median 85,570 miles, quartiles 55,000 and 115,000. -------------------------------------------------------------------------- 3. OWNER-STATED CONSUMPTION RATES (the 470) -------------------------------------------------------------------------- Narratives were scanned for an explicit rate in the form (1|one|a) (quart|qt) [of oil] (every|per|each|in|/) NUMBER [miles] Values below 100 or above 20,000 miles were discarded as parse errors (53 discards). 470 rates were parsed. Median: 1 quart per 1,000 miles; quartiles 654 and 1,000 miles. HONESTY NOTE, and it is a large one: owners round. 223 of the 470 stated values are exactly 1,000 miles, and other round numbers (500, 1,500, 2,000) dominate the rest. These are self-reported estimates, usually not the result of a measured test, so the distribution should be read as "what owners believe they are using," not as measured consumption. -------------------------------------------------------------------------- 4. OUTCOME READING (the 110) -------------------------------------------------------------------------- 1,588 of the 7,734 narratives mention a formal oil-consumption test (or an oil consumption study / burn test). 110 were drawn at random from that subset, read in full, and each assigned ONE primary outcome - the furthest point the owner's account reaches. Counts are in oil-consumption-test-outcomes-110-read.csv. Limitations of that table: a complaint is a snapshot, usually filed while the owner is still angry and often while the car is still at the dealer, so "test underway, no outcome stated" is the largest category by construction and "repair approved" is under-counted - nobody files a federal complaint to report that the engine was replaced under warranty without a fight. -------------------------------------------------------------------------- 5. MANUFACTURER THRESHOLDS -------------------------------------------------------------------------- Full-text search of bulletin summaries for "oil consumption" returns 667 distinct NHTSA document ids. 28 of those summaries state a quart-per-mileage rate or an acceptability statement; the six clearest published thresholds are transcribed in manufacturer-oil-consumption-thresholds.csv, quoting the summary text as NHTSA publishes it in the flat file. We did not obtain the underlying PDFs for these six documents - the static.nhtsa.gov PDF paths for them returned 404 in this pass - so the quotes are from the flat-file summary field, and the document id is given so anyone can look them up. -------------------------------------------------------------------------- 6. WHAT IS NOT IN THIS FOLDER -------------------------------------------------------------------------- Shop and chain oil-change prices. They are not in any federal file, and the estimator and chain pages that carry them refuse automated access, so no price on the article page was re-verified in this pass. ------------------------------------------------------------------------------ CASE SENSITIVITY — SECTION ADDED 19 SEPTEMBER 2026 ------------------------------------------------------------------------------ NHTSA stores complaint narratives in ALL CAPS for records filed through roughly 2020 and in mixed case for records filed from 2021 on. A case-SENSITIVE search of the narrative field loses about 92% of the complaints filed after 2020, so this is the single largest way an analysis of this database can go wrong while still looking finished. Every query behind this file is case-insensitive. SQLite's LIKE and its FTS5 full-text index both ignore case by default, and no query in this work uses GLOB or any case-sensitive comparison. This was not taken on trust. On 19 September 2026 every published folder was re-checked against the same database, comparing the figures here with what a case-sensitive query would have returned. The result: the complaint set published here (7,761) is LARGER than the absolute ceiling a case-sensitive search of the same terms could have returned (6,590 narratives). A case-sensitive query cannot produce a bigger set than itself, so the bug is ruled out for this folder by arithmetic, not by assurance.