Descartes' Rule of Signs Calculator
Somewhere in every algebra course there is a moment when a quintic appears on the board, somebody asks "so what are its roots?", and the honest answer is that nobody in the room can factor it. That is the moment Descartes' Rule of Signs was made for. It will not hand you the roots — but by doing nothing more than counting how often the signs flip along the coefficients, it tells you how many positive and negative real roots are even possible, before you lift a pencil to factor anything. The Descartes Rule of Signs calculator at the top of this page runs that count for any polynomial you give it: type the expression the way it is written in your homework — factored form included, brackets and all — press =, and you get the possible positive, negative and complex root counts, the complete combination chart, and every sign change marked on the actual coefficients.
What follows is the whole topic, taught properly: what the rule says, how to count sign changes without the classic mistakes, why the answer can drop "by an even number", how the f(−x) substitution catches negative roots, and what to do when x = 0 sneaks in as a root because the constant term is missing. Everything is free, and nothing stands between you and the tool.
What Is Descartes' Rule of Signs?
René Descartes published the rule in 1637, tucked into La Géométrie — the same appendix that gave the world coordinate geometry. It has survived four centuries in classrooms for one reason: it converts a hard question about roots into an easy question about counting.
Take a polynomial with real coefficients, written in standard form with the powers in descending order. The rule makes two statements:
Positive real roots. Count how many times the sign changes as you read the nonzero coefficients left to right. The number of positive real roots is either that count, or less than it by an even number.
Negative real roots. Replace x with −x, simplify, and count the sign changes of the new polynomial the same way. That count — or that count minus an even number — is the number of negative real roots.
Notice how modest the claim is, and how useful anyway. Before factoring, before graphing, before the rational root theorem, you already know the shape of the search: whether positive roots are worth hunting at all, whether negative ones can exist, and how many roots must be complex whatever else happens. Textbooks reach for it precisely because it costs thirty seconds and rules out half the wrong turns in advance.
How to Use This Descartes Rule of Signs Calculator
Most tools for this rule ask you to fill a row of little boxes — a₀ here, a₁ there — which means translating your polynomial into a coefficient list before the tool will even talk to you, and one forgotten zero ruins the count. This one skips that ritual. There is a single box, and you type the polynomial the way it already looks on the page:
Type naturally. x^4 - x^3 - x^2 + x works. So does 2x³, so does 3⁄4x^4, so do decimals, so do terms in any order, and a trailing "= 0" is quietly ignored. Powers can be written with the caret, with ** or pasted straight from a document as x². If you paste something ambiguous like "x2", the calculator refuses and asks whether you meant x^2 or 2x, rather than guessing behind your back.
Factored form goes straight in. This is the feature to remember: (x+1)(x-2) and (x−1)²(x+2) and 2(x^3−1) are all accepted and multiplied out for you, and the expanded polynomial is shown back in the answer band so you can confirm the reading. If your homework hands you the factored form, you never expand it by hand just to feed a calculator.
Use the keypad, or your own keyboard. Phone keyboards hide the caret and brackets three menus deep, so the calculator carries its own keypad with x, xⁿ, brackets, the fraction bar and digits. Prefer typing? The small keyboard icon inside the input box switches your system keyboard on and off. A line under the box echoes whatever you type in textbook notation — x^3 appears as x³ — so errors are visible before you press anything.
Press = and read the three tiles. Positive, Negative, Complex — each showing its maximum with a ≤ sign, because the rule gives possibilities, not certainties. Beneath them: the polynomial as understood, a plain sentence, and the "All possible combinations" button that expands the full chart. The step-by-step panel underneath rebuilds the whole argument — standard form, any zero root factored out, then the sign-change count on f(x) and on f(−x), with every flip marked with an arrow directly on your coefficients.
How to Count Sign Changes
Every mistake with this rule is a counting mistake, and there are only two of them: counting a pair that isn't adjacent, and letting a zero coefficient join the parade. The procedure that avoids both:
Write the terms in descending order of power. Read only the nonzero coefficients, left to right, and each time the sign differs from the previous nonzero sign, count one change. Zeros are invisible — skip them completely; they neither cause a change nor block one.
Run it on 2x⁵ − 3x⁴ + x² − 1, one of the examples under the calculator:
| Step along the coefficients | Signs | Change? |
|---|---|---|
| +2 → −3 | + to − | yes — 1st |
| −3 → (x³ missing, skip) → +1 | − to + | yes — 2nd |
| +1 → −1 | + to − | yes — 3rd |
Three sign changes, so the polynomial has 3 or 1 positive real roots. The x³ and x terms are absent and were skipped as though they were never there — had you treated those zeros as sign-bearers, the count would be wrong and everything downstream with it. The calculator's step panel draws exactly this walk, dimming the zero coefficients and putting a blue arrow at each genuine flip, which is the fastest way I know to teach the skip-the-zeros habit.
Finding Negative Roots with f(−x)
The rule as stated only sees positive roots. The trick for negative ones is a change of viewpoint: every negative root of f(x) is a positive root of f(−x). So build f(−x) and count again.
Building it is mechanical once you see the pattern: substituting −x flips the sign of every odd-power term and leaves even powers alone, because (−x)² = x² but (−x)³ = −x³. For the quintic above:
f(−x) = −2x⁵ − 3x⁴ + x² − 1
Reading the signs: − − + −. The first pair agrees, then − to + is one change, + to − is a second. Two sign changes, so 2 or 0 negative real roots. The odd-powers-flip shortcut is worth memorising for exams — it is faster than substituting term by term and much harder to fumble — and the calculator prints the flipped polynomial in full so you can check your own version against it.
Why "Or Less by an Even Number"?
The strangest phrase in the rule — why would roots vanish two at a time? — has a one-sentence answer: complex roots of a real polynomial only ever come in conjugate pairs. If 2 + 3i is a root, 2 − 3i must be one too. So whenever a real root fails to materialise, it hasn't wandered off alone; it has paired up with a partner and moved into the complex plane, taking exactly two slots from the real count.
That is why three sign changes means 3 or 1 positive roots, never 2: you can only step down from the maximum in twos. It is also why the parity of your answer is itself information. One sign change means exactly one positive root, guaranteed — one cannot drop to minus one — which is the rule at its most decisive and the case worth watching for in any exam question.
The Descartes Rule of Signs Chart: All Possible Combinations
Once both counts are in hand, the honest summary of what you know is a chart. Each row is one scenario the rule permits; the columns must always add up to the degree, because a degree-n polynomial has exactly n roots when complex ones are counted — that is the fundamental theorem of algebra keeping score. For 2x⁵ − 3x⁴ + x² − 1, with 3 sign changes for positive and 2 for negative:
| Positive | Negative | Complex | Total |
|---|---|---|---|
| 3 | 2 | 0 | 5 |
| 3 | 0 | 2 | 5 |
| 1 | 2 | 2 | 5 |
| 1 | 0 | 4 | 5 |
Four scenarios and no more: positive stepping through {3, 1}, negative through {2, 0}, complex filling whatever remains — and only rows where the complex count comes out even survive, since pairs cannot be split. Exam questions that say "list all possible combinations of roots" are asking for precisely this table, and it is the table this calculator builds for every polynomial under the "All possible combinations" button, with the complex column computed and checked against the degree for you.
Worked Examples
A cubic with no positive roots. f(x) = x³ + 7x² + 4. Every coefficient is positive, so reading + + + gives zero sign changes: no positive real roots at all, guaranteed. For the negative side, f(−x) = −x³ + 7x² + 4 reads − + +, one change — exactly one negative real root. One real root, and the remaining two must be a complex pair: the chart has a single row, 0–1–2. Twenty seconds of counting has pinned down the entire structure of a cubic nobody asked you to solve.
A factored quartic, entered as-is. Type (x−1)²(x+2) into the calculator — brackets, square and all. It expands the product to x³ − 3x + 2, then counts: + − + gives 2 changes, so up to 2 positive roots; f(−x) = −x³ + 3x + 2 gives 1 change, so exactly 1 negative root. And because we happen to know the factorisation, we can watch the rule being right: the roots are 1 (twice) and −2 — two positive, one negative, the maximum case realised. A repeated root counts once per multiplicity, which this example quietly demonstrates.
Nothing real at all. f(x) = x² + 1. No sign changes in + +, and f(−x) = x² + 1 again — none there either. Zero positive, zero negative, so both roots are complex (they are ±i). When both counts come out zero for a polynomial with a nonzero constant term, the rule has just told you the graph never touches the x-axis, without a single point being plotted.
The same three Descartes rule of signs examples, side by side — each one is also a tap-to-solve chip under the calculator:
| Polynomial | Positive | Negative | The lesson |
|---|---|---|---|
| x³ + 7x² + 4 | 0 | exactly 1 | all-positive coefficients rule out positive roots |
| (x−1)²(x+2) | ≤ 2 | exactly 1 | factored input; repeated roots count by multiplicity |
| x² + 1 | 0 | 0 | double zero means every root is complex |
When x = 0 Is a Root: The Missing Constant Term
Here is the case most calculators — and plenty of students — handle wrongly. Descartes' rule counts strictly positive and strictly negative roots; zero is neither, so a root at x = 0 is invisible to it. And a polynomial has a root at zero exactly when its constant term is missing.
The correct procedure is to factor the zero roots out first. Take the calculator's opening example, x⁴ − x³ − x² + x. No constant term, so factor: x(x³ − x² − x + 1). That leading x contributes the root x = 0 with multiplicity 1; the rule then applies to the cubic in the brackets, which gives 2 sign changes for positive and 1 for negative. Final account: up to 2 positive, exactly 1 negative, one zero root — and the combination chart gains a Zero column so the totals still reach the degree of 4.
This calculator performs the whole manoeuvre automatically: it detects the missing constant, factors out x to the correct multiplicity, announces the zero root in plain words, applies the rule to the reduced polynomial, and shows the factored form as its own step. If a tool ever reports sign-change counts for a constant-free polynomial without mentioning x = 0, its totals cannot reach the degree — a five-second sanity check worth running on any calculator, this one included.
What the Rule Cannot Tell You
A professor's duty is to state the limits as clearly as the powers. Descartes' rule of signs bounds the count of roots; it does not locate a single one of them. It will not tell you that the positive root of x³ − 3x + 2 sits at exactly 1, only that at most two positive roots exist. When the chart shows several surviving rows, the rule alone cannot pick between them — that takes more information: the rational root theorem to propose candidates, synthetic division to test them, or a graph to count actual crossings.
Treat it, then, as the reconnaissance step it was always meant to be. It clears the ground — no negative roots here, exactly one positive there, at least two complex everywhere — so the heavier tools are aimed only where roots can actually live. Used that way, thirty seconds of sign-counting routinely saves half a page of misdirected factoring.
What This Descartes Rule of Signs Calculator Does Differently
One box, your notation. No coefficient grid to fill, no a₀-a₁-a₂ translation, no forgotten-zero trap. The polynomial goes in as written — any term order, fractions, decimals, pasted superscripts — and the answer band restates f(x) so you can verify the reading at a glance.
Factored input, expanded honestly. (x−1)²(x+2) is multiplied out inside the calculator using error-tracked arithmetic, so cancellations collapse to genuine zeros instead of leaving microscopic false coefficients that would corrupt the sign count. That level of numerical care is invisible when it works — which is the point.
The zero root handled, not ignored. Missing constant term? The x is factored out, the multiplicity stated, the rule applied to the reduced polynomial, and the chart gets its Zero column. The totals always reach the degree.
The count you can see. Rather than announcing "3 sign changes" as a verdict, the step panel lays your actual coefficients in a row, dims the zeros, and marks each flip with an arrow — for f(x) and for f(−x) separately. You are not trusting the tool; you are watching the count.
A keypad built for polynomials. x, powers, brackets and the fraction bar on-screen where phone keyboards hide them, with a one-tap switch back to your own keyboard, and a live textbook-notation echo under the box so typos surface before you press =.
Refusals over guesses. "x2" could mean two things, so the calculator asks instead of assuming. Dividing by an expression containing x is not a polynomial, so it says so. Honest errors beat confident wrong answers — on that principle the whole Monkza toolkit is built.
Frequently Asked Questions
What is Descartes' rule of signs?
How do you use Descartes' rule of signs step by step?
What counts as a sign change?
Do zero coefficients count as sign changes?
Why can the number of roots be less by an even number?
Does Descartes' rule of signs give the exact roots?
What happens when the polynomial has no constant term?
Can Descartes' rule of signs find imaginary roots?
Is zero a positive or a negative root?
What is a Descartes rule of signs chart?
Can I type a factored polynomial into this calculator?
Does the rule work for polynomials of any degree?
Who invented the rule of signs?
Is this Descartes rule of signs calculator free?
Four centuries on, Descartes' little counting rule is still the fastest first question to ask of any polynomial: where can the roots even be? Count the flips in f(x), flip the odd powers and count again for f(−x), step down in twos, and hand zero roots their own line before you start. The calculator at the top of this page will run the whole argument for anything you type — factored or expanded — and show its working like a patient tutor. When the hunt moves from counting roots to computing with them, the scientific calculator is next door, and the rest of the free toolkit lives on the Monkza calculators page.