Originally published in 2015; rebuilt in 2026. This page began as a compact Mathematics note on intercepts, stationary points and asymptotes. The replacement preserves that RFE but raises it into a durable graph-reading and graph-sketching reference from O-Level and Additional Mathematics through A-Level, IGCSE and IB contexts.
Quick Read
A graph is a visual representation of a relationship between variables. Its important features—axes, intercepts, gradients, stationary points, turning points, asymptotes and end behaviour—tell us how that relationship behaves.
One-sentence answer: to understand a graph, identify where it crosses, where it flattens, where it turns, what it approaches and how it behaves far from the origin.
The RFE of this page
This page exists to answer one durable Mathematics question: what structural features should a student inspect when reading or sketching a graph?
1. The coordinate axes
In a standard Cartesian graph, the horizontal axis is the x-axis and the vertical axis is the y-axis.
A point is written as (x, y). The first coordinate tells us the horizontal position; the second tells us the vertical position.
Before reading any graph, check:
- what each axis represents;
- the scale;
- the units;
- whether the graph shows all relevant values or only a selected window.
A visually steep curve can look very different if the axis scales change.
2. The y-intercept
The y-intercept occurs where the graph crosses the y-axis. At every point on the y-axis, x = 0.
For a function y = f(x), substitute x = 0 to find the y-intercept:
y = f(0).
Example: for y = 2x + 5, the y-intercept is (0, 5).
3. The x-intercepts or roots
The x-intercepts occur where the graph crosses or touches the x-axis. At those points, y = 0.
For y = f(x), solve:
f(x) = 0.
These x-values are also called roots or zeros of the function.
A graph can:
- cross the x-axis;
- touch the x-axis and turn;
- have several roots;
- have no real x-intercept at all.
The visual behaviour often tells us something about root multiplicity, although the exact algebra depends on the function.
4. Gradient: how fast is y changing?
The gradient measures the rate of change of y with respect to x.
For a straight line:
gradient = change in y ÷ change in x.
For a curve, the gradient changes from point to point. Calculus expresses the instantaneous gradient as dy/dx or f′(x).
- positive gradient: graph rises as x increases;
- negative gradient: graph falls as x increases;
- zero gradient: tangent is horizontal;
- undefined or infinite-slope situations require more careful treatment, such as vertical tangents.
5. Stationary points
A stationary point is a point on a differentiable curve where the gradient is zero:
dy/dx = 0.
The tangent line there is horizontal.
Common types include:
- local maximum;
- local minimum;
- stationary point of inflection.
Not every stationary point is a turning point.
6. Turning points
A turning point is where the graph changes direction from increasing to decreasing or from decreasing to increasing.
For ordinary smooth curves studied at school, a local maximum or minimum is typically both stationary and a turning point.
But the terms are not identical. A graph can have a stationary point of inflection where dy/dx = 0 yet the graph continues increasing on both sides.
Conversely, in more advanced contexts, a graph can turn at a point where the derivative is not zero or not defined.
7. Maximum and minimum values
A local maximum is higher than nearby points. A local minimum is lower than nearby points.
A global or absolute maximum/minimum is the highest or lowest value over the whole domain under consideration.
This distinction matters in optimisation questions. A local best point is not automatically the best point over the entire allowed interval.
8. Points of inflection
A point of inflection is where the curve changes concavity.
Informally, the curve changes from bending one way to bending the other.
A stationary point of inflection has both a horizontal tangent and a change of concavity. But not every point of inflection is stationary.
This is one of the most common graph-language confusions: stationary, turning and inflection describe different properties.
9. What is an asymptote?
An asymptote is a line that describes the limiting behaviour of a curve.
School Mathematics commonly studies:
- vertical asymptotes;
- horizontal asymptotes;
- oblique or slant asymptotes.
The important idea is not merely that the graph “gets close” to a line. The asymptote describes what the function tends toward under a particular limit.
10. Vertical asymptotes
A vertical asymptote has equation x = a when the function grows without bound in magnitude as x approaches a from one or both sides.
For many rational functions, vertical asymptotes arise where a denominator becomes zero after common factors have been considered.
For example:
y = 1/(x − 2)
has a vertical asymptote at x = 2.
But students should not mechanically say “denominator zero means asymptote”. A removable discontinuity can occur when a common factor cancels.
11. Horizontal asymptotes
A horizontal asymptote has equation y = L when:
f(x) → L as x → ∞ or x → −∞.
A graph can cross a horizontal asymptote. The asymptote describes end behaviour, not a barrier the graph is forbidden to touch.
That is another common misconception.
12. Oblique or slant asymptotes
An oblique asymptote is a non-horizontal straight line:
y = mx + c
that the function approaches for large positive or negative x.
For rational functions, a common school case occurs when the degree of the numerator is exactly one more than the degree of the denominator. Polynomial division can then reveal the slant asymptote.
The old 2015 note stated this case too broadly. It is a useful sufficient pattern for many rational functions studied in school, not the universal definition of an oblique asymptote.
13. Holes are not asymptotes
Consider:
y = (x² − 1)/(x − 1).
For x ≠ 1, this simplifies to y = x + 1. The original function is undefined at x = 1, so the graph has a hole at (1, 2), not a vertical asymptote.
This distinction is important when sketching rational functions.
14. End behaviour
End behaviour asks what happens as x becomes very large positive or very large negative.
For polynomials, the highest-degree term dominates end behaviour. For rational functions, comparison of polynomial degrees often reveals horizontal or slant behaviour.
Before sketching the fine detail of a curve, understand where the graph is heading overall.
15. Domain matters
A graph only represents values where the function is defined and where the problem allows x to exist.
- denominators cannot be zero;
- real square roots impose restrictions;
- logarithms require positive arguments;
- real-world questions may impose additional restrictions such as time ≥ 0.
A mathematically valid branch of a graph may still be irrelevant if it lies outside the problem’s physical domain.
16. A practical graph-sketching order
- Identify the domain.
- Find x- and y-intercepts.
- Identify symmetry if relevant.
- Find vertical, horizontal or oblique asymptotes.
- Find stationary points where applicable.
- Classify maxima, minima or stationary inflection points.
- Inspect intervals where the graph increases or decreases.
- Inspect end behaviour.
- Plot enough additional points to connect the structure correctly.
The exact order can change with the function, but this prevents random point-plotting from replacing mathematical understanding.
17. Common misconceptions
- “A stationary point is always a turning point.” False.
- “A graph cannot cross an asymptote.” False for horizontal and oblique asymptotes in general.
- “Denominator zero always gives a vertical asymptote.” Not if the discontinuity is removable after cancellation.
- “An x-intercept and a root are unrelated ideas.” For y = f(x), the x-intercepts correspond to real roots of f(x) = 0.
- “A maximum is always the highest point on the entire graph.” A local maximum may only be highest nearby.
- “dy/dx = 0 proves maximum or minimum.” It identifies a stationary candidate; classification still matters.
18. Graphs as a translation layer
Graphs translate algebra into geometry.
| Algebraic statement | Graphical meaning |
|---|---|
| f(x) = 0 | x-intercept / root |
| f(0) | y-intercept |
| f′(x) > 0 | increasing locally |
| f′(x) < 0 | decreasing locally |
| f′(x) = 0 | stationary candidate |
| f(x) → L | approaches horizontal level y = L |
| f(x) → ±∞ as x → a | vertical asymptotic behaviour at x = a |
Students who can move both directions—equation to graph and graph to equation—have much stronger mathematical control.
19. From O-Level to A-Level and IB
The same graph features reappear at higher levels with increasing analytical depth.
- O-Level / IGCSE: intercepts, gradients, quadratic graphs, basic function features;
- Additional Mathematics: differentiation, stationary points, curve sketching, exponential/logarithmic/trigonometric functions;
- A-Level / IB: limits, more complex functions, optimisation, concavity, asymptotic behaviour and modelling.
The vocabulary becomes more formal, but the structural questions remain recognisable.
20. AI and graph learning
AI and graphing software can sketch a curve instantly. That is useful for checking, but it can hide whether the learner understands why the graph has that shape.
A stronger workflow is:
- predict the important features;
- calculate intercepts and asymptotes;
- find stationary points where required;
- sketch by reasoning;
- use software or AI to compare;
- explain any mismatch.
The tool should verify the model, not replace it.
The deeper principle
A graph is compressed information about a relationship.
Intercepts tell us where values reach zero or meet an axis. Gradients tell us how the relationship changes. Stationary and turning points reveal local structure. Asymptotes and end behaviour reveal limiting structure.
The goal is not to memorise graph vocabulary separately. It is to reconstruct the behaviour of the function from its visible and algebraic evidence.
Historical note: first published 8 June 2015 as “Graphs Axis, Stationary Points and Asymtotes”. The 2026 rebuild corrects several oversimplifications, expands the mathematical resolution and preserves the original RFE as a durable graph-features reference.