Transform nonlinear functions

Lesson progressPractice problems 0/3
Difficulty
Advanced
Estimated time
30 minutes
Techniques
Function-transformationsTranslationsReflectionsStretchesPoint-mapping

What you’ll learn

  1. Spot when a question takes a graph or rule you know and moves or reshapes it.
  2. Write shifts, reflections, and vertical stretches or compressions in function notation.
  3. Find where a named point on y=f(x)y=f(x) lands on the new graph.
  4. Handle the harder inside expression f(ax+b)f(ax+b) by solving for the new input.

Why this matters on the SAT

Tell inside and outside changes apart

The SAT might describe a graph moving in words, give you a new rule like g(x)=−f(x−3)+1g(x)=-f(x-3)+1, or ask where one named point ends up. All of these come down to one idea:

  • a change inside f( )f(\ ) changes the input, the xx-coordinate
  • a change outside f( )f(\ ) changes the output, the yy-coordinate

Solution to the example

Take the moves one at a time. Shifting 33 units right replaces xx with x−3x-3. Reflecting across the xx-axis multiplies every output by −1-1. Shifting down 22 subtracts 22 outside the function. Put together, that’s

y=−(x−3)2−2,y=-(x-3)^2-2,

so the answer is A.

Choice B falls for a common sign mistake. It’s easy to read x+3x+3 as “right 33,” but it moves the graph left. The next section shows why.

How to spot these questions. A graph or rule you already know gets shifted, reflected, stretched, or compressed into a new one. Often the question defines gg in terms of ff. If no new graph comes out of the old one, nothing is being moved. If the question only asks for a function value, that’s ordinary evaluation. Finding a feature of one quadratic that isn’t being changed, like its vertex, or building a model from a situation, doesn’t move a graph either.

SAT example

The graph of y=x2y=x^2 is shifted 33 units to the right, reflected across the xx-axis, and shifted 22 units down. Which equation represents the resulting graph?

  1. A

    y=−(x−3)2−2y=-(x-3)^2-2

  2. B

    y=−(x+3)2−2y=-(x+3)^2-2

  3. C

    y=(x−3)2+2y=(x-3)^2+2

  4. D

    y=−x2−5y=-x^2-5

Track what happens to one point

A graph is made of points. Transforming it sends every one of those points to a new place, so the question is always the same: where does each point go?

Start with the labeled point in the figure. On the parent graph, the original y=f(x)y=f(x), the point P(3,2)P(3,2) means f(3)=2f(3)=2. The new graph, called the image, is

g(x)=−2f(x−3)+1.g(x)=-2f(x-3)+1.
Transform the coordinates: 3+3=63+3=6 and −2(2)+1=−3-2(2)+1=-3.

Take the two coordinates one at a time.

The xx-coordinate. Whatever goes into ff still has to be 33, because f(3)=2f(3)=2 is the fact you’re carrying over. So the new xx has to make

x−3=3,x-3=3,

which gives x=6x=6. On the graph, the point moves 33 units right.

The yy-coordinate. Start with the old output, 22, and do what’s outside ff:

−2(2)+1=−3.-2(2)+1=-3.

The negative sign reflects the output across the xx-axis, to the other side. The 22 doubles its distance from that axis, and only then does the +1+1 shift the result up 11. So P(3,2)P(3,2) lands at

(6,−3),\boxed{(6,-3)},

the point P′P' in the figure. That’s four changes in one rule: right 33, a reflection, a stretch by 22, and up 11.

The same two steps work for any point. Call the old point (u,v)(u,v): uu is the old input, vv is the old output, and f(u)=vf(u)=v. For

g(x)=Af(x−h)+k,g(x)=A f(x-h)+k,

the input inside ff still has to be uu, so x−h=ux-h=u, which gives x=u+hx=u+h. Outside, vv becomes Av+kAv+k. So the point (u,v)(u,v) lands at

(u+h, Av+k).(u+h,\ Av+k).

In short: solve the inside, then apply the outside.

How common changes move a point (u,v)(u,v)

New ruleWhere the point goesWhat you see
f(x−h)f(x-h)(u+h,v)(u+h,v)right hh when h>0h>0
f(x)+kf(x)+k(u,v+k)(u,v+k)up kk when k>0k>0
−f(x)-f(x)(u,−v)(u,-v)reflection across the xx-axis
Af(x)A f(x)(u,Av)(u,Av)vertical stretch if $
Try it yourself:

For g(x)=−2f(x−3)+1g(x)=-2f(x-3)+1, make two calls before you calculate. Does a point move left or right? And does a point with a positive original output end up above or below the line y=1y=1? Then work out the point to check yourself.

When Desmos helps: the whole new graph

For one named point, mapping by hand is quick and exact, and it shows you why the point moves. But when you know the parent’s rule and the question asks about the whole new graph, like its vertex, Desmos may be faster. Define the parent first, then define the image from it:

f(x)=(x-1)^2/2
g(x)=-2f(x-3)+1

Desmos draws the whole image, and you never have to multiply out the new rule. Select the vertex of gg and you’ll see (4,1)(4,1). The point (6,−3)(6,-3) is on the graph too, right where the hand mapping put it. So for a vertex, a minimum or maximum, an intercept, or a function value as a number, these two lines can get you the answer directly.

Try it yourself:

Change x-3 to x+2. Which way should the graph move? Predict first, then look, then reset the example.

Calculator loads as you approach
Select the vertex of the new graph. Then edit the inside expression and watch the whole image move.

Write the transformed equation

Now flip the job around: the question describes the moves in words, and you write the rule. Horizontal changes go inside ff, with the input. Vertical changes go outside, on the output:

g(x)=Af(x−h)+k.g(x)=A f(x-h)+k.

Read the pieces in this order:

  1. Left or right: x−hx-h shifts right hh, and x+hx+h shifts left hh.
  2. Reflect and stretch: a negative AA reflects the graph across the xx-axis. The size of AA, written ∣A∣|A|, sets the vertical stretch or compression.
  3. Up or down: +k+k shifts up kk, and −k-k shifts down kk.

If you know the parent’s equation, put the whole new input into it first, then apply the outside changes.

Here’s one worked through. Let

f(x)=x3+2.f(x)=x^3+2.

Shift its graph 44 units left, compress it vertically by a factor of 12\frac12, reflect it across the xx-axis, and shift it down 11. In function notation, that’s

g(x)=−12f(x+4)−1.g(x)=-\frac12 f(x+4)-1.

To write it without ff, put x+4x+4 into the whole parent rule, +2+2 included:

g(x)=−12((x+4)3+2)−1=−12(x+4)3−2.\begin{aligned} g(x) &=-\frac12\left((x+4)^3+2\right)-1\\[1.4em] &=-\frac12(x+4)^3-2. \end{aligned}

The −12-\frac12 multiplies the +2+2 too, which gives an extra −1-1. Both forms describe the same graph, but the one written with ff makes the moves easier to read.

Check your understanding:

The function pp is defined by p(x)=x+1p(x)=\sqrt{x+1}. Write a function qq whose graph is the graph of pp shifted 55 units right, stretched vertically by a factor of 33, and shifted 22 units up.

Common mistake:

Outside, the signs do what they say: +2+2 means up 22. Inside, they look backward, because the new input has to solve x−h=ux-h=u. Say you wrote f(x+3)f(x+3) for a shift right 33. Test the parent input u=0u=0. After a shift right, the new graph should show the parent’s output f(0)f(0) at x=3x=3. But f(x+3)f(x+3) puts it at x=−3x=-3, since −3+3=0-3+3=0.

Example: where does a named point land?

Worked example

The point A(−1,4)A(-1,4) lies on the graph of y=f(x)y=f(x). Which choice gives the corresponding point on the graph of

y=−3f(x−2)+5?y=-3f(x-2)+5?
  1. A

    (−3,−7)(-3,-7)

  2. B

    (1,17)(1,17)

  3. C

    (1,−7)(1,-7)

  4. D

    (3,−17)(3,-17)

Step 1

Say what the point tells you

Since A(−1,4)A(-1,4) is on y=f(x)y=f(x),

f(−1)=4.f(-1)=4.

So the old input is u=−1u=-1, and the old output is v=4v=4.

Step 2

Find the new input

Whatever goes into ff has to equal the old input, −1-1. So the new xx has to solve

x−2=−1,x-2=-1,

which gives

x=1.x=1.

The point moves 22 units right, from −1-1 to 11.

Step 3

Change the output

Start from the old output, 44, and do what’s outside ff:

−3(4)+5=−12+5=−7.-3(4)+5=-12+5=-7.

The −3-3 reflects the old output and stretches it to three times its distance from the xx-axis. Then the +5+5 shifts the graph up 55.

Step 4

Name the point and check it

The new point is

(1,−7).\boxed{(1,-7)}.

To check, put x=1x=1 into the new rule:

−3f(1−2)+5=−3f(−1)+5=−3(4)+5=−7.-3f(1-2)+5=-3f(-1)+5=-3(4)+5=-7.

The check runs the mapping backward: 1−21-2 lands right back on the old input, −1-1. The answer is C.

Harder inside changes

So far the inside has been simple, like x−3x-3. A harder SAT question might put more inside ff, like this:

g(x)=4−2f(7−3x).g(x)=4-2f(7-3x).

This looks like a big jump, but you already have the tool for it: whatever goes into ff still has to equal the old input. The only new part is a slightly longer equation.

Say the point (5,−2)(5,-2) is on y=f(x)y=f(x), so f(5)=−2f(5)=-2. The pieces are out of their usual order, so sort them first. The 7−3x7-3x is inside ff. The −2-2 multiplies the output. The 44 out front is added last, so it shifts the graph up 44.

The xx-coordinate. Set the whole inside equal to the old input, 55, and solve:

7−3x=5⟹x=23.7-3x=5 \quad\Longrightarrow\quad x=\frac23.

The yy-coordinate. Start from the old output, −2-2, and do what’s outside ff:

4−2(−2)=8.4-2(-2)=8.

So (5,−2)(5,-2) lands at

(23,8).\boxed{\left(\frac23,8\right)}.

Notice that you never had to work out how far the graph shifted or which way it flipped. Solving the inside took care of all of it.

If a question asks what the graph itself does, the inside 7−3x7-3x tells you that too:

  • The 33 squeezes horizontal distances to 13\frac13 of their size.
  • The minus sign swaps left and right. That’s a reflection across the yy-axis, and it happens before the graph shifts.
  • The parent’s input 00 lands where the inside is 00. Solving 7−3x=07-3x=0 gives x=73x=\frac73.

The same steps work for any inside of the form ax+bax+b, as long as a≠0a\ne0. Written in the usual order, the rule is g(x)=Af(ax+b)+kg(x)=A f(ax+b)+k. Our example is g(x)=−2f(−3x+7)+4g(x)=-2f(-3x+7)+4, so A=−2A=-2, a=−3a=-3, b=7b=7, and k=4k=4.

Take any point (u,v)(u,v) on y=f(x)y=f(x). You solve ax+b=uax+b=u, just as you solved 7−3x=57-3x=5, and get x=u−bax=\frac{u-b}{a}. The output still becomes Av+kAv+k. So the point lands at

(u−ba, Av+k).\left(\frac{u-b}{a},\ Av+k\right).

The graph facts carry over the same way. Horizontal distances get multiplied by 1∣a∣\frac1{|a|}, like the 13\frac13 in the example. A negative aa swaps left and right. And the parent’s input 00 lands where ax+b=0ax+b=0, which is at x=−bax=-\frac ba.

Check your understanding:

If (8,3)(8,3) lies on y=f(x)y=f(x) and g(x)=1+4f(2x−6)g(x)=1+4f(2x-6), what point on y=g(x)y=g(x) corresponds to it?

Common mistake:

With 7−3x7-3x inside, it’s tempting to call it a shift of 77, or to divide the old coordinate by 33 right away. Instead, write one equation, 7−3x=u7-3x=u, with uu as the old input, and solve it. It handles the reversal, the scaling, and the shift in the right order.

By hand or with Desmos?

Look at what the question asks for. That tells you which way to go.

Use the point map by hand when…

  • the question gives one point, like a vertex, an intercept, or f(−1)=4f(-1)=4, and asks exactly where it lands.

  • you don’t know the rule for ff, as when you’re told only that A(−1,4)A(-1,4) is on its graph, so there’s nothing to enter.

  • the question describes the moves in words, like “shift left 44 and up 33,” and you have to write the exact equation.

  • the inside is something like 7−3x7-3x, and one short equation gives you the new input.

Let Desmos lead when…

  • you know the rule for ff, and the question asks for a number from the new graph, like its vertex, a minimum or maximum, an intercept, or a function value.

  • the parent’s rule is written out long, like x2−8x+20x^2-8x+20, so the point you need is hard to see until you graph the new function.

  • the answer choices are whole new rules, and you can compare them quickly by seeing which graphs overlap.

Either way, answer exactly what the question asks for. In Desmos, select or evaluate it on the new graph, not the parent. If you get a decimal, like 3.53.5, and the choices are exact, convert it to 72\frac72.

You won’t need regression or Log Mode for these. A transformation changes a rule or graph you already know. It doesn’t build a new model from data.

Practice problems

Three problems to try on your own: write an equation, map a point, then find a vertex through a harder inside change. The last one is the toughest, so take your time with it. Before each one, look at what it asks for and decide: by hand or Desmos?

Write a combined transformation

Practice problem

The function ff is defined by

f(x)=x3−2.f(x)=x^3-2.

The graph of y=g(x)y=g(x) is obtained by shifting the graph of y=f(x)y=f(x) 44 units left, reflecting it across the xx-axis, vertically compressing it by a factor of 12\frac12, and shifting it 33 units up. Which equation defines gg?

Answer choices
Calculator loads as you approach
Use Desmos if comparing the parent and new graphs helps.

Map a point through both coordinates

Practice problem

The point P(−2,5)P(-2,5) lies on the graph of y=f(x)y=f(x). Which choice gives the corresponding point on the graph of

y=3f(x+4)−2?y=3f(x+4)-2?
Answer choices
Calculator loads as you approach
There’s no rule for ff to graph here, so map the point by hand.

Find a transformed quadratic vertex

Practice problem

Let

f(x)=x2−8x+20.f(x)=x^2-8x+20.

A second function is defined by

g(x)=5−3f(11−2x).g(x)=5-3f(11-2x).

Which choice gives the coordinates of the vertex of the graph of y=g(x)y=g(x)?

Answer choices
Calculator loads as you approach
Define f and g, select the vertex of g, and match its decimals to an exact choice.

Finish the lesson

3 practice examples left

Finish the remaining questions correctly to complete this lesson.

Quick recap

  • Solve the inside, then apply the outside. In g(x)=Af(x−h)+kg(x)=A f(x-h)+k, that sends a point (u,v)(u,v) to (u+h,Av+k)(u+h,Av+k).
  • Inside signs look backward because the new input has to make x−h=ux-h=u.
  • A negative outside factor reflects the graph across the xx-axis. Its size, ∣A∣|A|, gives the vertical stretch or compression.
  • For g(x)=Af(ax+b)+kg(x)=A f(ax+b)+k, set the whole inside equal to the old input, ax+b=uax+b=u, and solve instead of guessing the horizontal move.
  • For one named point, mapping by hand is usually the quickest exact way. When you know the parent’s rule and need a number from the whole new graph, defining ff and gg in Desmos can get you the answer directly.

Related lessons

That’s the end of SAT Nonlinear Functions and Models. Go back to the course page to review any lesson or pick targeted practice.

Next lesson

Reason with ratios and proportions

Keep ratio order straight, connect part-to-part and part-to-whole comparisons, and use equivalent ratios to solve scale relationships.

Start next lesson

Practice

Practice this lesson

222 SAT questions use what this lesson teaches. Practice a few in a study session at the difficulty you choose.

Start practice