PULSCAR ANALYSIS — WHITE SMOKE FROM THE EXHAUST IN NHTSA PUBLIC DATA PulsCar analysis of NHTSA public data, September 2026. License: CC BY 4.0. Attribution: "PulsCar analysis of NHTSA public data, September 2026." Published alongside https://pulscar.io/blog/white-smoke-from-exhaust ------------------------------------------------------------------------------ 1. SOURCE FILES ------------------------------------------------------------------------------ All inputs are NHTSA's own published flat files, downloaded from static.nhtsa.gov and loaded into a local SQLite database. No third-party data was used. COMPLAINTS — 777,040 owner complaints received 1 January 2015 to 8 September 2026. Field definitions: https://static.nhtsa.gov/odi/ffdd/cmpl/CMPL.txt TSBs — manufacturer communications, including customer satisfaction programs. Field definitions: https://static.nhtsa.gov/odi/ffdd/tsbs/FLAT_TSBS.txt INVESTIGATIONS — ODI defect petitions, preliminary evaluations and engineering analyses. Complaints are self-reported. They are not failure rates and they have no denominator. Reporting volume rises sharply with publicity, owner forums and class-action activity — which is visible in this set and is reported on the page as a finding rather than hidden. Nothing here should be read as "make X is more or less reliable than make Y". ------------------------------------------------------------------------------ 2. BUILDING THE SET ------------------------------------------------------------------------------ Step 1 — full-text search of the narrative field: SELECT c.* FROM complaints c JOIN complaints_fts f ON f.rowid = c.rowid WHERE complaints_fts MATCH 'smoke OR smoking OR steam OR vapor'; -> 16,733 complaints. Step 2 — narratives upper-cased and whitespace-normalized, then kept only where a white-smoke phrase appears: (WHITE|WHITISH|GRAY-WHITE|WHITE-GRAY) + up to two words + (SMOKE|SMOKING|VAPOR|STEAM|CLOUD|EXHAUST) Step 3 — the white-smoke phrase must sit within 150 characters of an exhaust word: EXHAUST | TAIL PIPE | TAILPIPE | MUFFLER | OUT THE BACK | REAR OF THE (CAR|VEHICLE) -> 1,400 complaints, received 1 January 2015 - 8 September 2026. This is the set used on the page. The exhaust-proximity rule is what keeps the set honest: without it the query returns engine-bay smoke, tire smoke, cabin smoke and smoke from a fire, none of which is the symptom this page is about. ------------------------------------------------------------------------------ 3. SAMPLE READING AND FALSE POSITIVES (done before publishing any count) ------------------------------------------------------------------------------ 30 of the 1,400 were drawn at random (Python random.seed(31337), random.sample) and read in full. 27 of 30 were unambiguously white exhaust smoke reported as a fault. 3 of 30 were doubtful rather than wrong: two described smoke the owner called "white" but also "light blue" or reported alongside oil consumption, where the color split between coolant and oil is genuinely unclear from the text, and one named no color source for the smoke beyond the phrase itself. Observed error rate 10%, and it is a color-judgment error rather than a keyword error: the complaints are about exhaust smoke either way. Read every share below with that 10% in mind. WHAT THE SET CANNOT SEE. This file systematically under-counts cheap causes. Owners write to the federal government about failures that stranded them, cost four figures or were refused under warranty; nobody files a complaint about a $400 intake manifold gasket that fixed the problem, and only 2 of the 1,400 narratives mention an intake manifold gasket at all. The absence of a cheap cause here is evidence about who files complaints, NOT evidence that the cheap cause is rare. The article says so where the number appears. ------------------------------------------------------------------------------ 4. FILES ------------------------------------------------------------------------------ white-smoke-by-context.csv What the narrative names, as a count and share of the 1,400. Keyword matches, NOT exclusive, and they do not sum to 100%. "A head gasket" means the words appear, not that a head gasket was proven — owners write down the part a shop named out loud. white-smoke-by-make.csv / white-smoke-top-models.csv Counts by make and by make+model with median failure mileage, the number of complaints carrying a usable odometer, the NHTSA towed and fire flags, and how many narratives use the phrase "coolant intrusion", name an engine or short block, or name a head gasket. Rows with fewer than 3 complaints are omitted. Mileage statistics drop odometer values at or below 0 and at or above 400,000; 727 of the 1,400 complaints carry a usable figure. These are raw counts with no denominator. white-smoke-ford-ecoboost-by-model-year.csv The same columns for Escape, Edge and Fusion broken out by model year, because the concentration in this file is a model-year concentration rather than a make one. white-smoke-complaints-by-year-received.csv Complaints per calendar year of receipt, split into the Escape/Edge/Fusion group and everything else. 2026 is a partial year: the file ends 8 September 2026. This table is the clearest thing in the analysis and also the most easily misread. It shows FILING behavior, not failure behavior. A rise can mean more failures, or more owners learning there is a program, or a class action, or a news cycle. ------------------------------------------------------------------------------ 5. BASELINES USED FOR COMPARISON ------------------------------------------------------------------------------ Computed over all 777,040 complaints in the same file: towed 5.11% crash 4.37% fire 1.82% median miles 70,000 The white-smoke set returns 124 towed (8.86%), 2 crashes (0.14%), 38 fire flags (2.71%), 7 injuries, no deaths, and a median of 86,000 miles (quartiles 64,733 and 114,000), with 8.67% of usable odometer readings at or under 36,000 miles. ------------------------------------------------------------------------------ 6. LIMITATIONS ------------------------------------------------------------------------------ - Self-reported, unverified, no denominator, no failure rate. - Color is the owner's judgment, at a tailpipe, often in the dark. White, gray and light blue blur, and thin condensation on a cold morning is white smoke to somebody who has not seen it before. 169 of the 1,400 narratives mention condensation or steam explicitly, in both directions. - A keyword set, not an NHTSA component code. - Cheap causes are structurally invisible here; see section 3. - The diesel branch is nearly absent: 12 narratives mention a diesel and none mention glow plugs, so this file says nothing useful about the unburned-fuel version of white smoke. ------------------------------------------------------------------------------ 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, in two layers: * the complaint set is built with an FTS5 MATCH full-text search, which folds case before matching, so it cannot miss a mixed-case record; and * where the narrative is then tested with patterns, the narrative is upper-cased and whitespace-normalized first, so the patterns compare upper against upper. 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. Re-run here: step 1 is MATCH 'smoke OR smoking OR steam OR vapor', and step 2 upper-cases the narrative before testing it.