DAVID W. BARNES [*] I INTRODUCTION Judges and lawyers first encountering statistical want to believe that scientific standards are tougher than legal standards. A court will reject an assumption that there is no causal connection between an act and an injury if the makes causation likely than A scientist will reject an assumption that there is no relationship between two variables only if there is less than a five percent probability that the statistical showing a relationship is due to chance. The law appears willing to accept no more than a forty-nine percent chance of error while science appears willing to accept no more than a five percent chance of error. This perception is incorrect, but hard to change. It is a matter of such serious concern to statisticians and scientists that they often raise the issue, but lay people seldom understand it. This article offers those uninitiated into the statistical guild several reasons to look behind the probabilities when evaluating scientific evidence. The lack of congruity between legal and scientific standards is most egregious when testimony based on statistical/scientific methods is used to prove causation in tort law. Medical scientific testimony is often expressed in terms of two different probabilities. The first of these estimates the increased probability of harm if a person is exposed, for example, to a toxin. The second is the probability that the observed relationship is an artifact of the experimental method, rather than an actual causal relationship between the toxin and the injury. These two probabilities measure different phenomena, and neither measures whether causation is more likely than not. This article demonstrates that neither taken alone or together, measures whether the of the evidence test is met. In many cases involving statistical/scientific evidence, probabilistic observations may be conveniently captured under three headings: the probability, the probability,' and the error probability. The belief probability relates to evidentiary requirements imposed by the law, and the fact probability relates to the facts relevant to legal cases. These two probabilities are directly related to the civil law evidentiary requirement that the proponent of a claim must prove that the other's act is more likely than not a cause of harm. By contrast, the sampling error probability is a characteristic of statistical science. Appreciating the distinctions among these probabilities facilitates an understanding of the relationship between the preponderance of the standard and the probabilities reported by statisticians. II THE THREE PROBABILITIES The belief probability refers to the credibility--the believability--of the in support of a party's factual claims. In tort causation, the belief probability describes the factfinder's confidence in a party's about cause. In civil cases, the law requires that the proponent of a fact convince the factfinder by a preponderance of the evidence. The factfinder is instructed that the plaintiff's claim of causation must be more likely than not true. Thus, the belief probability for the proponent's factual assertion must exceed fifty percent in civil cases for the proponent's assertion to be accepted by the factfinder as true. [2] The fact probability describes a separate feature of a party's related to cause. The fact probability is the likelihood that the defendant's actions led to the adverse outcome. A fact probability may be based on, or stated in terms of, percentages. For example, when a physician delays his diagnosis of a patient's disease, the fact probability measures the percentage point reduction, due to the defendant physician's delay, in the plaintiff's chances of survival or recovery (or, in general, of obtaining a better outcome than she obtained). …