How To Use The Pythagorean Theorem

The theorem states a relationship about areas as much as about lengths. Squares drawn on the two shorter faces of a right triangle have a combined area exactly equal to the square on the long face. Reading it that way explains why the power is two and no other number would do: areas scale with the square of side length, so this is a statement about area being preserved under rearrangement.

The right angle is the whole precondition, and it is worth being explicit because it is the most common misuse. If the two known sides do not meet at ninety degrees, the formula gives a wrong answer rather than a slightly approximate one. In that situation the correct generalisation is the law of cosines, which reduces to Pythagoras when the included angle is exactly ninety degrees and the cosine term vanishes.

Identifying which side is the hypotenuse is the first practical step, and it never requires a diagram if you remember one thing: the hypotenuse is opposite the right angle and is necessarily the longest side. That gives a free error check. If your computed c comes out shorter than one of the inputs, you subtracted instead of added, and no amount of rechecking the arithmetic will fix it because the setup is wrong.

Rearranging for a missing leg is where most arithmetic slips happen, because the operation flips from addition to subtraction. Solving for a leg means a = sqrt(c^2 - b^2), and the most common error is adding anyway, which produces a side longer than the hypotenuse — physically impossible and therefore detectable. Always compare your result against the known longest side before writing it down.

Integer triples have been catalogued since Babylonian times and remain genuinely useful. The familiar ones — 3, 4, 5; 5, 12, 13; 8, 15, 17; 7, 24, 25 — satisfy the relation exactly with no rounding, which makes them ideal for checking a calculation and for laying out right angles physically. Multiplying a triple by any constant produces another triple, so 6, 8, 10 works identically.

The converse is equally true and arguably more useful in measurement: if a^2 + b^2 = c^2 holds for a triangle's three sides, then the angle between a and b is a right angle. Builders use this to square corners with tape measures, and it is why a 3-4-5 layout guarantees ninety degrees without any protractor. When measurements disagree slightly, the mismatch quantifies how far out of square the corner actually is.

Diagonals are the most frequent everyday application. The diagonal of a rectangle is the hypotenuse of the triangle formed by its two sides, so a 1920 by 1080 screen has a diagonal of sqrt(1920^2 + 1080^2) = 2202.9072 pixels. The same reasoning gives distances across fields, the size of television screens quoted diagonally, and the length of a cable routed corner to corner rather than around the edges.

The idea extends naturally into three dimensions with almost no extra machinery. The space diagonal of a rectangular box with sides p, q and r has length sqrt(p^2 + q^2 + r^2), obtained by applying the theorem twice. This result is what makes it possible to quote a single diagonal measurement for a room or a shipping container, and it generalises further to any number of dimensions where it becomes the Euclidean distance formula.

Finally, units need care even though the arithmetic is trivial. Both known lengths must be in the same unit before squaring, or the sum of squares is meaningless. The result is always in that same unit, never squared, because the final square root undoes the squaring. Mixing centimetres with inches and then reporting square units is a mistake that survives casual review surprisingly often.