Certainty versus probability in reasoning
Deductive reasoning moves from general principles to specific conclusions with absolute certainty, assuming the premises are true. All humans die, you are human, therefore you will die. The conclusion cannot be false if the premises are true. Deductive logic is binary: either the argument is valid (conclusion follows necessarily) or it's invalid (conclusion might be false).
Inductive reasoning moves from specific observations to general conclusions with varying probability. You've observed 10,000 white swans and never a black one, so you infer all swans are white. But one black swan disproves this. Inductive strength lives on a spectrum: stronger evidence increases confidence, but no amount of observation guarantees the conclusion. This distinction matters enormously in science, which relies on induction to build theories from observations.
When each reasoning type applies
Deduction excels at extracting consequences from known facts. If you know the rules of chess, deduction tells you which moves are legal. It cannot tell you whether those moves are strategically sound, only whether they follow the rules. Induction addresses the empirical world: why do objects fall? What's the relationship between smoking and disease? Induction gathers data and finds patterns.
Pure deduction cannot create new knowledge about reality, only reorganize existing knowledge. Pure induction without deduction is infinite: you could test a billion cases and still miss one counterexample. Effective thinking uses both: deduction to test implications of theories, induction to build and refine theories from evidence.