The study of video game effects is currently facing four dilemmas. The first is a dilemma of causality: research should strive to contribute to our understanding of generalised causal mechanisms rather than isolated effects, but our research methodologies are often misaligned with this goal. The second is a dilemma of measurement: inappropriate use of self-report is common, and phenomena which self-report cannot measure are suppressed. The third is a dilemma of interest: personal and professional interests are frequently at odds with normative goals of scientific enquiry. Finally, we face a dilemma of speed: horizon-scanning is limited, and available methodologies are slow to address societal challenges. One solution to problems is the use of digital trace data. We describe five distinct methods for the collection of such data, ranging from the development of bespoke data infrastructure to the brokerage of data sharing agreements with industry bodies. Using our dilemmas as a lens, and drawing from our own practice as data handlers, we highlight how these methods may exacerbate or resolve each dilemma: For example, while data sharing agreements with industry unlock rapid analysis paradigms, they also present novel sites for nonfinancial conflicts of interests. Finally, we propose next steps to make digital trace data collection strategies more widespread and accessible.