Academic performance optimization has increasingly become a central priority in contemporary educational systems; however, the psychological costs of this optimization drive remain underexplored in the context of data-driven, school-based predictive frameworks. This study investigates the quantitative relationships between five independent stressor variables — daily study hours, sleep duration, recreational screen time, perceived parental pressure, and exam frequency — and a composite measure of emotional strain among secondary school students (N = 412, aged 14–18). A structured survey instrument was administered across four urban and peri-urban school districts, yielding data subsequently analyzed through multiple linear regression, Pearson correlation analysis, and a purpose-formulated Strain Index Model (SIM). Findings indicate that parental pressure (β = 0.47, p < .001) and inadequate sleep duration (β = −0.38, p < .001) are the strongest predictors of elevated emotional strain, while high screen time moderates stress responses in a non-linear fashion. Students in lower-income cohorts demonstrated disproportionately higher composite strain scores, raising equity concerns regarding the use of performance-based predictive algorithms in school settings. The study argues for the integration of socio-emotional equity metrics into predictive systems and recommends targeted interventions targeting parental expectation management and sleep hygiene. The implications extend to educational policymakers, school counselors, and EdTech developers designing next-generation student support platforms.