In the marketing of video games made by self-publishing, independent developers, there is common advice to market a game using the social media platform Twitter. However, the advice that stems from industry sources is somewhat contradictory, and many self-publishing independent developers elect not to use Twitter at all. This presents an opportunity for researchers to investigate this tension and determine if using Twitter does have a causal influence on the successful release of a game by using relatively recent developments in causal data science techniques. In this sense, this paper highlights these causal inference developments while simultaneously informing self-publishing independent developers whether they should indeed market their games using Twitter. It was found that using Twitter results in an average increase of 85.4 reviews during release week, corresponding to a 314% positive difference. Using Twitter also doubles the chance of reaching a critical 10-review inflection point threshold on release week. These effects, however, are moderated by the characteristics of the game, expressed as 'tags', yet in no case is the effect of Twitter use reduced to 0. Based on these findings, it is advised that new, self-publishing video game developers do use Twitter to market their games.