Slick Nick Law

Crashes by California county

All 58 counties, ranked by reported crashes across 2023, 2024 and 2025. Read it as a ranking by size before you read it as a ranking by risk: Los Angeles County accounts for about 26 per cent of the statewide total, which is close to its share of the state's population. Nothing in this table is adjusted for how many people live somewhere or how far they drive.

Reported crashes, people killed and people injured by county, 2023–2025 combined
# County Crashes Killed Injured Pedestrian Bicycle Motorcycle 2026 so far part year, not ranked
1 Los Angeles 316,403 2,345 177,267 11,760 6,374 9,084 60,457
2 Orange 89,209 637 53,612 2,759 3,446 2,771 18,051
3 Riverside 83,617 864 46,018 1,998 1,330 2,775 15,285
4 San Bernardino 78,539 1,050 44,887 1,889 1,078 2,417 15,361
5 San Diego 75,348 834 57,723 3,191 2,221 4,746 15,214
6 Sacramento 61,273 525 35,424 1,932 1,695 1,539 12,166
7 Alameda 58,351 311 28,195 1,914 1,408 1,460 8,877
8 Santa Clara 42,686 300 26,496 1,543 1,492 1,120 7,839
9 San Joaquin 33,475 357 17,683 778 663 680 6,715
10 Fresno 32,102 464 16,185 953 599 671 4,977
11 Contra Costa 31,895 238 14,763 904 679 795 6,245
12 Kern 29,336 545 17,454 918 478 747 6,034
13 Ventura 25,974 198 16,008 664 716 811 5,476
14 Stanislaus 21,820 242 14,185 680 506 552 4,246
15 San Mateo 18,852 97 10,460 705 588 529 3,687
16 Tulare 18,522 243 8,763 371 264 369 3,142
17 San Francisco 18,392 121 12,956 1,822 1,079 913 3,588
18 Solano 16,609 133 8,135 412 316 415 3,364
19 Monterey 15,005 124 7,414 454 276 303 2,834
20 Santa Barbara 12,881 110 7,627 384 735 340 2,548
21 Placer 12,168 114 7,093 210 264 431 2,387
22 San Luis Obispo 11,910 81 4,604 203 242 303 2,284
23 Sonoma 11,512 117 7,229 336 304 382 2,541
24 Merced 11,164 151 6,150 264 183 244 1,828
25 Santa Cruz 8,745 64 4,149 264 373 360 1,635
26 Marin 7,822 40 3,970 234 312 280 1,513
27 Yolo 7,476 69 4,044 169 278 169 1,394
28 Madera 6,691 95 3,366 150 88 147 1,434
29 El Dorado 6,537 61 3,161 103 110 311 1,320
30 Butte 5,435 96 3,040 188 120 232 920
31 Napa 4,866 50 2,557 122 91 185 999
32 Kings 4,797 91 2,575 99 74 79 765
33 Shasta 4,648 75 2,775 152 86 205 932
34 Humboldt 4,584 71 2,093 139 96 114 801
35 Mendocino 3,614 53 1,384 51 43 137 737
36 Nevada 3,417 43 1,590 41 25 117 673
37 Sutter 3,212 57 2,123 91 70 85 645
38 Yuba 3,207 49 1,765 81 61 128 683
39 Imperial 3,196 82 2,295 58 35 75 771
40 Tuolumne 2,795 43 1,370 41 14 146 553
41 San Benito 2,440 32 1,369 54 48 75 476
42 Lake 2,417 61 1,331 76 20 106 465
43 Tehama 2,192 52 1,123 36 22 89 414
44 Calaveras 1,837 45 973 21 5 106 325
45 Siskiyou 1,521 41 775 6 5 29 313
46 Amador 1,295 26 790 16 2 90 289
47 Inyo 1,083 35 440 16 22 52 211
48 Colusa 1,030 26 507 11 0 20 204
49 Lassen 989 18 518 16 2 34 197
50 Del Norte 980 18 583 23 20 24 184
51 Mono 818 12 348 1 1 36 178
52 Glenn 809 30 480 8 0 16 164
53 Trinity 764 16 320 9 7 45 109
54 Plumas 736 11 367 14 2 33 191
55 Mariposa 686 6 330 7 1 31 138
56 Modoc 389 14 163 1 1 11 75
57 Sierra 216 3 131 1 1 31 52
58 Alpine 214 3 154 1 1 16 49
Statewide 1,228,560 11,689 699,318 39,344 28,980 38,011 234,955

The ranking and every total above cover covers 2023–2025. The 2026 figures are separate. They run from January to August 2026 — every whole month the state has published so far — and they are not part of the ranking or of any total above. Police agencies file at different speeds, so putting a part-reported year in a ranking would order agencies by how fast they file rather than roads by how often they are hit.

Recent months are still filling in, so treat them as a floor rather than a count. August 2026 held about 48% of what that month averaged across 2023–2025 when this data was compiled, so its real total will be higher. September 2026 is left out entirely: the state's file stops on 2026-09-01, so that month holds 15 records — a part of a month rather than a month that is late being filed.

Reading the table

The gap between the top and the bottom is enormous — Los Angeles County recorded 316,403 crashes over 2023–2025 and Alpine County recorded 214. Almost all of that gap is population. Orange County, second on the list, is not second by a small margin.

The killed and injured columns are people, not crashes. A single crash can contribute several to either column, or none. That is why a county can rank higher for deaths than for crashes: rural counties with fast roads tend to record fewer collisions and more severe ones.

The statewide row is slightly larger than the sum of the county rows, because a small number of records carry a county code outside the valid range and are left out of the county breakdown rather than assigned to a county. The methodology gives the exact counts.

Download it

California crash counts by county (CSV) — every county, every year, free and ungated. Citation requested, not required: Slick Nick Law, from CHP CCRS data.

Sources: California Crash Reporting System (CCRS), California Highway Patrol; county codes from California Secretary of State, California County Codes. Definitions and limitations: methodology.

Common questions

Why is Los Angeles County so far ahead?
Because it is far larger than anywhere else in the state. This table ranks by count, not by rate, so it ranks by population and traffic volume as much as by anything else. A county near the top is not necessarily more dangerous per mile driven than one near the bottom.
What does "crashes" count exactly?
One reported crash record in the California Crash Reporting System, for the calendar year shown. Records marked deleted are excluded. It counts crashes, not vehicles and not people.
Why do killed and injured not add up to the number of crashes?
They are different things. A crash can injure several people or none at all. "People killed" and "people injured" are sums of those columns across the crash records, so they can be higher or far lower than the crash count.
Where do the county names come from?
The dataset stores a county code, not a name. The mapping is the state’s own published list of California county codes, checked against the city names in the data before use. The methodology page shows the source and the check.