Analyze rational functions

Distinguish zeros, holes and asymptotes; preserve restrictions through equivalent expressions.

What to practice

  • Distinguish zeros, holes and asymptotes; preserve restrictions through equivalent expressions.

Start with a graph that hides something

Graph r(x)=x2−x−6x2−4r(x)=\frac{x^2-x-6}{x^2-4} in a standard calculator window and you will probably see two smooth branches separated by a vertical gap near x=2x=2. Nothing seems to happen at x=−2x=-2. Yet r(−2)r(-2) is undefined, exactly as r(2)r(2) is. Why does one excluded input produce a dramatic asymptote while the other removes a single point that many screens never show?

This chapter answers that question and the related ones the exam asks: where a rational function equals zero, where it is positive, how it behaves for inputs of large magnitude, and when two formulas define the same function. The central habit is easy to state and easy to abandon: record the domain of the original expression before you simplify it, and keep that record until the end. The chapter covers topics 1.7 to 1.11 of the AP Precalculus Course and Exam Description.

What a rational function measures

A rational function is a quotient of two polynomial functions, r(x)=p(x)q(x)r(x)=\frac{p(x)}{q(x)}, where qq is not the zero polynomial. For each allowed input, r(x)r(x) compares the size of p(x)p(x) with the size of q(x)q(x). That one interpretation explains almost every feature in this chapter. When q(x)q(x) is close to zero while p(x)p(x) is not, the quotient is huge. When p(x)p(x) is zero while q(x)q(x) is not, the quotient is zero. When both are zero at the same input, the quotient is undefined there, and you must look at nearby inputs to learn more.

The domain is every real number except the real zeros of the denominator. Write it first. For the opening example, x2−4=(x−2)(x+2)x^2-4=(x-2)(x+2), so the domain is all real xx with x≠−2x\neq-2 and x≠2x\neq2.

The Fall 2026 clarification of the course framework states that two functions are equal only when they have the same domain and give the same output for every input in it (CED clarifications and corrections). Factoring and cancelling produce a simpler expression, but that expression usually has a larger natural domain. Treat it as a tool for studying the original function, not as a replacement for it.

End behavior comes from leading terms

For inputs of large magnitude, every polynomial is dominated by its leading term. A rational function's end behavior therefore follows the quotient of the leading terms of its numerator and denominator. Three cases result.

Degree comparisonQuotient of leading termsEnd behavior of the graph
Numerator degree smallerconstant over a nonconstant powerhorizontal asymptote y=0y=0
Degrees equala nonzero constant bbhorizontal asymptote y=by=b
Numerator degree largera nonconstant polynomialfollows that polynomial; slant asymptote when the degrees differ by one

A horizontal asymptote y=by=b means the outputs get arbitrarily close to bb and stay arbitrarily close as the inputs increase or decrease without bound. In limit notation, lim⁡x→∞r(x)=b\lim_{x\to\infty} r(x)=b or lim⁡x→−∞r(x)=b\lim_{x\to-\infty} r(x)=b. The asymptote describes far-away behavior only. It says nothing about inputs in the middle of the graph.

Worked example 5.1: three end behaviors

Describe the end behavior of f(x)=6x2−x3x2+5f(x)=\frac{6x^2-x}{3x^2+5}, g(x)=5x+1x2−4g(x)=\frac{5x+1}{x^2-4} and h(x)=x3+2x−1h(x)=\frac{x^3+2}{x-1}.

For ff, the leading terms give 6x23x2=2\frac{6x^2}{3x^2}=2, so lim⁡x→±∞f(x)=2\lim_{x\to\pm\infty} f(x)=2 and the horizontal asymptote is y=2y=2. For gg, the leading terms give 5xx2=5x\frac{5x}{x^2}=\frac{5}{x}, which approaches 0, so y=0y=0 is the horizontal asymptote. For hh, the leading terms give x3x=x2\frac{x^3}{x}=x^2. The graph of hh has no horizontal or slant asymptote; both ends rise like x2x^2, so lim⁡x→∞h(x)=∞\lim_{x\to\infty} h(x)=\infty and lim⁡x→−∞h(x)=∞\lim_{x\to-\infty} h(x)=\infty.

Check. f(100)=5990030005≈1.996f(100)=\frac{59900}{30005}\approx1.996, close to 2 as predicted. When a function is given in factored form, multiply only the leading parts of the factors. In (2−x)(x+3)(x+1)(4x−1)\frac{(2-x)(x+3)}{(x+1)(4x-1)} the leading terms are (−x)(x)=−x2(-x)(x)=-x^2 and (x)(4x)=4x2(x)(4x)=4x^2, so the horizontal asymptote is y=−14y=-\frac{1}{4}. The constant terms, which give 6−1\frac{6}{-1}, play no part.

Worked example 5.2: a graph that meets its asymptote

Let k(x)=3x2+x−2x2−x+1k(x)=\frac{3x^2+x-2}{x^2-x+1}. The denominator has discriminant (−1)2−4(1)(1)=−3(-1)^2-4(1)(1)=-3, so it has no real zeros and the domain is all real numbers. Equal degrees give the horizontal asymptote y=3y=3.

Can the graph reach y=3y=3? Solve k(x)=3k(x)=3. Because the denominator is never zero, multiplying by it is safe: 3x2+x−2=3x2−3x+33x^2+x-2=3x^2-3x+3, so 4x=54x=5 and x=54x=\frac{5}{4}. The graph meets its horizontal asymptote at (54,3)(\frac{5}{4},3), rises above it, and then approaches 3 from above as xx increases (Figure 5.1a). A horizontal asymptote describes the ends of a graph; it is not a fence.

Worked example 5.3: find a slant asymptote by division

Let s(x)=x2+x−1x−2s(x)=\frac{x^2+x-1}{x-2}. The numerator's degree is one more than the denominator's, so expect a slant asymptote. Polynomial long division rewrites a dividend as divisor times quotient plus remainder, with the remainder's degree less than the divisor's. Divide x2x^2 by xx to get xx; subtract x(x−2)=x2−2xx(x-2)=x^2-2x to leave 3x−13x-1. Divide 3x3x by xx to get 3; subtract 3(x−2)=3x−63(x-2)=3x-6 to leave the remainder 5. So

x2+x−1=(x−2)(x+3)+5x^2+x-1=(x-2)(x+3)+5

and, for x≠2x\neq2, s(x)=x+3+5x−2s(x)=x+3+\frac{5}{x-2}. As ∣x∣|x| grows, 5x−2\frac{5}{x-2} approaches 0, so the graph approaches the line y=x+3y=x+3. Check at x=12x=12: s(12)=15510=15.5s(12)=\frac{155}{10}=15.5, the line gives 15, and the remainder term gives exactly 510=0.5\frac{5}{10}=0.5. The leading-term quotient alone, x2x=x\frac{x^2}{x}=x, gives the correct slope but the wrong line; only the full division supplies the constant 3.

Two side-by-side graphs. Panel a: the graph of k of x equals the quotient of 3x squared plus x minus 2 and x squared minus x plus 1, for x from negative 10 to 10, with a dotted horizontal line at y equals 3. On the far left the curve sits just below y equals 3; moving right, it falls to a minimum near y equals negative 2 just right of the origin, climbs through the marked point (5/4, 3), peaks near y equals 4 around x equals 2.4, and then decreases toward 3 from above. Panel b: the graph of s of x equals the quotient of x squared plus x minus 1 and x minus 2, with a dashed vertical asymptote at x equals 2 and a dash-dot slant asymptote y equals x plus 3. The left branch stays below the slant line and falls without bound as x approaches 2 from the left; the right branch comes down from very large values just right of 2, reaches a low point near y equals 9.5, and then rises along the slant line.

Figure 5.1. End behavior is about far inputs. (a) A graph can meet its horizontal asymptote and still approach it at both ends. (b) Long division gives the slant asymptote y=x+3y=x+3.

Zeros and sign changes

The real zeros of a rational function are the real zeros of its numerator that belong to its domain. Find the numerator's zeros, then test each against the original denominator. A numerator zero that also makes the original denominator zero is not a zero of the function; it is an excluded input, usually a hole.

To solve r(x)≥0r(x)\geq0 or r(x)≤0r(x)\leq0, mark every real zero of the numerator and of the denominator on a number line. These are the only places where the sign of rr can change, because a quotient of polynomials changes sign only at a zero or at an excluded input. Test one input in each interval, or count negative factors. Numerator zeros are included when the inequality allows equality; denominator zeros are never included. A factor raised to an even power does not change sign.

Worked example 5.4: solve a rational inequality

Solve w(x)=(x+2)(x−3)(x−1)2≤0w(x)=\frac{(x+2)(x-3)}{(x-1)^2}\leq0.

The boundary inputs are −2-2 and 3 (numerator zeros) and 1 (denominator zero). Test −3-3, 0, 2 and 4: w(−3)=(−1)(−6)16>0w(-3)=\frac{(-1)(-6)}{16}>0, w(0)=(2)(−3)1<0w(0)=\frac{(2)(-3)}{1}<0, w(2)=(4)(−1)1<0w(2)=\frac{(4)(-1)}{1}<0 and w(4)=(6)(1)9>0w(4)=\frac{(6)(1)}{9}>0. The sign does not change at 1 because (x−1)2(x-1)^2 is positive on both sides. Include −2-2 and 3, where ww equals 0; exclude 1. The solution is [−2,1)∪(1,3][-2,1)\cup(1,3] (Figure 5.2).

Multiplying both sides of an inequality by a denominator is valid only when you know the denominator's sign. Here (x−1)2>0(x-1)^2>0 for x≠1x\neq1, so multiplying gives (x+2)(x−3)≤0(x+2)(x-3)\leq0, whose solution [−2,3][-2,3] still needs 1 removed. For a denominator such as x−1x-1, whose sign changes at 1, the same shortcut produces wrong intervals.

A sign chart with four interval columns: x less than negative 2, between negative 2 and 1, between 1 and 3, and greater than 3. Row x plus 2: negative, positive, positive, positive. Row x minus 3: negative, negative, negative, positive. Row (x minus 1) squared: positive in all four columns. Row w of x: positive, negative, negative, positive. Below, a number line for w of x less than or equal to 0 is shaded from negative 2 to 3, with filled dots at negative 2 and 3 labeled zero: included, and an open dot at 1 labeled excluded.

Figure 5.2. Sign chart for w(x)=(x+2)(x−3)(x−1)2w(x)=\frac{(x+2)(x-3)}{(x-1)^2}. The squared factor never changes sign, but the input 1 is still excluded.

Vertical asymptotes and holes: compare multiplicities

Every real zero of the original denominator is excluded from the domain, but excluded inputs produce two different graph features. Factor both polynomials completely and compare the multiplicity of each zero aa of the denominator.

  • If aa has greater multiplicity in the denominator than in the numerator (including multiplicity 0 in the numerator), the graph has a vertical asymptote x=ax=a. Near aa the denominator is close to 0 while the reduced numerator is not, so the outputs increase or decrease without bound.
  • If aa has multiplicity in the numerator greater than or equal to its multiplicity in the denominator, the graph has a hole at x=ax=a. The function is undefined there, but outputs for inputs near aa approach a single value LL, and the hole is the point (a,L)(a,L).

Describe asymptote behavior with one-sided limits: lim⁡x→a+r(x)\lim_{x\to a^{+}} r(x) uses inputs slightly greater than aa, and lim⁡x→a−r(x)\lim_{x\to a^{-}} r(x) uses inputs slightly less. To find the sign, decide the sign of each factor for an input just to one side. At a hole, both one-sided limits equal LL, so lim⁡x→ar(x)=L\lim_{x\to a} r(x)=L, and you can compute LL by evaluating the reduced expression at aa.

Worked example 5.5: a hole and an asymptote in one function

Return to r(x)=x2−x−6x2−4=(x−3)(x+2)(x−2)(x+2)r(x)=\frac{x^2-x-6}{x^2-4}=\frac{(x-3)(x+2)}{(x-2)(x+2)}, with domain x≠−2x\neq-2, x≠2x\neq2.

At x=−2x=-2 the factor x+2x+2 appears once in each polynomial, so the multiplicities are equal: a hole. On the domain, r(x)=x−3x−2r(x)=\frac{x-3}{x-2}, and at x=−2x=-2 that reduced expression gives −5−4=54\frac{-5}{-4}=\frac{5}{4}. The hole is (−2,54)(-2,\frac{5}{4}). At x=2x=2 the denominator factor has no match: a vertical asymptote. Just left of 2, r(1.9)=−1.1−0.1=11r(1.9)=\frac{-1.1}{-0.1}=11; just right, r(2.1)=−0.90.1=−9r(2.1)=\frac{-0.9}{0.1}=-9. So lim⁡x→2−r(x)=∞\lim_{x\to2^{-}} r(x)=\infty and lim⁡x→2+r(x)=−∞\lim_{x\to2^{+}} r(x)=-\infty. The zero is x=3x=3, which is in the domain, and the horizontal asymptote is y=1y=1. The two exclusions behave differently because only at x=2x=2 does the denominator approach 0 while the numerator stays away from 0 (Figure 5.3).

Graph of r of x equals the quotient of x squared minus x minus 6 and x squared minus 4, for x from negative 6 to 7 and y from negative 6 to 6. A dotted horizontal asymptote is at y equals 1 and a dashed vertical asymptote at x equals 2. The left branch runs slightly above y equals 1 on the far left, passes an open circle at (negative 2, 1.25), and rises without bound as x approaches 2 from the left. The right branch rises from below the window just right of x equals 2, crosses the x-axis at the filled point (3, 0), and approaches y equals 1 from below. An enlarged inset around x equals negative 2 shows the open circle on an otherwise smooth curve at height 1.25.

Figure 5.3. One function, two exclusions: a hole at (−2,54)(-2,\frac{5}{4}) and a vertical asymptote x=2x=2. The inset enlarges the missing point that a calculator screen can hide.

Worked example 5.6: a hole on the x-axis

Let m(x)=(x+1)2(x−2)(x+1)(x−2)3m(x)=\frac{(x+1)^2(x-2)}{(x+1)(x-2)^3}, with domain x≠−1x\neq-1, x≠2x\neq2. At x=−1x=-1 the numerator multiplicity 2 is greater than the denominator multiplicity 1, so the graph has a hole. On the domain, m(x)=x+1(x−2)2m(x)=\frac{x+1}{(x-2)^2}, which gives 0 at x=−1x=-1: the hole is (−1,0)(-1,0). The missing point lies on the xx-axis, yet −1-1 is not a zero of mm, because m(−1)m(-1) does not exist. At x=2x=2 the denominator multiplicity 3 is greater than 1, so x=2x=2 is a vertical asymptote. The remaining factor (x−2)2(x-2)^2 is squared and the reduced numerator is 3 there, so both one-sided limits are ∞\infty; for example, m(1.9)=290m(1.9)=290 and m(2.1)=310m(2.1)=310. This function has no zeros at all.

Equivalent forms with a purpose

Different equivalent forms answer different questions. Factored form reveals zeros, excluded inputs, holes, vertical asymptotes and sign changes. Standard form, or the quotient-plus-remainder form from long division, reveals end behavior. Choose the form for the question, and carry the original domain through every rewrite.

Worked example 5.7: equal expressions, unequal functions

Compare u(x)=x2−16x−4u(x)=\frac{x^2-16}{x-4} with v(x)=x+4v(x)=x+4. For every x≠4x\neq4, u(x)=(x−4)(x+4)x−4=x+4u(x)=\frac{(x-4)(x+4)}{x-4}=x+4. But u(4)u(4) is undefined while v(4)=8v(4)=8. The domains differ, so uu and vv are not equal functions; the graph of uu is the line y=x+4y=x+4 with a hole at (4,8)(4,8). A correct rewrite keeps the restriction: u(x)=x+4u(x)=x+4 for x≠4x\neq4.

Worked example 5.8: expand with the binomial theorem

The binomial theorem expands (a+b)n(a+b)^n using one row of Pascal's triangle. Each row starts and ends with 1, and each interior entry is the sum of the two entries above it; row 4 is 1, 4, 6, 4, 1. The general term is (nk)an−kbk\binom{n}{k}a^{n-k}b^{k}, where (nk)\binom{n}{k} is entry kk of row nn, counting from 0.

Expand p(x)=(x−2)4p(x)=(x-2)^4 with a=xa=x and b=−2b=-2:

(x−2)4=x4+4x3(−2)+6x2(−2)2+4x(−2)3+(−2)4(x-2)^4=x^4+4x^3(-2)+6x^2(-2)^2+4x(-2)^3+(-2)^4

(x−2)4=x4−8x3+24x2−32x+16(x-2)^4=x^4-8x^3+24x^2-32x+16

The signs alternate because odd powers of −2-2 are negative. Check at x=1x=1: (1−2)4=1(1-2)^4=1 and 1−8+24−32+16=11-8+24-32+16=1. The factored form shows a single zero at x=2x=2 of multiplicity 4; the standard form shows the leading term x4x^4 and the yy-intercept 16.

Use technology without losing restrictions

On calculator sections, a graphing utility helps you find numerator zeros that do not factor, solve rational equations and inequalities numerically, and examine outputs near an excluded input. It cannot replace the domain record. A graph drawn from sampled points may not show a hole at all, and a table may give the same undefined message at a hole and at an asymptote. When a table reports no value at x=ax=a, evaluate inputs such as a−0.001a-0.001 and a+0.001a+0.001. Outputs that agree and settle near one number indicate a hole; outputs that grow in magnitude indicate an asymptote. For the opening function, r(−2.001)≈1.2499r(-2.001)\approx1.2499 and r(−1.999)≈1.2501r(-1.999)\approx1.2501, while r(1.999)=1001r(1.999)=1001 and r(2.001)=−999r(2.001)=-999.

When you clear denominators to solve r(x)=g(x)r(x)=g(x), the resulting polynomial equation is defined at the excluded inputs, so it can have solutions the original equation does not. Substitute every candidate into the original denominators and reject any that make one of them zero. A contextual model also has a domain set by the situation, such as nonnegative time or a tank's capacity; report only solutions inside it. Keep extra decimal places while you work, then round final answers to three decimal places unless told otherwise.

Common errors to repair

  • Cancelling and forgetting. Simplifying (x−3)(x+2)(x−2)(x+2)\frac{(x-3)(x+2)}{(x-2)(x+2)} to x−3x−2\frac{x-3}{x-2} and then treating −2-2 as allowed. Repair: write the domain before factoring and copy it beside every rewrite.
  • Every excluded input is an asymptote. Repair: compare multiplicities; a shared factor whose numerator multiplicity is at least as large gives a hole.
  • A cancelled numerator zero listed as an x-intercept. Repair: a zero must make the original function equal 0, so it must be in the domain.
  • The asymptote as a fence. Repair: solve r(x)=br(x)=b to see whether the graph meets its horizontal asymptote.
  • Constant terms deciding end behavior. Repair: compare only leading terms; in factored form, multiply the leading parts.
  • Brackets at denominator zeros. Repair: excluded inputs always get parentheses in a solution set.
  • A sign change at every boundary. Repair: even-power factors keep their sign.
  • The leading-term quotient as the slant asymptote. Repair: divide completely; the whole quotient polynomial is the asymptote.

Try it

Work these on paper before reading the solutions.

  1. For f(x)=x2−5x+6x2−x−6f(x)=\frac{x^2-5x+6}{x^2-x-6}, state the domain, the zeros, any hole, the vertical asymptotes and the horizontal asymptote.
  2. Solve x2−9x2+2x≤0\frac{x^2-9}{x^2+2x}\leq0.
  3. Rewrite x3−2x+3x2+1\frac{x^3-2x+3}{x^2+1} as a polynomial plus a fraction whose numerator has smaller degree than its denominator. State the slant asymptote and find where the graph meets it.
  4. Find the coefficient of x2x^2 in the expansion of (x+3)4(x+3)^4.
  5. Write a rational function with a zero at x=4x=4, a hole at x=−1x=-1, a vertical asymptote x=3x=3 and a horizontal asymptote y=2y=2. Give the hole's coordinates.

Check your try-it work

  1. f(x)=(x−2)(x−3)(x−3)(x+2)f(x)=\frac{(x-2)(x-3)}{(x-3)(x+2)}, so the domain is x≠3x\neq3, x≠−2x\neq-2. On it, f(x)=x−2x+2f(x)=\frac{x-2}{x+2}. Zero: x=2x=2. Hole: (3,15)(3,\frac{1}{5}). Vertical asymptote: x=−2x=-2. Horizontal asymptote: y=1y=1. If you listed 3 as a zero, reread the section on zeros.
  2. Factor: (x−3)(x+3)x(x+2)≤0\frac{(x-3)(x+3)}{x(x+2)}\leq0. The boundaries are −3-3, −2-2, 0 and 3, and the signs from left to right are ++, −-, ++, −-, ++. The solution is [−3,−2)∪(0,3][-3,{-}2)\cup(0,3], with brackets at the numerator zeros and parentheses at the excluded inputs.
  3. x3−2x+3=(x2+1)(x)+(−3x+3)x^3-2x+3=(x^2+1)(x)+(-3x+3), so the expression equals x+−3x+3x2+1x+\frac{-3x+3}{x^2+1}. The slant asymptote is y=xy=x. The graph meets it where the remainder is 0: −3x+3=0-3x+3=0, so at (1,1)(1,1). Check: 1−2+32=1\frac{1-2+3}{2}=1. The denominator is never zero, so there is no vertical asymptote.
  4. The term is (42)x2(3)2=6⋅9x2=54x2\binom{4}{2}x^2(3)^2=6\cdot9x^2=54x^2. The coefficient is 54, not 6.
  5. One answer is r(x)=2(x−4)(x+1)(x+1)(x−3)r(x)=\frac{2(x-4)(x+1)}{(x+1)(x-3)}. At x=−1x=-1, the reduced form 2(x−4)x−3\frac{2(x-4)}{x-3} gives −10−4=52\frac{-10}{-4}=\frac{5}{2}, so the hole is (−1,52)(-1,\frac{5}{2}). Other correct answers exist, such as using (x+1)2(x+1)^2 in both polynomials; check that the leading-term quotient is 2.

Review and retrieval

Close the book and write, from memory, the rule that separates a hole from a vertical asymptote and the three end-behavior cases. Then explain in two sentences why x2−16x−4\frac{x^2-16}{x-4} and x+4x+4 are different functions. Tomorrow, without notes, invent a function with one hole and one vertical asymptote, analyze it as in Try it 1, and check your hole with a table near the excluded input. In about a week, solve the worked inequality again with the inequality sign reversed. Chapter 4's multiplicity ideas support this chapter; Chapter 6 transforms these graphs; and Chapters 10, 12 and 16 reuse the same domain habit for compositions, logarithmic equations and trigonometric identities.

Practice

Twenty-four multiple-choice questions, Questions 73–96, and one free-response task, Written task 3, are printed after this chapter. Answer Questions 73–91, the first nineteen, without a calculator. Questions 92–96 and Written task 3 require a graphing calculator, and each of their stems says so; Written task 3 is a six-point function-concepts task with two pages of response space. Work under timed conditions before checking. Keys, the reasoning for every choice, the scoring guide and a complete sample response are under Chapter 5: answers and explanations in the answers section. Each explanation ends with a repair step that names the section of this chapter to revisit.

Use this chapter companion

Review the lesson, attempt the original practice, then check the explanation. Your printed study guide remains complete without the website.

Independence

Skill by Skill Prep is independent. SAT® and AP® are trademarks registered by the College Board, which is not affiliated with, and does not endorse, this website or book. PSAT/NMSQT® is a registered trademark of the College Board and the National Merit Scholarship Corporation, which are not affiliated with, and do not endorse, this site. Bluebook™ is a trademark of the College Board. ACT® is a trademark of ACT Education Corp., which is not affiliated with and does not endorse this website or book. Desmos is a trademark of Desmos Studio PBC. Practice results describe these questions, not official scores or admission predictions.

Last updated .