Researchers use statistically significant results involving astrology as examples of Type I errors where the techniques make chance results appear significant. In this paper, I examine two such studies and I replicate a result involving Pisces and heart disease using a different set of data, a result first found significant in the previous study by a chi square test but then deemed insignificant using Monte Carlo methods, and in the new data set the result is found significant using Fisher’s exact test (p = .0028). However, a closer examination of this data reveals that the significant result that suggests an association between Pisces and heart problems is better explained within astrology not by having Sun in Pisces but by the presence of medical astrology factors that might temporarily make Pisces people more susceptible to heart difficulties.