Driving risk is evolving, and driving behavior data reveals what’s changing

Key takeaways

• Driving risk is dynamic and can shift faster than traditional annual or semiannual pricing cycles can reflect.
• Macro forces like return-to-office mandates, extreme weather, and economic conditions can change exposure and concentrate risk in new time windows.
• Lagging indicators and static proxies can miss short-term spikes in risky behaviors (e.g., hard braking, distraction) that matter to near-term loss exposure.
• Misaligned risk assumptions compound into inaccurate pricing, hidden portfolio blind spots, and adverse selection over time.
• Continuously updated mobility data helps insurers detect emerging changes early and align pricing, underwriting, and segmentation with real-world conditions.

Introduction: Shifts in driving risk patterns

For decades, auto insurance risk assessment relied on relatively static assumptions: how often people drive, when they commute, and which behaviors signal higher risk. But those assumptions are no longer holding. Driving risk is changing faster than traditional models can absorb — and the consequences of lagging are showing up in losses and missed opportunities.

Our ongoing driving intelligence reporting makes this clear: Driving behavior is not only changing, it is changing unevenly, sporadically, and in ways that traditional proxies may not detect. For insurers, keeping one’s finger on the pulse of real-world risk now requires continuous visibility into mobility patterns — not ad hoc recalibration.

The question is no longer whether driving risk data is changing. The question is whether your organization has the visibility to recognize those changes before they show up as vulnerable portfolios, losses, or emergency rate filings.

Risk changes — even when premiums don’t

One of the most important takeaways from Arity’s ongoing mobility data reporting is that risk is dynamic. It shifts with macro forces like return-to-office mandates, extreme weather events, gas prices, and changing infrastructure usage — and those shifts can happen faster than annual or even semiannual pricing cycles can reflect.

Arity’s recent analyses show that:

• Commutes are growing longer, increasing exposure even among drivers whose historical profiles looked stable.
• Due to return-to-office mandates, rush hour patterns are evolving, concentrating risk in new time windows.
• With the 2026 gas price shock, certain cohorts are driving less while others are maintaining their typical mileage, which could lead to risky driving for those on newly spacious roads.

In other words, the risk environment a policy was priced for six months ago may already be outdated today.

Why traditional risk signals are falling short

Many insurers still rely heavily on lagging indicators: historical loss experience, annual mileage estimates, or proxies. While these inputs remain useful, they may fail to capture rapid changes in how and when risk manifests.

Our data shows that behavior-based risk indicators — such as sudden acceleration, hard braking, distraction, and response to extraordinary events — can fluctuate significantly across time, geography, and context.

For example, extreme weather events trigger sharp but temporary spikes in risky behaviors like hard braking and phone distraction. Our driving behavior data shows that, in Los Angeles, after the Palisades fire prompted an evacuation order, hard braking increased by 25.3% — a clear indication of congested roads as people fled the fire.

These kinds of patterns are invisible in static models but highly relevant to near-term loss exposure. Without mobility data that updates continuously, these shifts remain invisible.

What’s at stake when driving risk is incorrectly assessed

There is a cost to getting risk wrong. When driving risk data changes but pricing and underwriting assumptions do not, insurers face three compounding dangers.

1. Inaccurate pricing

Price too high, and low-risk drivers subsidize the system or leave. Price too low, and carriers unknowingly absorb increasing exposure. As commute times become less predictable, exposure-based assumptions become less relevant — eroding rating accuracy even if loss ratios initially appear stable.

2. Dangers lurk in the portfolio

Changing behavior patterns can introduce new pockets of risk that never surface in historical loss data until after claims occur. Mobility data reveals early warning signs — such as rising stop‑and‑go congestion or driving volatility during seasonal and climate‑related disruptions — before they result in losses.

Without this visibility, insurers are effectively underwriting based on assumptions.

3. Adverse selection over time

When risk is mispriced, the wrong drivers self-select into or out of coverage. Higher-risk drivers remain underpriced and retained, while safer drivers may feel unrewarded for their safe driving behavior. Over time, adverse selection compounds, weakening portfolio performance.

How mobility data can help

It’s not that driving has changed, it’s that the pace of change has accelerated. Driving behavior now responds quickly to external forces, from climate events to policy decisions like congestion pricing to economic shifts like the rapid rise in gas prices.

Mobility data enables insurers to:

• Detect changes in driving exposure and behavior as they emerge
• Update risk signals continuously rather than episodically
• Align pricing, underwriting, and segmentation with real-world conditions

This is not about replacing actuarial rigor but about supporting it with data that reflects how people actually drive today, not how they drove last year.

A strategic imperative for auto insurance leaders

For directors and senior leaders, the question is no longer whether driving risk data is changing. The question is whether your organization has the visibility to recognize those changes before they show up as vulnerable portfolios, increased losses, or emergency rate filings.

As mobility continues to evolve, insurers that rely solely on backward-looking indicators will increasingly operate with delayed signals. Those that integrate forward-looking mobility data will be better positioned to price accurately, surface emerging risks, and maintain a healthier book of business as conditions shift.

Driving risk data is changing. The auto insurers who adapt with it will define the next era of risk and pricing precision.