Studies on digital game-based language learning (DGBLL) have increased in numbers, creating a pool of studies that can be meta-analyzed to measure the overall effect of digital gaming on second language (L2) development. The current meta-analysis targets digital games that were available to the public at the time of data collection, January of 2020, aggregating their effects on L2 development overall and across a number of moderator variables. These moderator variables include the game developers' intended purpose of the game (educational or entertainment), outcome measures (e.g., vocabulary, overall proficiency), and several game design features such as the type of player interaction (single player, multiplayer, massively multiplayer online), among others. Results indicate that DGBLL has had a small to medium positive effect (Cohen's d[subscript weighted] = 0.50) for between-groups designs and a medium effect (d[subscript weighted] = 0.95) for within-group designs. Games designed for entertainment were found to be more effective than those designed for L2 education, although there is some overlap in the 95% confidence intervals of the two groups. The overall findings and those from additional moderator analyses are discussed in light of previous DGBLL findings while offering direction for future research and recommendations for improving the methodological rigor and transparency in DGBLL research.