During the covid-19 pandemic, the learning outcomes of students in class X IPS SMA Negeri 1 Watubangga were still not optimal due to the learning media used, namely LKPD media, so that treatment was needed, namely by applying video-based learning media that could provide opportunities for students to better understand the material presented. This study aims to determine the application of video-based learning media is more effective than LKPD media on the subject of the hydrological cycle of class X social studies students at SMA Negeri 1 Watubangga. This research is an experimental research using nonequevalent control group design. The population of this study were X grade students of SMA Negeri 1 Watubangga with samples, namely X IPS 1 class as the experimental class and X IPS 3 class as the control class, the sample was taken by purposive sampling technique. Data collection techniques used tests before and after treatment. Data analysis used normality test, homogeneity and normalized gain test. The results showed that video-based learning media was more effective than LKPD media. The results of the mann whitney hypothesis test Z value of -1.928 with a significant 0.05 = 0.05 then the hypothesis H0 is rejected and H1 is accepted with the conclusion that there is an average difference between the experimental class and the control class. The results of the gain test in the control class amounted to 0.42 medium category with a percentage of 41.96 interpreted as less effective while in the experimental class amounted to 0.58 medium category with a percentage of 58.10 interpreted as quite effective. The conclusion of this study is that the increase in learning outcomes in the experimental class using video-based learning media is more influential and effective compared to the control class using LKPD media in the learning process.