Abstract Background Most physicians sometimes apply therapies without specific active ingredients. Although patients seem to judge such placebo treatments as acceptable under certain circumstances, deception is still an ethical problem. Open-label placebos (OLPs) might be a promising approach to solve this dilemma. This study compared general acceptance and outcome expectations of OLPs and deceptive placebos (DPs). Methods In an experimental online study, 814 participants read a case vignette of a person with insomnia receiving a pill. The participants were then randomly allocated into two groups, where the second part of the vignette described the pill as either a deceptive placebo (DP group) or as an open-label placebo (OLP group). The Credibility/Expectancy Questionnaire (CEQ) assessed outcome expectations after the first (pre-assessment) and the second (post-assessment) parts of the vignette. Treatment acceptance was measured at post-assessment. Data from 798 participants were analyzed by a mixed multivariate analysis of variance (MANOVA), t-tests, and post-hoc mediation analyses. Results The MANOVA revealed a significant group main effect and a significant time × group interaction effect. At post-assessment, outcome expectations were higher in the DP group than in the OLP group. Acceptance of the placebo treatment was also higher in the DP group than in the OLP group. Mediation analyses confirmed that higher acceptance in the DP group was mediated by higher expectations. Conclusions When laypersons read about placebo treatment, their outcome expectations toward DPs were higher than toward OLPs. Surprisingly, the application of DPs was rated as more acceptable than OLPs. This result might be explained by indirect effects of treatment expectations.