Squeeze theorem: Evaluating limits by bounding

By Vegard Gjerde • Based on Masterful Learning • 9 min read • Published
squeeze-theorem calculus limits math learning-strategies

The squeeze theorem lets you evaluate the limit of a function ff by sandwiching it between two bounding functions gg and hh that both converge to the same limit LL at the point of interest. When both outer limits agree at aa, ff is forced to converge there too. Recognizing when to set up bounds and verifying that both outer limits match is the core fluency skill trained by the Unisium Study System.

Unisium hero image titled Squeeze theorem showing the principle equation and a conditions card.
The squeeze theorem: if g(x)≤f(x)≤h(x)g(x) \le f(x) \le h(x) near aa and lim⁡x→ag=lim⁡x→ah=L\lim_{x \to a} g = \lim_{x \to a} h = L, then lim⁡x→af(x)=L\lim_{x \to a} f(x) = L.

On this page: The Principle | Conditions | Failure Modes | EE Questions | Retrieval Practice | Practice Ground | Solve a Problem | FAQ


The Principle

The move: Infer the limit of ff by bounding it between two functions gg and hh — both with the same limit LL — and letting the bounds force ff to converge.

The invariant: This preserves the limiting value forced by the bounds: once ff is trapped between two functions both approaching LL, the only compatible limit for ff is LL.

Pattern: g(x)≤f(x)≤h(x),lim⁡x→ag(x)=lim⁡x→ah(x)=L⟹lim⁡x→af(x)=Lg(x) \le f(x) \le h(x),\quad \lim_{x \to a} g(x) = \lim_{x \to a} h(x) = L \quad\Longrightarrow\quad \lim_{x \to a} f(x) = L

Legal ✓Illegal ✗
−x2≤f(x)≤x2-x^2 \le f(x) \le x^2 near 00; lim⁡x→0(−x2)=0=lim⁡x→0x2\lim_{x\to 0}(-x^2) = 0 = \lim_{x\to 0} x^2 — bounds agree, squeeze applies−1≤f(x)≤cos⁡x-1 \le f(x) \le \cos x near 00; lim⁡x→0(−1)=−1≠1=lim⁡x→0cos⁡x\lim_{x\to 0}(-1) = -1 \ne 1 = \lim_{x\to 0}\cos x — bounds disagree, condition fails

The right column shows the applicability condition failing: the bounding functions converge to different values, so there is no single limit they force ff toward.


Conditions of Applicability

Condition: g(x)≤f(x)≤h(x)nearag(x) \le f(x) \le h(x) near a; lim⁡g=lim⁡h\lim g = \lim h

Before applying, check: evaluate lim⁡x→ag(x)\lim_{x \to a} g(x) and lim⁡x→ah(x)\lim_{x \to a} h(x) independently — they must equal the same value LL.

  • Both bounding inequalities must hold on a full punctured neighborhood of aa — near aa from both sides, except possibly at aa itself. One-sided bounds support only a one-sided limit conclusion.
  • If lim⁡g≠lim⁡h\lim g \ne \lim h, the theorem does not apply — even tight, valid bounding inequalities cannot produce a limit conclusion when the outer limits disagree.

Want the complete framework behind this guide? Read Masterful Learning.


Common Failure Modes

Failure mode: claim the squeeze applies when lim⁡x→ag(x)≠lim⁡x→ah(x)\lim_{x \to a} g(x) \ne \lim_{x \to a} h(x) → no single value is forced on ff; the conclusion lim⁡f(x)=L\lim f(x) = L is unwarranted.

Debug: before writing the conclusion, verify lim⁡g\lim g and lim⁡h\lim h are equal — compute both separately and confirm they match.


Elaborative Encoding

Use these questions to build deep understanding. (See Elaborative Encoding for the full method.)

Within the Principle

  • Why must both bounding functions converge to the same value LL? What exactly prevents ff from settling at a different limit when the two outer limits agree?
  • Can the bounding inequality g(x)≤f(x)≤h(x)g(x) \le f(x) \le h(x) fail at the point x=ax = a itself and still allow the theorem to apply? Why or why not?

For the Principle

  • How do you decide which bounding strategy to use for a given expression? Compare the strategy for f(x)=xnsin⁡(1/x)f(x) = x^n \sin(1/x) with the strategy for f(x)=(sin⁡x)/xf(x) = (\sin x)/x near x=0x = 0.
  • If the bounds only hold for x>0x > 0, what can you conclude? What additional work is needed to obtain a two-sided limit statement?

Between Principles

  • How does the squeeze theorem complement the limit sum rule and limit product rule? In what situations do algebraic limit rules fail and the squeeze theorem becomes the right tool?

Generate an Example

  • Construct a limit where the bounding inequality holds near aa but lim⁡g≠lim⁡h\lim g \ne \lim h. Write out exactly why the squeeze theorem cannot be used, and state what an incorrect application would (wrongly) claim.

Retrieval Practice

Answer from memory, then click to reveal and check. (See Retrieval Practice for the full method.)

State the squeeze theorem in one sentence: _____If g(x) ≤ f(x) ≤ h(x) near a and the limits of g and h at a both equal L, then the limit of f(x) at a also equals L.
Write the canonical pattern: _____g(x)≤f(x)≤h(x), lim⁡x→ag(x)=lim⁡x→ah(x)=L⇒lim⁡x→af(x)=Lg(x) \le f(x) \le h(x),\ \lim_{x \to a} g(x)=\lim_{x \to a} h(x)=L \Rightarrow \lim_{x \to a} f(x)=L
State the canonical condition: _____g(x)≤f(x)≤h(x)neara;lim⁡g=lim⁡hg(x) \le f(x) \le h(x) near a; \lim g = \lim h

Practice Ground

Use these exercises to build move-selection fluency. (See Self-Explanation for how to use worked examples effectively.)

Procedure Walkthrough

Starting from lim⁡x→0x2sin⁡(1/x)\lim_{x \to 0} x^2 \sin(1/x), reach a single numeric value.

StepExpressionOperation
0lim⁡x→0x2sin⁡(1/x)\lim_{x \to 0} x^2 \sin(1/x)—
1−x2≤x2sin⁡(1/x)≤x2-x^2 \le x^2 \sin(1/x) \le x^2 for x≠0x \ne 0∣sin⁡(1/x)∣≤1\lvert\sin(1/x)\rvert \le 1; multiply through by x2≥0x^2 \ge 0
2lim⁡x→0(−x2)=0=lim⁡x→0x2\lim_{x \to 0}(-x^2) = 0 = \lim_{x \to 0} x^2Both bounding functions are polynomials; evaluate by substitution
3lim⁡x→0x2sin⁡(1/x)=0\lim_{x \to 0} x^2 \sin(1/x) = 0Squeeze theorem: ff is sandwiched between two functions sharing limit 00

Drills

Action label (Format B)

What rule licenses this conclusion? Name it and state why the condition is satisfied.

0≤x2sin⁡2(1/x)≤x20 \le x^2 \sin^2(1/x) \le x^2 for x≠0x \ne 0, and lim⁡x→0x2=0\lim_{x \to 0} x^2 = 0, therefore lim⁡x→0x2sin⁡2(1/x)=0\lim_{x \to 0} x^2 \sin^2(1/x) = 0.

Reveal

Squeeze theorem. The function x2sin⁡2(1/x)x^2 \sin^2(1/x) is sandwiched between 00 and x2x^2. Since 0≤sin⁡2(⋅)≤10 \le \sin^2(\cdot) \le 1 everywhere, the lower bound 00 and upper bound x2x^2 are both valid for x≠0x \ne 0. Condition check: lim⁡x→00=0=lim⁡x→0x2\lim_{x \to 0} 0 = 0 = \lim_{x \to 0} x^2. ✓


What was done between these two intermediate steps? Name the rule and verify the condition.

−x2≤x2cos⁡(1/x)≤x2 near 0,lim⁡x→0(−x2)=0=lim⁡x→0x2-x^2 \le x^2 \cos(1/x) \le x^2 \text{ near } 0, \quad \lim_{x \to 0}(-x^2) = 0 = \lim_{x \to 0} x^2 ↓\downarrow lim⁡x→0x2cos⁡(1/x)=0\lim_{x \to 0} x^2 \cos(1/x) = 0

Reveal

Squeeze theorem applied. The bounds are valid (since ∣cos⁡(1/x)∣≤1|\cos(1/x)| \le 1 means ∣x2cos⁡(1/x)∣≤x2|x^2 \cos(1/x)| \le x^2) and both converge to the same limit 00. Condition: lim⁡g=lim⁡h=0\lim g = \lim h = 0. ✓


Is the squeeze theorem applicable here? Explain your decision.

−x≤f(x)≤x+2-x \le f(x) \le x + 2 for all xx near 00. Given: lim⁡x→0(−x)=0\lim_{x \to 0}(-x) = 0 and lim⁡x→0(x+2)=2\lim_{x \to 0}(x + 2) = 2.

Reveal

No — the condition fails. The lower bound approaches 00 while the upper bound approaches 22. The squeeze theorem requires lim⁡g=lim⁡h\lim g = \lim h; here 0≠20 \ne 2, so the theorem does not apply. Even though ff is bounded, no conclusion about lim⁡x→0f(x)\lim_{x \to 0} f(x) follows from these bounds. This is a near-miss: the bounding inequality holds, but the shared-limit condition is violated.


Forward step (Format A)

Apply the squeeze theorem to evaluate the limit. Set up your bounds first.

lim⁡x→0xsin⁡(1/x)\lim_{x \to 0} x \sin(1/x)

Reveal

Use ∣sin⁡(1/x)∣≤1|\sin(1/x)| \le 1 for x≠0x \ne 0: this gives ∣xsin⁡(1/x)∣≤∣x∣|x \sin(1/x)| \le |x|, so −∣x∣≤xsin⁡(1/x)≤∣x∣-|x| \le x \sin(1/x) \le |x|.

lim⁡x→0(−∣x∣)=0=lim⁡x→0∣x∣\lim_{x \to 0}(-|x|) = 0 = \lim_{x \to 0} |x|. Both bounds share limit 00. ✓

By the squeeze theorem: lim⁡x→0xsin⁡(1/x)=0\lim_{x \to 0} x \sin(1/x) = 0.


Apply the squeeze theorem to evaluate the limit.

lim⁡x→∞sin⁡xx\lim_{x \to \infty} \frac{\sin x}{x}

Reveal

Bound sin⁡x\sin x using −1≤sin⁡x≤1-1 \le \sin x \le 1. Divide by x>0x > 0 (inequality direction preserved):

−1x≤sin⁡xx≤1x\frac{-1}{x} \le \frac{\sin x}{x} \le \frac{1}{x}

lim⁡x→∞ ⁣(−1x)=0=lim⁡x→∞1x\lim_{x \to \infty}\!\left(-\dfrac{1}{x}\right) = 0 = \lim_{x \to \infty}\dfrac{1}{x}. Both share limit 00. ✓

By the squeeze theorem: lim⁡x→∞sin⁡xx=0\lim_{x \to \infty}\dfrac{\sin x}{x} = 0.


A student claims the squeeze theorem proves lim⁡x→0f(x)=0\lim_{x \to 0} f(x) = 0. Is the claim valid?

Proposed bounds: 0≤f(x)≤x20 \le f(x) \le x^2 for x>0x > 0. No information is given about f(x)f(x) for x<0x \lt 0.

Reveal

Invalid for the two-sided limit. The bounds 0≤f(x)≤x20 \le f(x) \le x^2 only hold for x>0x > 0, not on a full punctured neighborhood of 00. The squeeze theorem operates on a full neighborhood (both sides unless a one-sided limit is the goal). With only one-sided bounds, the correct conclusion is the right-hand limit: lim⁡x→0+f(x)=0\lim_{x \to 0^+} f(x) = 0. To claim lim⁡x→0f(x)=0\lim_{x \to 0} f(x) = 0, matching bounds for x<0x \lt 0 are also required.


Apply the squeeze theorem to evaluate the limit.

lim⁡x→0x4cos⁡(1/x2)\lim_{x \to 0} x^4 \cos(1/x^2)

Reveal

Use ∣cos⁡(1/x2)∣≤1|\cos(1/x^2)| \le 1 for x≠0x \ne 0: since x4≥0x^4 \ge 0, this gives −x4≤x4cos⁡(1/x2)≤x4-x^4 \le x^4 \cos(1/x^2) \le x^4.

lim⁡x→0(−x4)=0=lim⁡x→0x4\lim_{x \to 0}(-x^4) = 0 = \lim_{x \to 0} x^4. Both share limit 00. ✓

By the squeeze theorem: lim⁡x→0x4cos⁡(1/x2)=0\lim_{x \to 0} x^4 \cos(1/x^2) = 0.


Apply the squeeze theorem to evaluate the limit.

lim⁡x→∞cos⁡2xx\lim_{x \to \infty} \frac{\cos^2 x}{x}

Reveal

Bound cos⁡2x\cos^2 x using 0≤cos⁡2x≤10 \le \cos^2 x \le 1. Divide by x>0x > 0:

0≤cos⁡2xx≤1x0 \le \frac{\cos^2 x}{x} \le \frac{1}{x}

lim⁡x→∞0=0=lim⁡x→∞1x\lim_{x \to \infty} 0 = 0 = \lim_{x \to \infty}\dfrac{1}{x}. Both share limit 00. ✓

By the squeeze theorem: lim⁡x→∞cos⁡2xx=0\lim_{x \to \infty}\dfrac{\cos^2 x}{x} = 0.


Transition identification (Format C)

In the evaluation chain below, which step applies the squeeze theorem? State how you recognize it.

To evaluate lim⁡x→0x2sin⁡(1/x)\lim_{x \to 0} x^2 \sin(1/x), the following steps are taken:

(A) Use ∣sin⁡(1/x)∣≤1|\sin(1/x)| \le 1 to write −x2≤x2sin⁡(1/x)≤x2-x^2 \le x^2 \sin(1/x) \le x^2 for x≠0x \ne 0.

(B) Compute lim⁡x→0(−x2)=0\lim_{x \to 0}(-x^2) = 0 and lim⁡x→0x2=0\lim_{x \to 0} x^2 = 0.

(C) Conclude lim⁡x→0x2sin⁡(1/x)=0\lim_{x \to 0} x^2 \sin(1/x) = 0.

Reveal

Step (C) applies the squeeze theorem. That is the step that draws the limit conclusion from established bounds. Steps (A) and (B) provide the two prerequisites — the bounding inequality and the matching outer limits. The squeeze theorem is the rule that combines those two pieces into the final conclusion.


Which of the following setups satisfies all conditions for the squeeze theorem? Assume x→0x \to 0 and ff is defined for all x≠0x \ne 0 near 00.

(i) −x2≤f(x)≤x2-x^2 \le f(x) \le x^2 for all x≠0x \ne 0 near 00; lim⁡x→0(−x2)=0=lim⁡x→0x2\lim_{x\to 0}(-x^2) = 0 = \lim_{x\to 0} x^2

(ii) 0≤f(x)≤∣x∣0 \le f(x) \le |x| for all x≠0x \ne 0 near 00; lim⁡x→00=0=lim⁡x→0∣x∣\lim_{x\to 0} 0 = 0 = \lim_{x\to 0}|x|

(iii) 0≤f(x)≤x20 \le f(x) \le x^2 for x>0x > 0 only; no bound given for x<0x \lt 0

(iv) −1≤f(x)≤cos⁡x-1 \le f(x) \le \cos x for all x≠0x \ne 0 near 00; lim⁡x→0(−1)=−1\lim_{x\to 0}(-1) = -1, lim⁡x→0cos⁡x=1\lim_{x\to 0}\cos x = 1

Reveal

(i) and (ii) satisfy all conditions. Both have valid bounding inequalities holding near 00 from both sides, and both pairs of bounding functions share the same limit (00). The squeeze theorem gives lim⁡x→0f(x)=0\lim_{x\to 0} f(x) = 0 in each case.

(iii) fails because the bounds only hold one-sided (x>0x > 0). The squeeze theorem requires a full punctured neighborhood. Only lim⁡x→0+f(x)=0\lim_{x \to 0^+} f(x) = 0 can be concluded.

(iv) fails because lim⁡g=−1≠1=lim⁡h\lim g = -1 \ne 1 = \lim h. The outer limits disagree, violating the shared-limit condition — the most common near-miss when using this theorem.


Solve a Problem

Apply what you’ve learned with Problem Solving.

Problem: Evaluate lim⁡x→∞cos⁡xx\lim_{x \to \infty} \dfrac{\cos x}{x} using the squeeze theorem. Set up appropriate bounds, verify that both outer limits agree, and state the conclusion.

Full solution
StepExpressionMove
0lim⁡x→∞cos⁡xx\lim_{x \to \infty} \dfrac{\cos x}{x}—
1−1≤cos⁡x≤1-1 \le \cos x \le 1Bound cosine using its range [−1,1][-1, 1]
2−1x≤cos⁡xx≤1x\dfrac{-1}{x} \le \dfrac{\cos x}{x} \le \dfrac{1}{x} for x>0x > 0Divide through by x>0x > 0; direction preserved
3lim⁡x→∞ ⁣(−1x)=0=lim⁡x→∞1x\lim_{x \to \infty}\!\left(-\dfrac{1}{x}\right) = 0 = \lim_{x \to \infty}\dfrac{1}{x}Both bounds are rational with vanishing numerator; evaluate by limit at infinity
4lim⁡x→∞cos⁡xx=0\lim_{x \to \infty} \dfrac{\cos x}{x} = 0Squeeze theorem: ff is bounded between two functions both converging to 00

PrincipleRelationship
Limit statementThe target claim that squeeze ultimately establishes: bounding is one route to proving that a two-sided limit statement is true
Limit sum ruleAlgebraic decomposition for sums; only works when individual limits exist — squeeze fills the gap when a factor oscillates without a standalone limit
Limit product ruleDecomposes a product limit; also requires each factor to have a finite limit — squeeze handles the case where one factor has no limit on its own
L’Hôpital’s ruleDifferent technique for 0/00/0 and ∞/∞\infty/\infty indeterminate forms; complementary tool, not a substitute for bounding

FAQ

What is the squeeze theorem?

The squeeze theorem (also called the sandwich theorem or pinching theorem) states that if ff is bounded between gg and hh near a point aa, and both gg and hh approach the same limit LL, then ff must also approach LL. It is a standard tool for evaluating many oscillatory limits — such as xnsin⁡(1/x)x^n \sin(1/x) — where direct substitution and algebraic decomposition rules do not apply.

When does the squeeze theorem apply?

Two conditions must hold simultaneously: (1) the inequality g(x)≤f(x)≤h(x)g(x) \le f(x) \le h(x) must hold on a punctured neighborhood of aa — near aa from both sides, except possibly at aa itself; and (2) both lim⁡x→ag(x)\lim_{x \to a} g(x) and lim⁡x→ah(x)\lim_{x \to a} h(x) must equal the same value LL.

What happens if the bounding functions have different limits?

If lim⁡x→ag(x)≠lim⁡x→ah(x)\lim_{x \to a} g(x) \ne \lim_{x \to a} h(x), the squeeze theorem is not applicable. The bounds do not force ff to converge to any particular value, even if the bounding inequality itself holds. The shared-limit condition is not optional — it is what makes the squeeze argument work.

How do I find the right bounds?

The most common strategy uses ∣sin⁡θ∣≤1|\sin \theta| \le 1 and ∣cos⁡θ∣≤1|\cos \theta| \le 1 for all θ\theta. When f(x)=xnsin⁡(1/x)f(x) = x^n \sin(1/x) or f(x)=xncos⁡(1/x)f(x) = x^n \cos(1/x), bound the oscillating factor between −1-1 and 11, then multiply through by xnx^n (or ∣xn∣|x^n| if xnx^n can be negative near aa). Both resulting bounds will share the same limit whenever xn→0x^n \to 0 as x→ax \to a.

How is the squeeze theorem different from L’Hôpital’s rule?

L’Hôpital’s rule handles indeterminate quotient forms (0/00/0, ∞/∞\infty/\infty) by differentiating numerator and denominator. The squeeze theorem handles limits where ff is hard to differentiate or has no quotient structure, by bounding it between simpler functions. For lim⁡x→0xsin⁡(1/x)\lim_{x\to 0} x\sin(1/x), the squeeze theorem is the natural tool; L’Hôpital’s rule is not directly applicable since the function does not form a differentiable quotient at x=0x = 0.

Does the squeeze theorem work for one-sided limits?

Yes. If the bounding inequality holds only for x>ax > a (or only for x<ax \lt a), you can conclude the corresponding one-sided limit. For the full two-sided limit lim⁡x→af(x)=L\lim_{x \to a} f(x) = L, the bounding inequality must hold on a full punctured neighborhood of aa.


How This Fits in Unisium

In Unisium, calculus fluency means knowing when to reach for a technique — not just executing it once chosen. The squeeze theorem trains a specific diagnostic habit: when direct substitution fails and algebraic decomposition rules are unavailable, ask whether ff can be bounded. Through retrieval practice and the drill formats above, you build automatic recognition of the bounding setup and the two-condition check, so the theorem becomes a tool you deploy confidently rather than vaguely recall.

Explore further:

  • Calculus Subdomain Map — Return to the calculus hub to see where squeeze sits inside the broader limits cluster
  • Limit statement — The theorem’s job is to justify a limit statement when direct substitution and basic algebra are not enough
  • Elaborative Encoding — Build deep understanding of why the shared-limit condition is necessary
  • Retrieval Practice — Make the condition and bounding pattern instantly accessible
  • Self-Explanation — Use the worked example above to practice explaining each bounding step out loud

Ready to master the squeeze theorem? See Unisium’s 14-day free trial or explore the full learning framework in Masterful Learning.

Masterful Learning book cover

Masterful Learning

The book behind these guides: a study system for physics, math, & programming built on retrieval, connection, explanation, and problem solving.

Ready to apply this strategy?

Unisium turns these evidence-based techniques into guided study sessions for math and physics. Places are limited during early access. Check current availability to start a trial; joining the mailing list is optional.

See plans and availability Read More Guides

Already have access? Sign in