Studies have shown that while we almost always associate terms such as never or always with probabilities of 0 and 100% respectively, we vary considerably in our interpretations of terms such as unlikely or likely. As Hakel said, one person’s rarely might be another’s almost never! This can make it challenging for scientists to effectively communicate nuanced concepts such as relative risk, as it can be difficult to identify the words or phrases that most accurately convey the statistical probabilities associated with specific outcomes. This problem is probably (!) more pronounced when writing about effect sizes, which are notoriously hard to interpret. Using metaanalysis of 20+ years of published accounts of parasite responses to pollution, I explore the relationships between probability, effect sizes, and the words authors choose to report them. Specifically, I analyze the extent to which p values and standardized effect sizes (Cohen’s d) correlate with (for example) probability phrases (e.g., always, rarely) and verbal markers of author confidence (i.e., positive or negative emotion; certainty vs. tentativeness) generated by text analysis (LIWC2015). I also track how author language has changed over time. I conclude with a discussion of what I think this all means for the use of parasites as biological indicators of pollution.