* This blog was originally published on WayPoint, Waymo’s official blog, on July 7, 2026
Not all miles are created equal. Navigating a highway commute on a Tuesday morning is fundamentally different from driving through downtown nightlife at 2:00am on a weekend. Waymo has conducted two new studies, peer-reviewed and accepted for publication in the journal Traffic Injury Prevention, which aim to close this gap by diving into two critical factors often overlooked in crash risk analysis: time and location.
Waymo has historically benchmarked its safety record against human drivers using regional averages, but says this approach misses important variables, such as time of day. The company gives an example: if its vehicles drive disproportionately at night in dense urban areas while human drivers mostly travel during daylight on familiar routes, a simple overall comparison would be misleading.
To enable a more accurate comparison, in both studies, researchers paired human crash databases with granular traffic volume data to map exactly when and where humans drive. By unlocking the ability to break down human crash data by location and time, the company has built unprecedented, highly precise benchmarks against which to evaluate Waymo’s performance.
Feng Guo, professor of statistics at Virginia Tech and lead data scientist for the Virginia Tech Transportation Institute (VTTI) said, “Evaluating autonomous vehicle safety requires moving past abstract, aggregated national averages. Meaningful safety assessment must be context-specific, accounting for the disparities in risk across different regions, infrastructure types and times of day. This new research advances understanding of autonomous vehicle safety, by developing a framework to establish comparable human driver benchmarks that incorporate these critical spatial and temporal conditions.”
The fatal crash baseline
Waymo’s research across the top 50 most populous US urban areas revealed a massive disparity in fatal crash involvement rates between different regions in the country.
For example, on surface streets, human drivers in Memphis were involved in fatal crashes at a rate 8.4 times higher than drivers in Boston. Relying on a single national average to judge safety would be unfair in both cities – it overestimates the risk of driving in Boston by three times, while underestimating the hazards in Memphis by the same threefold margin.
Furthermore, the road type plays a major role: across all 50 regions, driving on surface streets carries a fatal crash rate 2.3 times higher than driving on freeways. This confirms a pattern identified in Waymo’s previous research, which has consistently shown that urban streets present a higher crash risk than freeways.
Human fatal crash rates
While Waymo has accumulated the immense mileage required to show statistically significant reductions in serious injuries, fatal crashes are too rare to yield immediate, direct comparisons. As the company works toward building scientific consensus, establishing these localized fatal crash baselines proactively will help create a clear framework to evaluate autonomous safety as the industry matures.
Waymo’s research shows that human fatal crash risk surges during late-night hours and weekends. Fatigue, darkness and impaired driving completely change the safety landscape.
Waymo’s second study extended the company’s earlier location-based research by adding time-of-day and day-of-week analysis across its main operating areas: Maricopa County (Phoenix), San Francisco, Los Angeles and Travis County (Austin). This allows the company to compare its performance against human driving benchmarks matched more precisely by time as well as location.
The research found that human crash rates increase sharply between midnight and 3:59am, especially on weekends. Because overnight driving makes up only about 1.5% of total human mileage, the company says these high-risk hours are obscured in conventional crash statistics by the much larger volume of daytime driving. Within that midnight-to-4am window specifically, Waymo reports crash rates two to five times higher than average on weekdays, and 2.5 to six times higher on weekends. The company attributes this to combining crash records with detailed, hour-by-hour traffic volume data, allowing risk to be measured per hour rather than just by raw crash counts.
“The data points to a significant increase in crash risk during late-night and weekend hours, when road safety is most unpredictable and impaired driving is most prevalent,” said Jonathan Adkins, chief executive officer of the Governors Highway Safety Association (GHSA). “GHSA has long recognized the potential of autonomous technology to intervene when human decision-making is impaired, helping prevent behavior-related crashes and save lives.”
Road safety when it matters most
Waymo notes that as a ride-hailing service, it serves a large share of riders during late-night hours when nightlife activity peaks and alternative transportation options are limited. The company states its fleet drives proportionally about four times more overnight miles than the average human driver, placing it in the highest-risk driving windows more often. Despite this, Waymo reports its fleet had lower crash rates than the human benchmark across every time window studied, and says a substantial share of its overall safety advantage comes specifically from these higher-risk night and weekend periods.
When comparing the Waymo Driver’s real-world performance across 127,000,000 autonomous miles (204,000,000km) – and regardless of fault – against a human driver navigating the same combination of locations, days of the week and times of day, the study found that Waymo was involved in 359 fewer crashes with injuries. Crucially, 189 (53%) of those avoided crashes were during the overnight hours between 8:00pm and 3:59am.
While Waymo’s most recent Safety Impact Hub analysis features data from over 220,000,000 miles (354,000,000km), the company believes the findings from this foundational study remain relevant and representative at its current scale.
Waymo concludes that these two papers demonstrate the importance of accounting for location- and time-specific risk factors when evaluating driving safety. The company argues that using dynamic benchmarks, adjusted for where and when driving occurs, allows for more accurate assessment of real-world safety impact, tailored to the specific conditions of each city. Waymo has stated that it is publishing these findings and its methodology in the hope of encouraging a shared, industry-wide approach to safety evaluation.