Toss a ball straight up to a friend on a balcony and it rises quickly, slows down, stops for an instant at the top, then falls faster and faster. A motion sensor on the floor can record the ball's height every 0.2 seconds. When you plot those heights against time, the points do not lie on a straight line: they form a smooth arch that is the same on the way up as on the way down. That arch is called a parabola, and relations whose graphs are parabolas are called quadratic relations. They show up whenever something speeds up or slows down at a steady rate: thrown and kicked balls, water from a fountain, the braking distance of a car compared with its speed, and the area of a square compared with its side length. Here is one set of measured data.
| Time t (s) | 0 | 0.2 | 0.4 | 0.6 | 0.8 | 1.0 | 1.2 | 1.4 | 1.6 |
|---|---|---|---|---|---|---|---|---|---|
| Height h (m) | 1.0 | 2.5 | 3.4 | 3.9 | 4.2 | 4.1 | 3.4 | 2.3 | 1.0 |
To graph data like this, put the independent variable (time, the one you control) on the horizontal axis and the dependent variable (height) on the vertical axis, then plot each pair. Real measurements contain small errors, so do not join the dots with a ruler. Instead draw a curve of best fit: one smooth curve that follows the overall pattern, passing through or close to as many points as possible, with points scattered evenly on both sides. Here the curve rises from m, peaks at about m near s, and comes back down to m at s. Graphing software can do this for you with quadratic regression. For this data it gives an equation very close to . The rest of this lesson explains how to tell that a relation is quadratic and how to describe its graph.

What makes a relation quadratic
A quadratic relation is one that can be written in the form , where , and are numbers and . The highest power of is 2, so it is called a second-degree relation. The condition matters: if , the term disappears and is linear. Examples of quadratic relations: , (here ), and , because expanding gives . Not quadratic: (no term) and (the highest power is 3). Every quadratic relation has a parabola as its graph.
Finite differences
You can spot a quadratic relation in a table of values without graphing. First make sure the -values go up by equal steps. Then find the first differences: subtract each -value from the one after it (later minus earlier). For a linear relation, the first differences are all the same. For a quadratic relation they are not, but if you subtract each first difference from the next one, you get the second differences, and those are all the same. Here is the table for , where each first difference compares one row with the row below it.
| x | y = x² − 4x | First difference | Second difference |
|---|---|---|---|
| −1 | 5 | 0 − 5 = −5 | −3 − (−5) = 2 |
| 0 | 0 | −3 − 0 = −3 | −1 − (−3) = 2 |
| 1 | −3 | −4 − (−3) = −1 | 1 − (−1) = 2 |
| 2 | −4 | −3 − (−4) = 1 | 3 − 1 = 2 |
| 3 | −3 | 0 − (−3) = 3 | 5 − 3 = 2 |
| 4 | 0 | 5 − 0 = 5 | |
| 5 | 5 |
The parts of a parabola
- Vertex: the turning point of the parabola, the lowest point if it opens up or the highest point if it opens down. In the figure it is .
- Axis of symmetry: the vertical line through the vertex. Fold the graph along it and the two halves match. Name it with an equation: here .
- Zeros (also called x-intercepts): the -values where the parabola crosses or touches the -axis, so . Here they are and . A parabola can have two zeros, one zero (the vertex sits on the -axis) or no zeros (it never reaches the -axis).
- y-intercept: the -value where the parabola crosses the -axis, found by setting . In it is , because . Here it is . A parabola of this form has exactly one -intercept.
- Minimum or maximum value: the -coordinate of the vertex. A parabola that opens up has a minimum value (here ); one that opens down has a maximum value.
Symmetry is the most useful property of all. Any two points on a parabola at the same height are mirror images across the axis, so the axis lies exactly halfway between them. In the table for , the value appears at and at , so the axis is . The vertex is on the axis, so it is , which matches the lowest -value in the table. In the figure, the -intercept has a partner on the other side: both are 3 units from the axis .
- Second differences positive
- Vertex is the lowest point
- Has a minimum value
- Example: , minimum
- Second differences negative
- Vertex is the highest point
- Has a maximum value
- Example: , maximum