How likely is "likely to reoffend" in sex offender civil commitment trials?

Many sexually violent predator (SVP) laws are ambiguous regarding the degree of reoffense risk that would indicate that an offender is sufficiently "likely to reoffend" to justify civil commitment. We review how SVP statutes operationalize likelihood of reoffending. We then examine what likelihood of recidivism actual SVP jurors considered to indicate that an offender was likely to reoffend. Real jurors (N = 153) from 14 actual SVP hearings completed a questionnaire after deliberating to a verdict. Most jurors (81.7%) considered a 15% estimated chance of recidivism to mean that the respondent was "likely" to reoffend, and many (53.6%) even considered a 1% chance to indicate likely reoffense. Jurors who heard lower risk estimates in trials were more likely to report that a low chance of recidivism (as low as 1%) indicated an offender was likely to reoffend. Results suggest that jurors view risk more in terms of the severity of potential harm than in terms of strict statistical probability. Results also suggest that when laws give jurors discretion to define tolerable risk, jurors consider even a statistically low degree of risk intolerable.