Dover by the Numbers from AskDover
About 1 in 4 Dover crashes happens between 3 and 6 p.m.
Police reported 2,380 crashes inside the city in 2025, about 6.5 a day. Of those, 316 involved an injury and 5 were fatal.
- 2,380
- crashes reported inside Dover in 2025 (source).
- 24.7%
- of crashes from 2021 to 2025 happened between 3 and 6 p.m., just 12.5% of the day.
- 5
- fatal crashes in 2025. The state file lists 63 since 2010.
Each line adds up to 100%. The hour is the time the crash started, in Eastern time. Weekdays are Monday to Friday. Between 3 and 6 p.m., 26.0% of weekday crashes and 20.8% of weekend crashes happened.
Source: Delaware Public Crash Data, crashes inside the Census city limits. Download the yearly data (CSV).
See the numbers as a table
| Hour starting | Weekday crashes, share | Weekend crashes, share | All crashes, count |
|---|---|---|---|
| 12 a.m. | 1.3 | 3.6 | 219 |
| 1 a.m. | 0.8 | 3.1 | 157 |
| 2 a.m. | 0.5 | 2.2 | 109 |
| 3 a.m. | 0.5 | 1.6 | 89 |
| 4 a.m. | 0.4 | 0.9 | 62 |
| 5 a.m. | 1.0 | 1.1 | 119 |
| 6 a.m. | 2.0 | 1.4 | 214 |
| 7 a.m. | 4.3 | 1.4 | 423 |
| 8 a.m. | 5.4 | 1.8 | 527 |
| 9 a.m. | 4.2 | 3.3 | 466 |
| 10 a.m. | 5.3 | 4.4 | 595 |
| 11 a.m. | 6.5 | 5.9 | 743 |
| 12 p.m. | 7.7 | 7.2 | 886 |
| 1 p.m. | 7.8 | 7.4 | 900 |
| 2 p.m. | 8.3 | 7.9 | 958 |
| 3 p.m. | 8.6 | 7.3 | 965 |
| 4 p.m. | 8.9 | 6.9 | 987 |
| 5 p.m. | 8.5 | 6.6 | 938 |
| 6 p.m. | 5.2 | 6.5 | 648 |
| 7 p.m. | 3.8 | 5.9 | 506 |
| 8 p.m. | 3.3 | 3.9 | 398 |
| 9 p.m. | 2.5 | 3.9 | 331 |
| 10 p.m. | 1.8 | 3.2 | 249 |
| 11 p.m. | 1.4 | 2.6 | 199 |
About 2,136 crashes a year
Since 2010, the state's crash file has recorded an average of 2,136 crashes a year inside Dover. The yearly total has ranged from 1,812 in 2020 to 2,385 in 2023. In 2025 it recorded 2,380: 316 with injuries (13%), 5 fatal, and the rest classed as property damage only or not reportable (source).
The state's file runs through April 2026, so 2025 is the latest full year.
- 2,380crashes in 2025, about 6.5 a day
- 316crashes with injuries in 2025
- 5fatal crashes in 2025
Crashes in which someone was hurt but nobody died. Fatal crashes are in their own chart below. Year-to-year changes of this size can come from how crashes are reported and classed as well as from changes on the road, so we do not read a trend into them.
Source: Delaware Public Crash Data. Download the yearly data (CSV).
See the numbers as a table
| Year | All crashes | Property damage only | Injury | Fatal | Not reportable | No class recorded | On private property |
|---|---|---|---|---|---|---|---|
| 2,010 | 1,977 | 428 | 256 | 1 | 1,292 | 0 | 556 |
| 2,011 | 1,893 | 368 | 249 | 2 | 1,274 | 0 | 518 |
| 2,012 | 2,076 | 440 | 299 | 4 | 1,333 | 0 | 538 |
| 2,013 | 2,151 | 478 | 355 | 4 | 1,314 | 0 | 548 |
| 2,014 | 2,088 | 553 | 353 | 4 | 1,178 | 0 | 505 |
| 2,015 | 2,070 | 569 | 440 | 6 | 1,055 | 0 | 342 |
| 2,016 | 2,174 | 692 | 440 | 6 | 1,036 | 0 | 411 |
| 2,017 | 2,013 | 729 | 395 | 3 | 886 | 0 | 356 |
| 2,018 | 2,056 | 860 | 390 | 5 | 801 | 0 | 393 |
| 2,019 | 2,171 | 1,077 | 387 | 6 | 699 | 2 | 400 |
| 2,020 | 1,812 | 878 | 332 | 3 | 599 | 0 | 321 |
| 2,021 | 2,374 | 1,094 | 392 | 3 | 885 | 0 | 618 |
| 2,022 | 2,229 | 934 | 362 | 4 | 929 | 0 | 557 |
| 2,023 | 2,385 | 1,014 | 372 | 4 | 995 | 0 | 615 |
| 2,024 | 2,320 | 954 | 373 | 3 | 990 | 0 | 641 |
| 2,025 | 2,380 | 1,221 | 316 | 5 | 838 | 0 | 636 |
What the state counts as reportable
The state sorts every crash into one of four classes. A crash on a public road with only property damage is "reportable" when the damage is $2,000 or more. That limit was $1,500 before September 30, 2021. Smaller crashes, and most crashes on private property such as a parking lot, are classed "not reportable" (source).
Because of that rule and the 2021 change, we compare injury and fatal crashes across years and treat the total with care.
Weekday afternoons are the busiest time
From 2021 to 2025, 2,890 of Dover's 11,688 crashes happened between 3 and 6 p.m. That is 24.7%, about 1 in 4, in 3 hours of the day. The share was between 23.0% and 26.1% in every year from 2010 to 2025.
The busiest single hour is 4 to 5 p.m., with 987 crashes over the five years. Crashes that caused injuries or deaths follow the same pattern: 24.3% of them happened between 3 and 6 p.m.
Weekends look different. The weekday peak is sharper (26.0% in those 3 hours, against 20.8% on weekends), and weekends have more crashes in the first hours after midnight: 3.6% of weekend crashes happened between 12 and 1 a.m., against 1.3% on weekdays. Crashes are rarest from 2 to 5 a.m., at 2.2% of the total (source).
Tuesdays and Fridays have the most crashes, Sundays the fewest
Across 2021 to 2025, Dover averaged 7.2 crashes on a Tuesday and 7.1 on a Friday. Sundays averaged 4.6 (source).
Average number of crashes on each kind of day, counting every day in the period.
Source: Delaware Public Crash Data.
See the numbers as a table
| Day | Average crashes a day |
|---|---|
| Monday | 6.5 |
| Tuesday | 7.2 |
| Wednesday | 6.7 |
| Thursday | 6.5 |
| Friday | 7.1 |
| Saturday | 6.1 |
| Sunday | 4.6 |
Crashes peak in the fall, when evenings get dark
October is Dover's busiest month, at 7.5 crashes a day on average, and January the quietest, at 5.6. September (7.2), December (6.9) and November (6.8) follow (source).
Light matters in fall and winter. Among crashes on public roads with a known light condition, 43% in December happened in the dark, against 18% in June. Deer are part of the fall picture too: 95 of the 189 crashes with a deer as the main contributing factor, 50%, happened from October to December, 45 of them in November.
Each month is the average per day across the five years, so shorter months are not undercounted.
Source: Delaware Public Crash Data. Share in the dark counts crashes on public roads with a known light condition.
See the numbers as a table
| Month | Average crashes a day | Share in the dark |
|---|---|---|
| January | 5.6 | 38% |
| February | 5.7 | 34% |
| March | 6.1 | 27% |
| April | 6.3 | 22% |
| May | 6.5 | 20% |
| June | 6.1 | 18% |
| July | 6.1 | 20% |
| August | 5.9 | 20% |
| September | 7.2 | 24% |
| October | 7.5 | 33% |
| November | 6.8 | 40% |
| December | 6.9 | 43% |
Inattention is the most common recorded cause
For each crash, the officer records one main contributing circumstance. From 2021 to 2025, 29.5% of crashes list the driver's inattention, distraction or fatigue. Following too close (10.2%) and failing to yield (8.1%) come next, then improper backing (7.1%), most of which happens on private property (source).
This is the officer's note at the scene, one per crash. It is not a finding of fault, and 16% of crashes have no circumstance recorded.
Most crashes happen on ordinary days. Among crashes with a known condition, 86% were on a dry road, 80% in clear weather and 70% in daylight.
The six most common circumstances. Other circumstances, 451 crashes marked other and 1,906 unknown, are not shown. The share is out of all 11,688 crashes in the period.
Source: Delaware Public Crash Data, field "primary contributing circumstance".
See the numbers as a table
| Circumstance | Crashes | Share of crashes |
|---|---|---|
| Driver inattention, distraction, or fatigue | 3,451 | 29.5 |
| Following too close | 1,189 | 10.2 |
| Failed to yield right of way | 946 | 8.1 |
| Improper backing | 835 | 7.1 |
| Improper lane change | 519 | 4.4 |
| Driving in a careless or reckless manner | 460 | 3.9 |
Fatal crashes more often involve darkness, pedestrians and alcohol
From 2010 to 2025, 1.6% of crashes on public roads involved a pedestrian, but 22 of the 63 fatal crashes did, 34.9%. Crashes in the dark are 28.9% of crashes with a known light condition and 58.7% of fatal ones. Suspected alcohol use is noted in 4.1% of crashes and 34.9% of fatal ones (source).
Motorcycles and scooters were in 1.7% of crashes and 22.2% of fatal ones. Bicycles were in 197 crashes, 3 of them fatal.
With only 63 fatal crashes in 16 years, these shares can move a lot from a few cases. They show who is most at risk when a crash does happen. They do not show the cause of any one crash. Suspected alcohol use covers anyone involved, including a pedestrian or cyclist.
The open dot is all 18,063 crashes on public roads. The filled dot is the 63 fatal crashes. A crash is counted once for each factor it has.
Source: Delaware Public Crash Data. Public roads means crashes the state classes as property damage only, injury or fatal.
See the numbers as a table
| Factor | All crashes, count | All crashes, share | Fatal crashes, count | Fatal crashes, share |
|---|---|---|---|---|
| In the dark (known light condition) | 5,107 | 28.9 | 37 | 58.7 |
| A pedestrian involved | 294 | 1.6 | 22 | 34.9 |
| Suspected alcohol use | 746 | 4.1 | 22 | 34.9 |
| A motorcycle or scooter | 312 | 1.7 | 14 | 22.2 |
63 fatal crashes in 16 years
The state's file lists 63 fatal crashes inside Dover from 2010 to 2025, between 1 and 6 a year. In 14 of the 16 years there were 3 to 6. There were 5 in 2025 (source).
The federal Fatality Analysis Reporting System (FARS) counts people as well as crashes. For 2010 to 2024 it lists 61 fatal crashes in the city, in which 66 people died. From 2020 to 2024, 18 people died (FARS).
The two files are close but not identical. FARS has 61 fatal crashes for 2010 to 2024 and the state file has 58. We matched each FARS crash to the state file by date and place. 55 match a fatal crash. 4 are listed as injury crashes in the state file, and 2 have no match within 500 meters and a day. The state's count may therefore run a little low.
FARS does not have 2025 yet. The two lines match exactly in 10 of 15 years and are within one crash in 13.
Sources: Delaware Public Crash Data and NHTSA FARS, crashes inside the Census city limits.
See the numbers as a table
| Year | State file, fatal crashes | FARS, fatal crashes | FARS, people who died |
|---|---|---|---|
| 2,010 | 1 | 1 | 1 |
| 2,011 | 2 | 2 | 2 |
| 2,012 | 4 | 4 | 5 |
| 2,013 | 4 | 6 | 7 |
| 2,014 | 4 | 4 | 4 |
| 2,015 | 6 | 6 | 6 |
| 2,016 | 6 | 6 | 6 |
| 2,017 | 3 | 2 | 2 |
| 2,018 | 5 | 7 | 7 |
| 2,019 | 6 | 6 | 8 |
| 2,020 | 3 | 2 | 2 |
| 2,021 | 3 | 3 | 4 |
| 2,022 | 4 | 5 | 5 |
| 2,023 | 4 | 4 | 4 |
| 2,024 | 3 | 3 | 3 |
| 2,025 | 5 | Not yet out | Not yet out |
About 1 in 4 crashes is on private property
The state flags crashes on private property, which includes private parking lots. From 2021 to 2025, 3,067 of Dover's 11,688 crashes were flagged that way, 26%, or 1 in 4. The state classes most of them (91%) as not reportable, and many have no weather or light condition recorded (source).
Were you in a crash? The state says official crash reports are confidential, and that authorized parties can contact the police agency that took the report for an official copy (state data page).
How we did this
- Crashes: the state's Public Crash Data, one row per crash with a location, from the Department of Safety and Homeland Security, downloaded on 2026-11-06. It runs from 2009 through April 2026. We used 2010 to 2025 for yearly counts and 2021 to 2025 for patterns (data).
- Inside Dover: the file has no city field, so a crash counts as in Dover when its point falls inside the Census city limits, which are today's boundary applied to every year. The state's own mapping server gave the same count as our script for all 16 years (Census).
- Times: the state stores times in UTC. We converted them to Eastern time before counting hours and days, and checked the result against the state's own light-condition field.
- Cross-checks: fatal crashes against the federal FARS file, matched case by case. Dover Police's 2025 annual report says officers responded to 2,076 collisions with 4 fatal. The state file has 2,380 and 5 for the city. It covers crashes reported by any police agency and uses the Census boundary, so it can differ from one department's count (Dover Police).
- AI and review: this report is AI-assisted. An AI model wrote the text and code from a computed data file, and a person reviews it before publishing. A script checks every number on the page against that file. The scripts and saved raw data are kept with this report.
What this data can't tell you
- Which roads or intersections are worst. The file has no street names, and we report no crash locations.
- How risky driving is. It counts crashes, not miles driven or traffic, so a busy hour is not necessarily a riskier hour.
- Why a crash happened. The contributing circumstance is an officer's note at the scene, not a finding of fault.
- Crashes nobody reported. Small crashes that never reach police are not in the data, and the state says the file is the best available information, not an official record of any one crash.