Incentive-based demand response is a program where participant users are paid to reduce energy consumption from an established baseline. Counter-factual models to estimate the baselines are vulnerable for gaming. In this paper, a novel demand response contract between a user and an aggregator is developed to induce individual rationality (voluntary participation) and asymptotic incentive-compatibility (truthfulness) through the probability of call, which is the chance of a consumer to be selected by the aggregator to serve as demand response resource at a given period. In this approach, a consumer self-reports his baseline and reduction capacity, given a payment scheme that includes cost of electricity, incentive price, and penalty caused by any deviation between self-reported and actual energy consumption. Another important feature of this approach, different from the classic solutions, is that a participant agent does not require reporting marginal utility (energy preference), and only announces information in terms of energy. A two-stage stochastic programming problem is proposed from the demand side to understand the consumer rational decisions under this contract. As result, the aggregator decides randomly what users are called to perform the energy reduction in order to manage the truth-telling behavior of each agent through the probability of call. Mathematical proofs and numerical studies are provided to demonstrate the properties and advantages of this contract in limiting gaming opportunities and in terms of its implementation.