Self-efficacy is perhaps the most important parameter associated with behavioral changes. The main aim of this study was to provide insight into the diet and exercise self-efficacy of Greek adolescents and how they could be modified via a multilevel multicomponent school-based lifestyle intervention. Secondary aims were to study the associations of students’ dietary and exercise self-efficacy indices with their anthropometric and sociodemographic parameters. A representative sample of the adolescent population in Attica, consisting of 1610 adolescents aged 12–17 years, recruited from 23 public high schools in three municipalities of the Attica area in Greece, received a three-component lifestyle educational intervention for health promotion and underwent screening for characteristics of metabolic syndrome with the use of portable telemedicine. All assessments and anthropometric measurements were performed at baseline and after the 6-month intervention. Anthropometric measurements included body mass index, waist circumference (WC), waist-to-height ratio (WHtR) and waist-to-hip ratio (WHR). Assessment tools included the Self-efficacy for Diet and the Self-efficacy for Exercise questionnaires, as well as the Mediterranean Diet Quality Index in Children and Adolescents (KIDMED). Analysis included 1020 adolescent students (421 males and 599 females), who completed the self-efficacy questionnaires pre- and post-intervention. Overall, the dietary (p < 0.001) and exercise (p < 0.001) self-efficacy increased significantly post-intervention. Post-intervention, all adolescents decreased their abdominal obesity indices (WC, WHtR, WHR), and this improvement was even more pronounced and significant (p = 0.019, p = 0.019, p = 0.023 respectively) in the adolescents with overweight/obesity. Post-intervention, the proportion of adolescents with normal weight increased from 73.9% to 78.6%, whereas the proportion of adolescents with overweight and obesity decreased from 20.4% to 15.9% and from 5.7% to 5.5%, respectively. Abdominal obesity also decreased from 10.4% to 9.0%. Female adolescents achieved significantly (p = 0.010) higher changes in diet self-efficacy than males. Other sociodemographic characteristics such as family structure, parental age, parental educational level and family income showed non-significant differences. Adolescents with higher KIDMED scores manifested significantly higher dietary and exercise self-efficacy than those with lower KIDMED scores. Both adolescents with normal weight and overweight/obesity manifested a reciprocal relation between diet and exercise self-efficacy. Multicomponent lifestyle interventions in the school environment may provide a first step in students’ behavior changes and provide grounds for future prevention programs in youth.