Warning Signs in Experimental Design and Interpretation

I think the broad appeal of statistics is because it seems to offer a way of finding truth automaticly, kind of like the allure of neural networks.