Fundamental to all evidence rules is the division of responsibility between the judge, who determines the admissibility of evidence, and the jury, which gauges its weight. In most evidentiary contexts, such as those involving hearsay and character, threshold admissibility obligations are clear and relatively uncontroversial. The same is not true for scientific evidence. The complex nature of scientific inference, and in particular the challenges of reasoning from group data to individual cases, has bedeviled courts. As a result, courts vary considerably on how they define the judge’s gatekeeping task under Federal Rule of Evidence 702 and its state equivalents. This Article seeks to reconceptualize gatekeeping analysis in scientific evidence cases based on the nature of science itself, specifically, the division between general and case-specific scientific findings. Because expert testimony describing basic science, “framework” science, and the scientific methods an expert uses to reach conclusions transcend the case at hand, the validity of these facts ought to be determined by the judge. In contrast, when an expert claims to have used a methodology approved by the judge, but a dispute arises as to whether the expert in fact did so, the question becomes one of credibility specific to the case, and is for the jury. This division between general and case-specific preliminary facts is simpler to administer than other admissibility–weight frameworks, which have relied primarily on problematic attempts to distinguish scientific methods from scientific conclusions. It is also fully consistent with, and helps implement, basic principles of both constitutional and evidentiary jurisprudence by ensuring that the trial judge—presumptively better attuned to matters of general import—decides reliability issues, while the jury—historically viewed as trier of the facts—is the ultimate arbiter of those case-specific matters requiring a credibility assessment. Because the general–specific divide likewise argues for a stiff standard of appellate review on scientific reliability issues, our alignment of evidence law with the nature of scientific research also provides the best court-monitored mechanism for ensuring that courtroom use of science is both sophisticated and consistent across cases.