Marathon Strategies’ “Corporate Verdicts Go Thermonuclear: 2025 Edition” recorded 135 nuclear verdicts totaling $31.3 billion in 2024, a 52% increase in count and 116% jump in total dollars over 2023. Five states accounted for $23.8 billion, or 76%, of that total: Nevada ($8.4 billion), California ($6.9 billion), Pennsylvania ($3.4 billion), Texas ($3.0 billion), and New York ($2.1 billion).
The concentration went beyond dollars. Texas led in verdict count with 23, followed by California (17) and Pennsylvania (12). At the upper tail, 49 verdicts exceeded $100 million (“thermonuclear” verdicts), nearly doubling the 27 recorded in 2023. Five verdicts crossed $1 billion. The median nuclear verdict reached $51 million, up 15.9% year over year.
Who it affects
Self-insured fleet operators, commercial auto programs, and GL programs with multi-state exposure face the sharpest risk from this geographic concentration. A trucking company running corridors through Nevada and Pennsylvania carries a categorically different severity profile than one concentrated in Ohio or Minnesota. Public entities with vehicle fleets operating in high-verdict jurisdictions face the same dynamic.
Captive programs and excess casualty towers anchored to national blended factors are also exposed. Lockton’s May 2026 analysis flagged $1.8 billion in commercial auto adverse development concentrated in accident years 2022 and 2023, a pattern consistent with national severity picks that lag jurisdiction-specific reality.
The reserve mechanism: jurisdiction-specific severity factors
Geographic concentration makes blended national development factors unreliable. When five states drive three-quarters of all nuclear verdict dollars, a severity assumption built from a national blend systematically underreserves the high-concentration jurisdictions and overreserves the rest. The blended national pick can look adequate in total while hiding offsetting errors at the state level.
The thermonuclear tail compounds the problem. At a $51 million median and 49 verdicts above $100 million, the right tail of the severity distribution has fattened enough that a single jurisdiction’s verdict experience can swing an entire program year. Frequency and severity have diverged: traffic fatalities fell to a record low in 2024 while verdict severity climbed for the fifth consecutive year.
State courts produced $20.1 billion across 85 verdicts compared to $11.2 billion from 50 federal cases, reinforcing that venue selection and transfer motions are material cost drivers for self-insured programs with multi-state exposure.
What this means for your next review
Before your next reserve study, confirm whether your actuary is applying jurisdiction-specific severity factors or a blended national trend. For programs with material exposure in any of the top five states, a blended pick is likely underreserving the tail. Three questions to raise:
- Are our auto liability and GL development factors split by jurisdiction, or are we applying a single national severity trend?
- For Nevada and Pennsylvania, the two fastest-growing verdict states by dollar volume, have we stress-tested severity assumptions against the thermonuclear tail?
- Should we model a $100 million-plus scenario in our highest-exposure state and test our SIR and excess attachment against it?
Marathon’s data will update next year with 2025 results. Watch whether Florida’s 2023 tort reform continues to suppress verdict volume (Florida dropped to tenth in dollar volume in 2024) and whether Georgia’s and Oklahoma’s 2025 reforms produce a similar bend.