This paper discusses the challenge of studying play and gaming activities in situations where participants are reluctant to disclose their participation, or in which researchers face limitations that prevent direct observations. To answer this question, we have focused on the material traces and clues of play left by the participants, terming this approach 'ludological semiotics': the study and interpretation of traces, clues and remnants of game and play activities. We base our methodology on historical and archaeological studies of traces, notably drawing from Carlo Ginzburg's method of clues. We will scrutinize the traces and clues of play we have identified and documented, presenting their categorization and our interpretation model. We will contemplate what assumptions about play can be made based on these clues. Finally, our approach addresses some methodological challenges in contemporary game and play studies.