Lets keep the con out of experimental a methodological note

When Edward Learner (1983) wrote the well known critique of econometric practice whose title I have adapted and adopted, he was concerned that the credibility and utility of econometric research had suffered because of differences between the way econometric research was conducted and the way it was reported1. He wrote (p36–37): “The econometric art as it is practiced at the computer terminal involves fitting many, perhaps thousands, of statistical models. One or several that the researcher finds pleasing are selected for reporting purposes. This searching for a model is often well intentioned, but there can be no doubt that such a specification search invalidates the traditional theories of inference. The concepts of unbiasedness, consistency, efficiency, maximum-likelihood estimation, in fact, all the concepts of traditional theory, utterly lose their meaning by the time an applied researcher pulls from the bramble of computer output the one thorn of a model he likes best, the one he chooses to portray as a rose. The consuming public is hardly fooled by this chicanery.”