For decades, the repair-versus-total decision was arithmetic at the margins. Most damaged vehicles were obviously repairable, a small share were obviously gone, and appraisers spent their judgment on a thin band in between. That world has quietly ended.
The industry’s largest claims dataset, analyzed in CCC Intelligent Solutions’ 2026 Crash Course report, put total loss frequency at a record 23.1 percent in 2025, the highest ever recorded, with average repair costs climbing toward five thousand dollars. Calibrations for camera and sensor systems now appear on more than a quarter of repairable appraisals, up 30 percent in a single year, each adding hundreds of dollars and another step where documentation can fail. Meanwhile the fleet keeps aging, which pushes vehicle values down while repair complexity pushes costs up. Those two curves are converging, and every point of convergence turns another band of vehicles into borderline calls.
Nearly one in four claims now ends in a total loss, which means the borderline is no longer the margin. It is the job. And that changes which moment in the claim carries the most leverage. The industry has spent years optimizing the back half of the file: repair cycle time, shop throughput, supplement management. But when a growing share of vehicles sits within a calibration or a hidden-damage finding of the total threshold, the decisive act is the first appraisal. A thorough one, with complete photographs and calibration requirements captured up front, routes the vehicle correctly on day one. A thin one sends a doomed repair into a shop, burns weeks, and arrives at the same total loss with rental, storage and teardown costs attached.
This is why supplement frequency deserves more attention than it gets. A supplement is often described as a cost event. On a borderline vehicle it is really a decision-quality event: evidence that the original appraisal missed enough of the damage picture that the repair-versus-total call was made on incomplete information. As repair complexity rises, every missed calibration line and undocumented sensor is not just dollars added later. It is a wrong turn taken earlier.
The operational conclusion runs against the industry’s reflexes. The instinct under cost pressure is to make the first inspection faster and cheaper. The math now argues for the opposite: make the first inspection more complete, because the expensive error is no longer a slow appraisal. It is a wrong one. Front-loading appraisal quality is the cheapest severity control available in a market where the totaled quarter of the claim volume is decided, one borderline vehicle at a time, by how good the first look was.
Vehicle technology will keep compounding this. Collisions themselves are getting rarer, with Allstate’s driver data showing the average US driver now goes 10.86 years between crashes, but the vehicles involved carry ever more sensor content spreading through the fleet faster than vehicle values are rising. Fewer claims, each more complex and closer to the total threshold: the borderline band widens every model year. The organizations that treat the first appraisal as the highest-skill, highest-leverage step in the claim, rather than a commodity task to be compressed, are positioning themselves on the right side of arithmetic that is not going to reverse.
Sources
- CCC Intelligent Solutions, Crash Course 2026 report (total loss frequency, average repair cost, calibration frequency).
- Allstate America’s Best Drivers data (collision frequency of once every 10.86 years for the average US driver).
