This book is dynamite: George E. P. Box, Statistics for Experimenters: An Introduction to Design, Data Analysis, and Model Building
It starts from zero knowledge of Statistics but it doesn't insult the reader's intelligence. It's incredibly practical but with no loss of rigour; in fact, it underscores the danger of ignoring underlying assumptions (which are often false in real life) of common tests.
It's out of print but it's very easy to find a copy. Follow the link for a few options.