Educational games are nowadays gaining popularity among many science practitioners for its potential to improve students' performance in science by enhancing their conceptual understanding, motivation and peer collaboration. However, its efficacy in improving achievement has not yet been explored in the Pakistani context. Therefore, this study aimed at exploring the efficacy of educational games in improving grade VI students' achievement in selected science concepts on Science Achievement Test (SAT). This study employed quasi-experimental pre-and posttest control group design. For the sample, two intact classes of grade VI were recruited from two private community based schools and then assigned as comparison group (CO) and intervention group (10). In total, 60 students participated constituting of equal number of students from both groups. For data collection, SAT which was developed and validated as part of this study was administered to both groups before and after intervention. The intervention comprised of four weeks during which 10 was pre-dominantly taught with educational games whereas, CO was taught with the pre-existing teaching method. The data obtained were analyzed for students' overall performance, performance on SAT content and cognitive domains on both pretest and posttest. Findings of the study illustrated that before intervention, both the groups evenly matched on their overall performance on SAT as well as their performances in the content and cognitive domains with no significant difference (p> 0.05 for all). Conversely, after intervention, both the groups improved; however, the overall performance on SAT favored 10 (p< 0.01) with the large magnitude of difference (r=0.70; r2= 0.49). Likewise, on SAT content domains, 10 outperformed CO with significant difference (p