Factorial

No design is best on every row, and which row matters depends on what the experiment is for: A/p or E for precise coefficients, I or G for prediction over the region, max |r| for keeping second-order effects apart, and power for detecting an effect of a stated size. Each measure's name links to its definition, and the book's comparison of design families reads such a table row by row.

Each row is shaded to guide the eye: the greener a value, the nearer it is to that row's best, which is in bold. Only the designs shown that fit the model are shaded. Each row prints as many digits as tell its designs apart; the Python code below prints four significant figures.

To narrow the choice for reasons the table does not hold (the cost of a run, a setting the plant cannot hold, a design the team already knows), turn a design's switch off. It moves past the thick line at the right, in grey, and leaves the shading and the charts. Turn it on and it comes back to its place.