An Insurance Journal analysis published August 17, 2026 documents the shift: premises liability has joined trucking and commercial fleet as a primary driver of nuclear verdicts above $10 million. The median jury award in cases exceeding $10 million grew from $20 million in 2015 to $27 million by 2020, per Advisen loss data, and the precipitating events are not exotic. Wet floors, uneven pavement, inadequate lighting, loose handrails, and delayed maintenance response are the incident types now producing eight-figure verdicts.
A $15.8 million Oregon jury award against Topgolf for a child’s skull fracture and traumatic brain injury at its Hillsboro entertainment complex illustrates the new severity floor. That claim type, a guest injured at a commercial recreation venue, has historically settled in the low-to-mid six figures.
Who it affects
Self-insured employers with large physical footprints carry the highest premises liability frequency and now face nuclear verdict severity on any high-injury claim. Hospitals, school districts, municipalities, retailers, and hospitality operators are the primary exposed categories. A hospital with 10 campuses, a county with 40 public buildings, or a regional retailer with 80 locations generates premises incidents at a rate that was manageable under frequency-severity assumptions built from pre-2018 data.
The SIR layer is the structural problem. Self-insured employers with GL retentions between $500,000 and $5 million set those figures assuming premises claims rarely exceed $2 million. At a $27 million median for nuclear premises verdicts, a single traumatic brain injury or spinal cord case will exhaust the entire retention and activate excess coverage. For public entities operating below that median, the program is exposed on any claim that reaches verdict territory.
The reserve mechanism
Three effects follow from this severity shift.
First, case reserves on severe premises injury claims are calibrated to pre-2018 frequency-severity patterns. For claims involving traumatic brain injury, spinal cord damage, or wrongful death at a covered premises, those reserves are understated relative to current verdict risk, often by a factor of three to ten.
Second, expected claim ratio assumptions in GL actuarial models predate the nuclear verdict era in premises cases. Accident years 2021 and later are generating emergence patterns inconsistent with those assumptions, producing IBNR understatement that grows with each late-developing severe claim.
Third, GL premises cases now routinely take three to five years to reach verdict as plaintiff counsel extends litigation timelines on high-stakes matters. Tail factors derived from pre-2019 development triangles do not reflect this extended pattern. The same revision problem that hit fleet auto tail factors after 2019 is now arriving in GL premises, with a lag of several years.
The plaintiff bar is applying the “preventable failure” narrative to premises cases that drove trucking nuclear verdicts higher. Failure to inspect, failure to maintain records, failure to correct known hazards, and absence of written protocols are standard evidentiary themes in GL premises trials. Insurance Journal identifies delayed claims reporting and incomplete maintenance documentation as the two most common internal failures that convert a manageable claim into a high-severity outcome.
What this means for your next review
Ask your actuary whether the severity assumption in the GL model has been updated for nuclear verdict potential in premises claims, or whether it remains calibrated to historical averages below $2 million. Confirm that open case reserves for severe premises injuries, specifically traumatic brain injury, spinal cord injury, and wrongful death, are stress-tested against a $15 million to $25 million scenario. Ask what tail factor applies to GL premises claims and whether it reflects the post-2018 verdict environment. The public entity GL IBNR framework describes where these assumptions typically appear in the actuarial report for government and institutional self-insureds.
Combined ratios for GL were already running above 100% entering 2026. Premises severity is not a future problem; it is appearing in current development triangles as claims initially reserved in the $200,000 to $500,000 range reopen at five to ten times that figure on their way to verdict or settlement.
Sources
- Insurance Journal: “Premises Liability Emerging as Significant Driver of Nuclear Verdicts” (August 17, 2026)
- Advisen loss database: nuclear verdict median data cited in Insurance Journal analysis (Advisen Ltd.)
- CaseGlide: “What Counts as a Nuclear Verdict? Definition and 2026 Data”
- Sedgwick: “Inside the Verdict: What Is Driving the Rise in Nuclear and Thermonuclear Awards?” (2026)