What Are Critical Points?
A critical point is a place on a function where the graph may change behavior in an important way. In single-variable calculus, these points are usually found by looking at the derivative. The derivative tells you the slope of the graph, so a zero derivative means the graph has a horizontal tangent, while an undefined derivative can signal a sharp corner, cusp, vertical tangent, or boundary of the function’s real-valued domain.
For a real-valued function \(f(x)\), a number \(c\) in the domain of the function is a critical number when:
$$
f^\prime(c) = 0
$$
or when:
$$
f^\prime(c) \text{ does not exist, while } f(c) \text{ is defined.}
$$
The corresponding critical point is the coordinate:
$$
(c, f(c)).
$$
Critical points matter because local maximum and local minimum values can only occur at special places: critical points or, when working on a closed interval, endpoints. Not every critical point is a maximum or minimum, but critical points are the natural starting places for studying where a function turns, flattens, or changes direction.
Why Critical Points Matter
Critical points help connect algebra, derivatives, and graphs. Instead of guessing from a graph alone, you can use the derivative to identify important candidate locations, then classify what happens near each one.
Students use critical points when learning curve sketching, optimization, and the first derivative test. Educators use them to show how the sign of \(f^\prime(x)\) explains increasing and decreasing behavior. In applied problems, critical points can represent possible best or worst values, such as the highest height, lowest cost, largest area, or smallest error, as long as the model and interval are appropriate.
The key word is “possible.” A critical point is a candidate for important behavior, not a final conclusion by itself. You still need to check the derivative sign, the function’s domain, and sometimes the interval endpoints.
Key Terms to Know
-
Function: A rule that assigns an output value \(f(x)\) to an input value \(x\).
-
Derivative: The function \(f^\prime(x)\) that describes the instantaneous rate of change or slope of \(f(x)\).
-
Critical number: An input value \(c\) where \(f^\prime(c)=0\) or where \(f^\prime(c)\) is undefined while \(f(c)\) exists.
-
Critical point: The coordinate \((c, f(c))\) associated with a critical number.
-
Stationary point: A point where the derivative is zero. It may be a maximum, a minimum, or neither.
-
Local maximum: A point where the function value is higher than nearby values.
-
Local minimum: A point where the function value is lower than nearby values.
-
First derivative test: A method that classifies a critical point by checking whether \(f^\prime(x)\) changes sign around it.
-
Domain: The set of input values where the function is defined as a real-valued function.
-
Numerical derivative: An approximate derivative found from nearby function values rather than from an exact symbolic derivative.
How Critical Points Work
The basic process is simple:
-
Start with a function \(f(x)\).
-
Find the derivative \(f^\prime(x)\), exactly or approximately.
-
Look for values of \(x\) where \(f^\prime(x)=0\).
-
Also consider values where \(f^\prime(x)\) is undefined but \(f(x)\) still exists.
-
Evaluate the original function at those \(x\)-values to get coordinates.
-
Use the derivative sign on each side of the point to classify the behavior.
The derivative sign tells you whether the function is increasing or decreasing. When \(f^\prime(x)>0\), the function is increasing. When \(f^\prime(x)<0\), the function is decreasing. This is why the first derivative test works.
| Derivative behavior near \(c\) |
Likely classification |
| \(f^\prime(x)\) changes from positive to negative |
Local maximum |
| \(f^\prime(x)\) changes from negative to positive |
Local minimum |
| \(f^\prime(x)\) does not change sign |
Stationary point, but not a local extremum |
| \(f^\prime(x)\) is undefined near the point |
Possible corner, cusp, vertical tangent, or domain boundary |
For polynomial functions, the derivative can often be found exactly with the power rule:
$$
\frac{d}{dx}\left(ax^n\right)=anx^{n-1}.
$$
For example, if:
$$
f(x)=x^3-3x^2-9x+4,
$$
then:
$$
f^\prime(x)=3x^2-6x-9.
$$
The critical numbers from the stationary-point condition are found by solving:
$$
3x^2-6x-9=0.
$$
For supported non-polynomial functions, a calculator may estimate the derivative numerically. A common central-difference estimate uses function values slightly to the left and right of \(x\):
$$
f^\prime(x) \approx \frac{f(x+h)-f(x-h)}{2h}.
$$
This estimate can be very useful, especially for functions involving trigonometric, logarithmic, exponential, square-root, cube-root, or absolute-value behavior. However, it is still an approximation. A numerical scan can miss some derivative roots, especially if the feature is very narrow or if the derivative touches zero without changing sign.
When a derivative changes sign over a small interval, a root-finding method such as bisection can refine the estimated solution. Bisection repeatedly narrows an interval where a sign change occurs. It is reliable when the sign-change condition is present, but it does not find every possible root in every situation.
Examples of Critical Points in Practice
Example 1: A cubic function
Consider:
$$
f(x)=x^3-3x^2-9x+4.
$$
Differentiate:
$$
f^\prime(x)=3x^2-6x-9.
$$
Factor:
$$
f^\prime(x)=3(x-3)(x+1).
$$
Set the derivative equal to zero:
$$
3(x-3)(x+1)=0.
$$
So the critical numbers are:
$$
x=-1 \quad \text{and} \quad x=3.
$$
Now evaluate the original function:
$$
f(-1)=(-1)^3-3(-1)^2-9(-1)+4=9,
$$
$$
f(3)=3^3-3(3)^2-9(3)+4=-23.
$$
The critical points are:
$$
(-1, 9) \quad \text{and} \quad (3, -23).
$$
The derivative changes from positive to negative at \(x=-1\), so \((-1,9)\) is a local maximum. It changes from negative to positive at \(x=3\), so \((3,-23)\) is a local minimum.
Example 2: A stationary point that is not always the same kind of point
Consider:
$$
f(x)=x^4-4x^2.
$$
Differentiate:
$$
f^\prime(x)=4x^3-8x=4x(x^2-2).
$$
Set the derivative equal to zero:
$$
4x(x^2-2)=0.
$$
The critical numbers are:
$$
x=-\sqrt{2}, \quad x=0, \quad x=\sqrt{2}.
$$
Evaluate the function:
$$
f(-\sqrt{2})=-4,
$$
$$
f(0)=0,
$$
$$
f(\sqrt{2})=-4.
$$
So the critical points are:
$$
(-\sqrt{2}, -4), \quad (0,0), \quad (\sqrt{2}, -4).
$$
Using the derivative sign, the two points with \(x=\pm\sqrt{2}\) are local minima, and \((0,0)\) is a local maximum. This example shows why the derivative alone is not enough; classification depends on how the derivative behaves around each point.
Example 3: A point where the derivative is undefined
Consider:
$$
f(x)=|x|.
$$
The function is defined at \(x=0\), and:
$$
f(0)=0.
$$
But the derivative is not defined at \(x=0\) because the graph has a sharp corner there. Since the function exists but the derivative does not, \(x=0\) is a critical number. The critical point is:
$$
(0,0).
$$
The function decreases as it approaches \(0\) from the left and increases after \(0\), so this point is a local minimum.
Example 4: A stationary point that is not a maximum or minimum
Consider:
$$
f(x)=x^3.
$$
Differentiate:
$$
f^\prime(x)=3x^2.
$$
The derivative is zero at:
$$
x=0.
$$
The critical point is:
$$
(0,0).
$$
However, \(f^\prime(x)=3x^2\) is positive on both sides of \(0\) except at the point itself. The function is increasing before and after \(0\), so the point is stationary but not a local maximum or local minimum.
How to Interpret the Result
A coordinate such as \((3,-23)\) means the calculator found a critical-point candidate at \(x=3\) and then evaluated the original function to get \(y=f(3)=-23\).
A result labeled Likely local maximum means the derivative changed from positive on the left to negative on the right. A result labeled Likely local minimum means the derivative changed from negative on the left to positive on the right.
A result labeled Stationary point means the derivative is near zero at that point, but the sign samples around it did not show a clear positive-to-negative or negative-to-positive change. This can happen at points like \(f(x)=x^3\) at \(x=0\), where the graph flattens but keeps increasing.
A result labeled Critical point with undefined derivative points to a place where the derivative appears not to exist near the candidate point. Common examples include sharp corners from absolute values and certain square-root or cube-root expressions.
If the result says No points found, read that as “no critical points were detected in the selected interval by this method.” For simple exact cases, that may be enough. For harder numerical cases, it should not be treated as a formal proof that no critical points exist.
If the result says that every \(x\) in the interval is stationary, the function is behaving like a constant polynomial on that interval. A constant function has derivative zero everywhere, so it does not have isolated stationary points.
The graph preview and point table are useful for checking whether the listed points match the visible shape of the curve. The derivative sign hint helps explain the classification by showing the sign of \(f^\prime(x)\) on each side of the point.
Common Mistakes and Misconceptions
One common mistake is assuming every critical point is a maximum or minimum. A critical point can also be a flat point where the function keeps increasing or keeps decreasing.
Another mistake is ignoring the domain. A point only counts as a critical point if the original function is defined there. If \(f(x)\) is not real-valued at a certain \(x\), then that coordinate cannot be a critical point of the real-valued function.
Students also often confuse local extrema with absolute extrema. A local maximum is high compared with nearby points. An absolute maximum is the highest value over the whole interval or domain being studied. To find absolute extrema on a closed interval, you normally compare function values at both endpoints and all relevant critical points. This calculator focuses on critical points and does not separately compute endpoint maxima or endpoint minima.
Trigonometric inputs are another common source of confusion. Functions such as \(\sin(x)\), \(\cos(x)\), and \(\tan(x)\) use radians, not degrees. For example, \(\sin(\pi/2)=1\), while \(\sin(90)\) means the sine of \(90\) radians, not \(90^\circ\).
It is also easy to choose an interval that excludes the point you want to study. If a critical point is outside the range from \(x\ \text{min}\) to \(x\ \text{max}\), it will not appear in the results.
Finally, numerical output should not be mistaken for exact proof. Decimal coordinates, numerical derivative estimates, and graph markers are helpful, but exact symbolic work or a separate proof may be needed for formal calculus assignments.
When to Use Critical Points
Use critical points when you want to understand where a function may turn, flatten, or change behavior. They are especially useful when:
-
Sketching the shape of a function.
-
Finding candidates for local maxima and local minima.
-
Preparing for an optimization problem.
-
Checking where a polynomial has horizontal tangent lines.
-
Exploring functions that include supported trigonometric, exponential, logarithmic, square-root, cube-root, or absolute-value terms.
-
Comparing symbolic derivative work with a graph-based or numerical check.
Critical points are also useful as a first step before deeper analysis. After finding them, you may still need to check intervals, endpoints, second derivatives, domain restrictions, or the original context of the problem.
Limitations and Things to Keep in Mind
Critical-point calculations depend on the function, the interval, and the method used to locate derivative roots. Exact polynomial differentiation is strongest when the expression is a simple polynomial. More general supported functions may require numerical derivative estimates.
This calculator works with one variable, \(x\). It supports common functions such as \(\sqrt{x}\), \(\sqrt[3]{x}\) through cbrt, \(|x|\) through abs, natural logarithms through ln or log, exponentials through exp, and the trigonometric functions \(\sin(x)\), \(\cos(x)\), and \(\tan(x)\). It also supports constants such as \(\pi\) and \(e\). Variables other than \(x\) and unsupported functions are not part of the calculation.
The exact polynomial method is limited to simple polynomial forms. Expressions with division, negative powers, fractional powers, non-integer exponents, or parentheses may be handled numerically instead of as exact polynomial derivatives.
Numerical scanning can miss critical points. This is most likely when a derivative root does not create a sign change, when the graph has a very narrow feature, when the function has difficult domain behavior, or when the derivative becomes undefined in a way that is not detected by the supported checks.
Endpoint behavior is another important limitation. A function can have an absolute maximum or minimum at an endpoint of a closed interval even when that endpoint is not a critical point in the usual interior sense. This calculator is designed to find critical points, not to complete a full absolute-extrema comparison.
Rounding can also affect how results look. Standard numeric coordinates are shown with compact formatting and up to about seven decimal places. Very small values may be displayed as \(0\), and very large or very small nonzero values may use exponential notation. Closely repeated candidates may be merged so that the same point is not listed many times.
Use the result as a strong guide for learning and checking work, especially for standard classroom examples. For a proof, a graded symbolic solution, or a high-stakes technical application, verify the result with exact derivative analysis and domain reasoning.
How to Use This Calculator
-
Enter a function using \(x\) as the variable.
-
Enter the lower bound in x min and the upper bound in x max.
-
Make sure x max is greater than x min.
-
Use an example preset if you want to load a sample cubic, quartic, or trigonometric function.
-
Review the main result for detected coordinate pairs.
-
Use the graph preview and summary table to compare point labels, coordinates, and classifications.
-
Check the derivative detail to see whether the result used an exact polynomial derivative or a numerical derivative scan.
-
Use the graph download option if you need a PNG copy of the preview.
-
Use Clear to reset the function input.
You can enter expressions with operators such as \(+\), \(-\), \(*\), \(/\), powers, and parentheses. The input also accepts common constants and supported functions, including pi, e, sqrt, cbrt, abs, ln, log, exp, sin, cos, and tan.
Frequently Asked Questions
Is a critical point the same as a local maximum or minimum?
No. A critical point is a candidate for important behavior. It may be a local maximum, a local minimum, or neither. The derivative sign on each side of the point is what helps classify it.
Can a point be critical if the derivative is undefined?
Yes, as long as the original function is defined at that input value. For example, \(f(x)=|x|\) has a critical point at \((0,0)\) because the function exists there but the derivative does not.
Why are endpoints not always listed as critical points?
Endpoints can be important for absolute maxima and minima on a closed interval, but they are not automatically critical points in the usual interior derivative-test sense. A complete absolute-extrema problem should compare endpoints with critical points.
What does “No points found” mean?
It means no critical points were detected within the selected interval by the calculator’s method. For numerical cases, this is not the same as a mathematical proof that none exist.
What does “Stationary point” mean?
A stationary point is a point where the derivative is zero or very close to zero. It may not be a maximum or minimum if the derivative does not change from positive to negative or from negative to positive.
Why do trigonometric functions use radians?
Radians are the standard angle measure used in calculus formulas for derivatives of trigonometric functions. Inputs such as \(\sin(x)\), \(\cos(x)\), and \(\tan(x)\) should be interpreted in radians.
How exact are the results?
Some polynomial results can be based on exact differentiation, but many supported functions are handled with numerical derivative estimates. Rounded decimal coordinates and graph markers are useful for interpretation, but exact symbolic work may still be needed for formal proof.
Sources and References
Books and Open Textbooks
-
Gilbert Strang and Edwin “Jed” Herman. Calculus Volume 1. OpenStax, 2016. Sections 4.3 “Maxima and Minima” and 4.5 “Derivatives and the Shape of a Graph.” OpenStax Calculus Volume 1, Section 4.3, Section 4.5.
-
Matthew Boelkins. Active Calculus: Single Variable. Active Calculus, first edition, major update August 2019. Section 3.1 “Using derivatives to identify extreme values.” Active Calculus section 3.1.
-
Qingkai Kong, Timmy Siauw, and Alexandre M. Bayen. Python Programming and Numerical Methods: A Guide for Engineers and Scientists. 1st ed., Academic Press/Elsevier, 2020. Chapters 19 “Root Finding” and 20 “Numerical Differentiation.” Elsevier book page, Python Numerical Methods finite-difference notebook.
Online Educational Sources
-
Autar Kaw and contributors. “Numerical Differentiation of Continuous Functions.” Numerical Methods with Applications, LibreTexts, accessed June 28, 2026. LibreTexts numerical differentiation.
-
Jeffrey R. Chasnov. “Bisection Method.” Numerical Methods, LibreTexts, accessed June 28, 2026. LibreTexts bisection method.
