Split

Split taught the first Advanced idea by breaking the rule on purpose, to show the wreck.

"Watch what happens if I get greedy," Split said, and made a second pair — but this time changed two things at once, a new title and a new picture. One card won. "Now tell me why it won." The young creators opened their mouths and stopped. "Exactly. You can't. Was it the title? The picture? Both? Neither, and just luck? You changed two things, so the credit is tangled between them forever. That tangle has a name — a confound — and it ruins the test. Two changes at once don't give you two answers. They give you zero."

The young creators felt the trap close. It was so tempting to improve two things at once — and that temptation was exactly what made the result meaningless.

Split named the two sides of a fair test.

"The card you leave alone is the control," Split said. "The card with the one deliberate change is the treatment. The control is your ruler — without something held steady to compare against, a change tells you nothing, because you don't know what would have happened anyway. Maybe both versions would have done better today for some outside reason. The control catches that. Always keep one thing steady to measure the other against."

Then Split taught the deepest idea, the one that outlives any single test.

"Even in the wild, without an experiment, remember this," Split said. "When two things happen together, that is correlation — they moved alongside each other. It does not prove one caused the other. Ice cream sales and sunburns rise together, but ice cream doesn't cause sunburns — the sun causes both. A fair test is precious because it's the one way to turn 'these happened together' into 'this actually caused that.' Outside a fair test, hold your conclusions loosely: alongside is not the same as because."