You divide P(x) = x^3 - 6x^2 + 11x - 6 by (x - 3) using long division. After three rounds of subtraction the remainder comes out 0. Clean. No leftover number.
Now the question — what does that tell you? About P, and about the number 3?
It tells you two things at once, and they are really the same thing. The number 3 is a root of P: plugging x = 3 into P gives zero. And (x - 3) is a factor of P: it divides P exactly, leaving nothing behind. This is the Factor Theorem, and it is the single most-used fact in polynomial root-finding.
The Remainder Theorem, stated
Before the factor version, one step back. When you divide any polynomial P(x) by the linear divisor (x - a), the remainder is not some leftover polynomial — it is a single number, and that number is exactly P(a).
When P(x) is divided by (x - a), the remainder is P(a).
Not "a constant we can compute by continuing the division." The remainder equals the value of P evaluated at x = a. You can find it without doing any long division at all — just plug in.
Derivation
The polynomial division algorithm gives you the identity
where Q(x) is the quotient and R is the remainder. Because the divisor (x - a) has degree 1, the remainder must have degree strictly less than 1 — so R is a constant, a single number, not a polynomial in x.
Now substitute x = a into that identity:
The first term vanishes because (a - a) = 0. Whatever Q(a) is, it gets multiplied by zero and drops out. What is left is R.
So R = P(a). The remainder equals the evaluation. That is the whole proof.
What "remainder = 0" means
Now tighten the screws. Suppose the division leaves remainder zero — R = 0. Four things follow, and they all say the same thing from different angles:
- R = P(a) = 0, because the remainder equals P(a).
- a is a zero (or root) of P — a number at which P evaluates to zero.
- (x - a) divides P(x) exactly — the division leaves nothing behind.
- Equivalently, (x - a) is a factor of P(x).
These are four vocabularies for one fact. "Root," "zero," "(x - a) divides exactly," and "(x - a) is a factor" are synonyms in this context. If a problem hands you any one of them, you get the other three for free.
Factor Theorem, stated
Wrapping those two implications together gives the Factor Theorem. This is the if-and-only-if form:
(x - a) is a factor of P(x) if and only if P(a) = 0.
The "if and only if" is doing real work — both directions are true.
- If P(a) = 0, then by the Remainder Theorem, dividing P(x) by (x - a) gives remainder 0. So (x - a) divides P exactly, which is what "factor" means.
- If (x - a) is a factor of P(x), then you can write P(x) = (x - a) \cdot Q(x) for some quotient Q(x). Substitute x = a: P(a) = 0 \cdot Q(a) = 0.
Every root corresponds to a linear factor, and every linear factor corresponds to a root. The sign flips — root a gives factor (x - a), not (x + a) — but the correspondence is exact.
Worked example: is 1 a root of x^3 - 6x^2 + 11x - 6?
Take P(x) = x^3 - 6x^2 + 11x - 6. Is 1 a root?
Do not divide. Just substitute.
Yes. So by the Factor Theorem, (x - 1) is a factor of P. Divide to find the other factor:
The quadratic x^2 - 5x + 6 factors as (x - 2)(x - 3). Putting it all together:
All three roots exposed: 1, 2, 3. Each root contributes one linear factor, exactly as the factor theorem in reverse would build it.
Using the Factor Theorem to find roots
Here is the general recipe for factoring a polynomial you cannot crack by inspection:
- Try small integer candidates. Start with \pm 1, \pm 2, \pm 3, and the small factors of the constant term.
- For each candidate a, compute P(a). This is just arithmetic — no division needed.
- If P(a) = 0, then (x - a) is a factor. Divide P by (x - a) to get the quotient Q(x).
- Repeat on Q(x). Look for roots of Q, peel off another linear factor, continue until the remaining polynomial is a quadratic (which you can solve directly) or is irreducible.
Each root found reduces the degree by one. A cubic becomes a quadratic after one root; a quartic becomes a cubic. The algorithm walks down the degree ladder one rung at a time.
Rational Root Theorem
"Try small integers" is fine for textbook cubics, but which values should you try in general? The Rational Root Theorem formalises the answer.
For a polynomial a_n x^n + a_{n-1} x^{n-1} + \cdots + a_0 with integer coefficients, any rational root \tfrac{p}{q} in lowest terms must satisfy:
- p divides a_0 (the constant term), and
- q divides a_n (the leading coefficient).
So the candidates are (factors of a_0) over (factors of a_n). For P(x) = x^3 - 6x^2 + 11x - 6, a_0 = -6 and a_n = 1, so the rational-root candidates are \pm 1, \pm 2, \pm 3, \pm 6 — the factors of 6. The roots 1, 2, 3 are all on that list.
This narrows the search from "infinitely many real numbers" to "a handful of rationals." If none of the candidates is a root, the polynomial has no rational roots at all — only irrational or complex ones.
Worked example: P(x) = 2x^3 - 3x^2 - 11x + 6
Here a_0 = 6 and a_n = 2. Factors of 6: \pm 1, \pm 2, \pm 3, \pm 6. Factors of 2: \pm 1, \pm 2. Rational-root candidates:
Work through the list. Try x = 3:
So (x - 3) is a factor. Divide:
The quadratic 2x^2 + 3x - 2 factors as (2x - 1)(x + 2), giving roots \tfrac{1}{2} and -2. All three roots:
And \tfrac{1}{2} was on the candidate list — the Rational Root Theorem told you to look there.
If you don't find a root easily
Sometimes the candidate list runs out with no hits. That does not mean the polynomial has no roots — it means it has no rational roots. The polynomial may have only irrational roots (like \sqrt{2}), or only complex roots (like i), or a mix.
At that point the Factor Theorem has no more easy moves. You reach for numerical methods — the Newton-Raphson iteration, the bisection method, or approximate graphical solving. For cubics, Cardano's formula gives exact roots in radicals. For higher degrees, exact formulas exist up to degree 4 and no further (that is Abel's theorem — no general formula in radicals for degree \geq 5).
Checking a claim — substitute, don't divide
A teacher asks: "Is 5 a root of x^4 - 6x^3 + 2x^2 - x + 30?" The textbook approach would be to divide by (x - 5) and check whether the remainder is zero. That is a lot of arithmetic.
The Remainder Theorem gives you the shortcut. Just substitute:
Not zero, so 5 is not a root, and (x - 5) is not a factor. Done in three lines. The theorem guarantees that substitution and division-then-remainder give the same answer — so always substitute when you only need the remainder. Only actually divide when you need the quotient as well.
Use in equation-solving
The whole point of factoring a polynomial is to solve the equation P(x) = 0. Once P is in factored form
a product is zero if and only if one of the factors is zero, so the roots are exactly r_1, r_2, \dots, r_n. Reading roots off a factored polynomial is immediate.
The Factor Theorem is the engine that takes you from unfactored to factored form:
- Find one root a (by inspection, rational-root candidates, or a lucky substitution).
- Divide by (x - a) to reduce the degree.
- Repeat on the quotient.
Each rung of the ladder exposes one root. After n rungs, a degree-n polynomial is fully factored and all its roots are in hand.
Recognition drill
For each, check whether the given value is a root by direct substitution. No division.
- P(x) = x^2 - 4, a = 2: P(2) = 4 - 4 = 0. Yes, root. (x - 2) is a factor.
- P(x) = x^3 + 1, a = -1: P(-1) = -1 + 1 = 0. Yes. (x + 1) is a factor.
- P(x) = x^2 + 1, a = 1: P(1) = 1 + 1 = 2 \neq 0. No.
- P(x) = 2x^2 - 3x + 1, a = \tfrac{1}{2}: P(\tfrac{1}{2}) = \tfrac{1}{2} - \tfrac{3}{2} + 1 = 0. Yes, root. (2x - 1) (or equivalently (x - \tfrac{1}{2})) is a factor.
Each check is one substitution and a line of arithmetic. Faster than any division.
The takeaway
"Remainder is zero after dividing by (x - a)" is the Factor Theorem in action. It tells you three things packaged as one: a is a root, (x - a) is a factor, and you can divide out (x - a) to reduce the degree and continue factoring.
This is the fundamental trick of polynomial root-finding — find a root by substitution, peel off the corresponding linear factor by division, and repeat on a smaller polynomial. Every factoring recipe, every rational-root hunt, every equation-solving procedure for polynomials comes back to this one fact. Zero remainder means root. Root means factor. Factor means you can divide and keep going.