In the latest issue of Regulation, Paul H. Rubin explains the differences between "Reagan and the Wonks." The latter mine the data but they are up against the limited data processing capabilities of their audience. The everyman audience is equipped with data processing talents honed in our evolutionary past. We make fairly good snap judgments but require the science of statistical inference when up against large data sets.
Court room success or political success are not the same as economteric success. Rubin writes about the natural problems that masters of the latter have when attempting to (or being asked to) make their findings widely known and accepted.
What is a wonk to do? Rubin concludes: "An important remedy would be to devise ways of expressing policies in terms that will resonate with with the decisionmaking process of voters. Ronald Reagan was a master of advocating free-market policies using rhetoric that reflected the biases of voters, and others should be able to learn from his methods."