SAT® Quadratic Discriminant: Count Real Solutions Without Solving Everything

Put the whole equation into standard form

The coefficients a, b, and c belong to an equation with zero on one side. If the equation is x² + 6x = 7, rewrite it as x² + 6x − 7 = 0. Then a = 1, b = 6, and c = −7. Treating c as positive 7 would analyze a different equation and produce a different discriminant.

Combine like terms before reading the coefficients. For 3x² + 2x + 1 = x² − 4x + 6, move everything to the left to obtain 2x² + 6x − 5 = 0. The relevant coefficients are 2, 6, and −5. You cannot simply use the first three numbers visible in the original equation.

Check that a is nonzero. If a parameter choice makes the x² coefficient zero, the equation is no longer quadratic for that choice. A linear equation can have one solution without a zero discriminant. Parameter problems involving the leading coefficient therefore require checking this special case separately.

The discriminant rule applies to a quadratic in standard form. Moving a term across the equals sign changes its sign.

Why the discriminant counts roots

The quadratic formula is x = (−b ± √(b² − 4ac))/(2a). The expression under the square root controls whether the two formula branches produce different real numbers, the same real number, or no real numbers. That expression is the discriminant, often abbreviated D.

If D is positive, its square root is positive. Adding it and subtracting it give two different real values because 2a is nonzero. If D is zero, both branches use a square root of zero, so they give the same value, −b/(2a). This is one distinct real solution, also called a repeated root. If D is negative, its square root is not real, so there are no real solutions.

You usually do not need to calculate the roots when asked only how many real solutions exist. For 2x² + 3x − 4 = 0, D = 3² − 4(2)(−4) = 9 + 32 = 41. Because 41 is positive, there are two distinct real solutions. Their exact values involve √41, but those values are unnecessary for a counting question.

Read the three cases on a graph

Real solutions of ax² + bx + c = 0 are the x-coordinates where the graph y = ax² + bx + c meets the x-axis. A positive discriminant corresponds to two distinct intercepts. A zero discriminant corresponds to the vertex touching the axis at one point. A negative discriminant corresponds to no x-axis intercepts.

For y = (x − 3)², the graph touches the axis at x = 3. Expanding gives x² − 6x + 9, whose discriminant is 36 − 36 = 0. For y = (x − 3)² − 4, there are intercepts at x = 1 and x = 5. For y = (x − 3)² + 4, every output is at least 4, so there are no real zeros.

Opening upward or downward does not by itself determine the number of intercepts. An upward-opening parabola can cross, touch, or miss the x-axis depending on its position. Likewise, knowing that c is positive does not alone establish that no real roots exist. Use the full expression or the vertex's position.

Solve a parameter question

Suppose x² − 10x + k = 0 has exactly one distinct real solution. Here a = 1, b = −10, and c = k. Set the discriminant equal to zero: (−10)² − 4(1)(k) = 0. This gives 100 − 4k = 0, so k = 25. Substitution produces x² − 10x + 25 = (x − 5)², which confirms one distinct root.

If the question instead asks for no real solutions, require 100 − 4k < 0. Then −4k < −100, and dividing by −4 reverses the inequality, giving k > 25. For two distinct real solutions, require 100 − 4k > 0, which gives k < 25. The boundary value belongs to the one-root case.

Pay attention to restrictions such as ‘k is an integer.’ If k must be an integer and the question asks for the least value producing no real solutions, the answer is 26. If k is unrestricted and must merely exceed 25, there is no least real number greater than 25. An answer based on rounding only makes sense when the problem supplies an appropriate discrete restriction.

Watch negative signs and squared expressions

In x² − 8x + 20 = 0, b is −8, so b² is (−8)² = 64. It is not −64. Parentheses are essential when substituting a negative coefficient into a square. The discriminant is 64 − 80 = −16, so there are no real solutions.

The product −4ac can become positive when a and c have opposite signs. For 3x² + 2x − 1 = 0, D = 2² − 4(3)(−1) = 4 + 12 = 16. A careful written substitution prevents the common error of subtracting 12 instead of adding it.

When a coefficient contains a variable expression, square the whole expression. If b = k + 2, then b² = (k + 2)² = k² + 4k + 4. It is not k² + 4. Keeping the factored expression intact until you need to solve can reduce expansion errors, especially if the resulting equation has a recognizable structure.

Connect a line and a parabola

A system involving a line and a parabola can become a discriminant question after substitution. Consider y = x² − 4x + 7 and y = 2x + k. Equating their outputs gives x² − 4x + 7 = 2x + k, or x² − 6x + (7 − k) = 0. Each real solution for x identifies an intersection of the two graphs.

The discriminant is (−6)² − 4(1)(7 − k) = 36 − 28 + 4k = 8 + 4k. One intersection requires 8 + 4k = 0, so k = −2. At that value the quadratic becomes x² − 6x + 9 = 0, giving x = 3. Substitution into the line gives y = 4, so the graphs meet at (3, 4).

For k > −2 the discriminant is positive, so there are two intersections. For k < −2 there are none. In this example the one-intersection case is tangency. The important step is analyzing the equation formed by setting the two functions equal, not analyzing the parabola's own x-intercepts.

Use vertex form when it is simpler

The discriminant is useful, but it is not always the shortest route. In (x + 2)² = 9, taking square roots gives x + 2 = 3 or x + 2 = −3, so x = 1 or x = −5. In (x + 2)² = 0 there is one distinct solution, x = −2. In (x + 2)² = −9 there are no real solutions because a real square cannot be negative.

Similarly, y = 2(x − 4)² + 3 has minimum value 3, so its graph cannot meet y = 0. You do not need to expand it and compute a discriminant to reach that conclusion. A question about its intersections with y = 3 has one point; a question about y = 5 has two. The target horizontal line matters.

Choose the representation that makes the requested property easiest to see. Standard form exposes the discriminant. Factored form exposes roots. Vertex form exposes the vertex and a minimum or maximum. Flexibility across forms is more valuable than using the same formula on every problem.

Original practice set with explanations

Problem 1: how many distinct real solutions does 4x² + 4x + 1 = 0 have? The discriminant is 4² − 4(4)(1) = 16 − 16 = 0, so there is one distinct real solution. Factoring as (2x + 1)² = 0 confirms x = −1/2. The repeated factor does not create two distinct values of x.

Problem 2: for what value of c does 2x² − 12x + c = 0 have one distinct real solution? Set 144 − 8c = 0 to get c = 18. The equation becomes 2(x − 3)² = 0. Problem 3: if c is an integer, what is the smallest c producing no real solutions in that same equation? Require c > 18, so the answer is 19.

Problem 4: how many real solutions does x² + 5 = 2x have? Rewrite as x² − 2x + 5 = 0. The discriminant is 4 − 20 = −16, giving none. Completing the square yields (x − 1)² + 4 = 0, which confirms the conclusion independently.

Problem 5: the equation (k − 1)x² + 2x + 1 = 0 has exactly one real solution when k = 1. Explain why setting its discriminant to zero is unnecessary for that case. Substitution removes the quadratic term and leaves 2x + 1 = 0, a linear equation with one solution. This example demonstrates why the nonzero-leading-coefficient condition must be checked.

Check a calculator result intelligently

A graph can help confirm the number of intersections, but the viewing window can hide a distant root or make a near-tangent crossing hard to see. If an exact algebraic condition is available, use it to support the graph. Zooming is not a substitute for interpreting what the equation actually asks.

When solving a parameter boundary, substitute your candidate back into the discriminant and the original equation. The discriminant should become exactly zero for a quadratic with one distinct real root. Then check whether the leading coefficient remains nonzero. These checks are particularly valuable when parameters appear in more than one coefficient.

Build a focused practice routine

Practice a short mixed set containing a standard-form equation, an equation requiring rearrangement, a parameter inequality, a line-parabola system, and a vertex-form expression. For each, explain why your chosen method is efficient. A session containing only identical standard-form examples may build arithmetic fluency without testing whether you recognize when the tool applies.

Record mistakes by cause: wrong sign after rearrangement, incorrect square of a negative coefficient, missed inequality reversal, or failure to check whether the equation remains quadratic. Review those categories on a later day using new numbers. Your goal is a reliable decision process, not memorization of one worked example.

Frequently asked questions and sources

Does a positive discriminant mean integer roots? No. It means two distinct real roots. For integer coefficients, a perfect-square discriminant can lead to rational roots, but those roots need not be integers. Does a negative discriminant mean the quadratic expression never has any output? No; it means the equation setting that expression equal to zero has no real solutions.

The examples in this guide are original instructional material. College Board's Math overview describes the current domains, including Advanced Math. For related work, revisit quadratic expressions in your Math practice and use official practice tests for the authentic exam interface and scoring context.

College Board Math content overview

SAT Math practice guide