Self-efficacy is an important affective factor in English as a Foreign Language learning because students’ belief in their own ability may influence their willingness, persistence, confidence, and performance in speaking tasks. Students who believe that they can complete speaking activities are more likely to participate actively, control anxiety, use vocabulary confidently, and continue communicating despite difficulties. This study investigated self-efficacy as a predictor of junior high school EFL students’ speaking ability at SMP Katolik Trisakti 2. The study employed a quantitative correlational design with simple linear regression. The participants were 35 junior high school students. The independent variable was self-efficacy, while the dependent variable was speaking ability. Self-efficacy was measured using a Likert-scale questionnaire, while speaking ability was measured using an oral performance test. Both variables were scored on a scale of 0 to 100. The data were analyzed using descriptive statistics, category distribution, Pearson product-moment correlation, and simple linear regression. The findings showed that students’ self-efficacy was at a moderate-to-high level, with a mean score of 68.07 and a standard deviation of 8.70. Students’ speaking ability was categorized as good, with a mean score of 70.56 and a standard deviation of 8.75. Pearson correlation analysis showed a very strong positive and statistically significant relationship between self-efficacy and speaking ability, r = .998, p < .001. The regression analysis showed that self-efficacy significantly predicted speaking ability, B = 1.00, t = 83.71, p < .001. The coefficient of determination showed that self-efficacy explained 99.5% of the variance in speaking ability.