California Crash Guide
Where crashes are reported, when they happen, and how California is changing.
3 PM stands out in California’s crash records
California recorded 312,851 crashes in the selected period. 3 PM recorded 22,825 crashes, 7.3% of the selected total. Friday had the highest day-of-week count, with 49,505 reported crashes.
rear end is the most common recorded collision type in this selection, appearing in 93,872 reports.
Counts describe reports, not individual risk. They do not adjust for traffic volume, distance travelled or time spent on the road. Statewide totals include records outside named incorporated cities.
California crash map
Explore approximate concentrations of reported crashes. Markers combine records in roughly 5 km grid cells; they are not individual crash locations or intersection rankings.
Loading the 2026 aggregate map…
Map coverage varies by year and place. Records with missing or unusable coordinates, or coordinates outside the displayed extent, remain in the guide totals but are not plotted. Areas without markers cannot be interpreted as having no crashes. Map controls and attribution are provided by Leaflet and OpenStreetMap.
Reported crashes by year
2026 runs through Nov 7, 2026. A partial year is not directly comparable with a full calendar year. Counts can change as reports are added or revised. Some reports may be preliminary.
Monthly counts across years
The latest month may be incomplete. Future months are not plotted as zero. These charts show all loaded years, independently of the overview period.
Time of day
Reported local clock time. Unknown and invalid times are excluded from hourly rankings; details are in methods.
Day of the week
Recorded collision types
Source descriptions, not party or person classifications. Up to 12 known categories; unknown descriptions remain in the total.
What this first California release supports: crash counts, dates, known hours, collision descriptions and aggregate maps. Party/person breakdowns and serious-injury measures are not included. Their absence is not a zero value.
Download aggregate data (CSV) ↓Methods, definitions and limitations
Source: California Highway Patrol — California Crash Reporting System (CCRS), annual Crashes files for 2022–2026. Each source is downloaded, archived with a SHA-256 checksum, normalized, independently audited and copied privately before aggregate publication. Crash counts represent eligible crash reports, not people.
Geography: all eligible California crash records in these annual files, including unincorporated locations. Official geographic reference: U.S. Census TIGERweb.
Explicitly deleted records are excluded. Identical duplicate identifiers count once; conflicting identifiers are held for review. A source with unknown deletion status or cross-year identifier conflicts cannot be published by this workflow. Invalid or missing casualty fields do not erase an otherwise usable crash event; those fields remain unavailable. No party or person file is required for this crash-only view, and their measures are not inferred.
Dates use the reported local calendar date. Hours require a valid 0000–2359 source time matching the reported timestamp. Codes such as 2500 and inconsistent times are treated as unknown. See the CHP reporting manual, date and time conventions. “Last 12 calendar months” ends at the last calendar-month end represented in the source; this does not certify that reporting is complete.
For the selected period, 3,249 records have unknown/unusable hour, 4 lack a known primary road and 587 lack a known collision description. Casualty counts are missing in 60,862 records for deaths and 60,033 for injuries; 0 records have invalid casualty values retained as unavailable. These are quality notes, not story findings.
Map cells are coarse approximations, 0.05 degrees on each axis. Coordinate checks use a fixed California/Los Angeles map extent, not a historical municipal boundary. 72,745 records in the selected overview period are not plotted. Count rankings do not account for traffic or pedestrian exposure and do not establish risk or causation. California and Texas use different reporting systems; these counts should not be used for direct state risk comparisons.
| Year | Source rows | Eligible crashes | Observed through | Source revision / checksum |
|---|---|---|---|---|
| 2022 | 405,239 | 405,239 | 2022-12-31 | ccrs-core-2022-e27b7549d658d172fa70b59b SHA-256: 0b83a520f42aadac8c12462428633012a813a9734ea8e1cd5ba1e2a73807e15e Source modified: 2026-11-07T03:18:29.696852 |
| 2023 | 409,836 | 409,836 | 2023-12-31 | ccrs-core-2023-cc7a113533eb21cc97a23cc5 SHA-256: 0e67a9fc04d29d9a87f57fb82824818a1223947f02dad086b22ec8166ef4712c Source modified: 2026-11-07T03:14:20.974293 |
| 2024 | 417,173 | 417,173 | 2024-12-31 | ccrs-core-2024-37e10137e262fd26dd77dffa SHA-256: 16a8bf5a533cd129feb36d0356722799f36658d28a7333a90b2e6fa05de9802a Source modified: 2026-11-07T03:10:08.194695 |
| 2025 | 402,447 | 402,447 | 2025-12-31 | ccrs-core-2025-09dc8344ac724110a36c1b76 SHA-256: 11e1b395905750168b3c51e362f9938cf2752515bb47aa1723111c5146c488fe Source modified: 2026-11-07T03:06:01.978409 |
| 2026 | 312,851 | 312,851 | 2026-11-07 | ccrs-core-2026-29d9f5fc8fd9d8a4b357a006 SHA-256: 0ff12232181b104ec569f2905bf1f137675bd618f032daad64103a4f946cf875 Source modified: 2026-11-08T03:02:22.270495 |
Guide revision: 1ffe1c08-5d70-434b-ab0f-38b6c093f773
Source binding: 5628129b2946d2f45539ab983a1259f943e377272a8d513e096bcf520ef3ed1a
Prepared: 2026-11-08T03:13:17.290Z
CrashDNA is not affiliated with CHP or any government agency. Questions and corrections: alerts@crashdna.com.