Exploring Algebraic Identities
Why This Matters
Try a small game. Pick any three “square numbers” that come one after another. The squares of 1, 2, 3 are 1, 4, 9. Now do this: add the smallest and the largest, then take away twice the middle one.
So 1 + 9 = 10. Then 10 − (2 × 4) = 10 − 8 = 2.
Try another set. The squares of 5, 6, 7 are 25, 36, 49. So (25 + 49) − (2 × 36) = 74 − 72 = 2 again!
It is always 2. Every single time. That feels like magic. But it is not magic — it is algebra. By the end of this chapter you will be able to prove why the answer must always be 2, in two short lines.
This chapter is about algebraic identities — a small set of ready-made rules. They do two big jobs for you. They let you square or multiply numbers in your head, like finding 43² without long multiplication. And they let you factorise — break a big expression into smaller multiplied pieces. Both skills show up again and again, in this class and in higher classes too.
The Big Idea
You already know what an equation is. Something like x² − 1 = 24 is an equation. It is only true for some values of x — here, only x = 5 or x = −5. Put in any other number and it breaks.
An identity is special. It is an equation that stays true for every value you put in.
The Big Idea: An identity is an equation that is true for all values of the variables. For example, (a + b)² = a² + 2ab + b² holds no matter what a and b are — positive, negative, fractions, anything. So we can trust it like a tool and use it again and again.
That word “all” is the whole point. Because an identity never fails, you can swap one side for the other any time you like. That swap is what makes calculations and factorising fast.
Is 3x + 6 = 12 an equation or an identity?
It is an equation. It is only true for one value, x = 2. Put x = 5 and you get 21 = 12, which is false. An identity would stay true for every x.
Seeing (a + b)² as a Picture
Let us start with the most useful identity of all and prove it with a picture.
This identity uses areas of squares and rectangles, so here is a quick reminder of those formulas first.
Imagine two short sticks. One has length a, the other has length b. Join them end to end and you get one stick of length (a + b).
Now build a square whose side is this full length, (a + b). Mark the point where a ends and b begins on each side, and draw lines across. The square breaks neatly into four pieces. Figure 4.1 shows them.
Look at what we have. The big square’s area is (a + b)² — that is just side × side. But it is also made of four pieces:
- one square of area a²,
- one square of area b²,
- two rectangles, each of area ab.
Add them: a² + b² + ab + ab = a² + 2ab + b². The pieces must add up to the whole. So we have proved, just by looking, that
(a + b)² = a² + 2ab + b²
This picture only works when a and b are real lengths, so they are positive. But the identity is true for all numbers, even negative ones. Let us check one negative case, then prove it with algebra so we are fully sure.
Check with negatives. Let a = −2 and b = −3. Then (a + b) = −5, so (a + b)² = 25. On the other side: a² = 4, b² = 9, and 2ab = 2 × (−2) × (−3) = 12. Adding, 4 + 12 + 9 = 25. It matches!
The algebra proof (works for every number). We use the distributive property — “multiply each part by each part”:
(a + b)² = (a + b)(a + b)
= a(a + b) + b(a + b)
= a² + ab + ba + b²
= a² + 2ab + b²
This proof never assumed anything about a and b. So it holds for all numbers, and (a + b)² = a² + 2ab + b² is a true identity.
Which is bigger, (a + b)² or a² + b²?
A very common slip is to think (a + b)² is a² + b². It is not. The difference is exactly that extra 2ab.
Take a = 10 and b = 2. Then (a + b)² = 12² = 144, but a² + b² = 100 + 4 = 104. The gap is 40, which is exactly 2ab = 2 × 10 × 2 = 40. When a and b are both positive, 2ab is positive, so (a + b)² is bigger.
When is (a + b)² equal to a² + b²?
Only when the extra term 2ab = 0, which means a = 0 or b = 0. If one of the numbers is zero, the two middle rectangles vanish and the square has no extra area.
Using (a + b)² as a tool
Once we trust the identity, we can use it two ways: to expand expressions, and to square numbers fast.
Here is the identity expanding a binomial — a two-term expression — squared.
Expand (5x + 2y)² using (a + b)² = a² + 2ab + b².
- Match the parts. Compare (5x + 2y)² with (a + b)². So a = 5x and b = 2y.
- Write the identity with these parts: (5x + 2y)² = (5x)² + 2(5x)(2y) + (2y)².
- Work out each piece. (5x)² = 25x². The middle term 2(5x)(2y) = 20xy. And (2y)² = 4y².
- So (5x + 2y)² = 25x² + 20xy + 4y².
Now the number trick. To square a number, split it into an easy round part plus a small part.
Find 43² without long multiplication.
- Split 43 into a round number plus a small number: 43 = 40 + 3. So we want (40 + 3)².
- Use (a + b)² = a² + 2ab + b² with a = 40 and b = 3.
- That is 40² + 2 × 40 × 3 + 3² = 1600 + 240 + 9.
- Add them up: 1600 + 240 + 9 = 1849. So 43² = 1849.
Using Identities to Factorise
Factorising means writing an expression as things multiplied together. It is the reverse of expanding. The same identity, read backwards, becomes a factorising tool.
If you ever see an expression shaped like a² + 2ab + b², you can fold it straight back into (a + b)².
Let us factorise x² + 4x + 4. We hunt for the pattern. Notice x² = (x)², and 4 = 2², and the middle term 4x = 2 × (x) × (2). So it matches a² + 2ab + b² with a = x and b = 2. Therefore
x² + 4x + 4 = (x + 2)²
So (x + 2) is a factor of x² + 4x + 4.
Sometimes you first need to pull out a common factor. Here is one of those.
Factorise 50p² + 60pq + 18q².
- Look for a number that divides all three terms. 50, 60 and 18 are all even, so 2 is a common factor. Take it out: 50p² + 60pq + 18q² = 2(25p² + 30pq + 9q²).
- Now focus on the bracket 25p² + 30pq + 9q². Find the squares: 25p² = (5p)² and 9q² = (3q)².
- Check the middle term fits 2ab. With a = 5p and b = 3q, 2ab = 2 × 5p × 3q = 30pq. Yes, it matches.
- So 25p² + 30pq + 9q² = (5p + 3q)². Putting back the 2, the full answer is 50p² + 60pq + 18q² = 2(5p + 3q)².
The (a − b)² Identity
What if there is a minus sign? Just replace b with −b in the identity you already have. Since (−b)² = b² and 2a(−b) = −2ab, you get:
(a − b)² = a² − 2ab + b²
Only the middle sign flips. This is also a true identity, and it works the same way for expanding, squaring numbers, and factorising.
We can prove it with a picture too. This time draw a square of side a. Then mark a length b in from one corner, which leaves a length (a − b). The lines cut the square into pieces, shown in Figure 4.2.
Reading the figure: the big square has area a². To find the green corner square, which is (a − b)², we subtract the two strips. The bottom strip has area ab. The right strip has area b(a − b). So:
(a − b)² = a² − ab − b(a − b)
= a² − ab − ba + b²
= a² − 2ab + b²
The picture and the algebra agree.
The number trick now uses a minus. It is best when a number is just below a round value.
Find 29² using (a - b)² = a² - 2ab + b².
- 29 is one less than 30. So write 29 = 30 − 1, and we want (30 − 1)².
- Use (a − b)² with a = 30 and b = 1: 30² − 2 × 30 × 1 + 1².
- That is 900 − 60 + 1.
- So 900 − 60 + 1 = 841. Hence 29² = 841.
Proving the “always 2” magic trick
Remember the trick from the start? Now we can prove it in two lines. Any three numbers in a row can be written as (n − 1), n, and (n + 1). Their squares are (n − 1)², n², and (n + 1)².
Add the smallest and largest squares, using our two identities:
(n − 1)² + (n + 1)² = (n² − 2n + 1) + (n² + 2n + 1) = 2n² + 2
Now subtract twice the middle square, which is 2n²:
(2n² + 2) − 2n² = 2
The n² and the −2n and +2n all cancel out, leaving just 2. That is why you always get 2, no matter which three consecutive squares you pick. Magic explained.
The Difference of Two Squares
Here is another identity you met earlier, and it is a real favourite in exams:
(a + b)(a − b) = a² − b²
You can check it by multiplying out: (a + b)(a − b) = a² − ab + ba − b² = a² − b². The middle terms cancel, leaving a² − b².
Written the other way round, a² = (a + b)(a − b) + b², this becomes a clever shortcut for squaring numbers. Figure 4.3 shows why this rearrangement works as areas.
A famous Indian mathematician, Śhrīdharāchārya (around 750 CE), used exactly this idea to square numbers fast.
Find 55² using a² = (a + b)(a - b) + b².
- Pick b so that a + b and a − b are easy round numbers. With a = 55 and b = 5, we get a + b = 60 and a − b = 50.
- Use 55² = (55 + 5)(55 − 5) + 5² = 60 × 50 + 25.
- 60 × 50 = 3000, and 5² = 25.
- So 55² = 3000 + 25 = 3025.
The (a + b + c)² Identity
What if we square the sum of three numbers? We can work it out using a trick: treat b + c as one single block, call it d.
We already know (a + d)² = a² + 2ad + d². Now put d = b + c back in. After expanding 2a(b + c) and (b + c)², and tidying up, we get:
(a + b + c)² = a² + b² + c² + 2ab + 2bc + 2ca
In words: square each of the three terms, then add twice each pair product. There are three pairs — ab, bc, ca — so three “twice” terms.
This has a beautiful picture too. Draw a square of side (a + b + c), and split each side into a, b, and c. You get a 3-by-3 grid of nine pieces, in Figure 4.4.
See how the diagonal gives you a², b², c², and each off-diagonal area shows up twice? That is exactly where the 2ab, 2bc and 2ca come from. The picture shows the doubling instead of just stating it.
Find 119² using (a + b + c)² = a² + b² + c² + 2ab + 2bc + 2ca.
- Split 119 into easy parts: 119 = 100 + 10 + 9. So a = 100, b = 10, c = 9.
- Squares: 100² = 10000, 10² = 100, 9² = 81.
- Twice the pairs: 2(100)(10) = 2000, 2(100)(9) = 1800, 2(10)(9) = 180.
- Add everything: 10000 + 100 + 81 + 2000 + 1800 + 180 = 14161. So 119² = 14161.
Multiplying (x + a)(x + b)
Now a different shape. Multiply two brackets that both start with x. Using the distributive property:
(x + a)(x + b) = x² + bx + ax + ab = x² + (a + b)x + ab
So the middle coefficient is the sum a + b, and the last term is the product ab. We can see this with algebra tiles — little squares and strips that stand for x², x, and 1.
Take the example (x + 3)(x + 4). Lay it out as a rectangle with sides (x + 3) and (x + 4), shown in Figure 4.5.
The whole rectangle is x² + 7x + 12. Its sides are (x + 3) and (x + 4). Notice the 7 in 7x is split as 3 + 4 — exactly our a + b — and the 12 is 3 × 4 = ab. The tiles make the rule visible.
Factorising x² + (sum)x + (product) by splitting the middle term
To factorise without tiles, just reverse the rule. For x² + bx + c, find two numbers that add to b and multiply to c.
Factorise x² + 11x + 30.
- We want x² + 11x + 30 = x² + (a + b)x + ab. So we need a + b = 11 and ab = 30.
- List pairs that multiply to 30: 1 and 30, 2 and 15, 3 and 10, 5 and 6. Check which pair adds to 11.
- 5 + 6 = 11 and 5 × 6 = 30. That is the pair.
- So x² + 11x + 30 = (x + 5)(x + 6).
Watch the signs carefully when the numbers are negative.
Factorise x² - 5x + 6.
- Compare with x² + (a + b)x + ab. So a + b = −5 (negative!) and ab = 6 (positive).
- The product is positive but the sum is negative. That can only happen if both numbers are negative.
- Try −2 and −3: their sum is −5 and their product is (−2)(−3) = 6. Both conditions hold.
- So x² − 5x + 6 = (x − 2)(x − 3).
The Cube Identities
Squares grew into nice patterns. Cubes do too. What is (a + b)³? Use the distributive property, with the square we already know:
(a + b)³ = (a + b)(a + b)² = (a + b)(a² + 2ab + b²) = a³ + 3a²b + 3ab² + b³
This is a brand-new identity. And just as (a + b)² came from cutting a square, (a + b)³ comes from cutting a cube of edge (a + b). Figure 4.6 shows the eight pieces inside.
The volume of the whole cube is (a + b)³. Inside, you can count: one cube of volume a³, one of volume b³, three cuboids each a²b, and three each ab². Add them and you get a³ + 3a²b + 3ab² + b³. The volumes prove the identity, exactly the way the areas did before.
For a minus sign, replace b with −b. The signs alternate (plus, minus, plus, minus):
(a − b)³ = a³ − 3a²b + 3ab² − b³
These cube identities let you spot the side of a cube from its volume.
A cube has volume p³ + 6p²q + 12pq² + 8q³ cubic units. Find the length of its edge.
- This should match (a + b)³ = a³ + 3a²b + 3ab² + b³. Find a and b.
- The first term p³ = (p)³, so a = p. The last term 8q³ = (2q)³, so b = 2q.
- Check the middle terms: 3a²b = 3(p²)(2q) = 6p²q ✓, and 3ab² = 3(p)(4q²) = 12pq² ✓. Both fit.
- So the volume is (p + 2q)³, which means the edge of the cube is (p + 2q) units.
Sum and difference of two cubes
Multiply (x − y)(x² + xy + y²) using the distributive property and almost everything cancels:
(x − y)(x² + xy + y²) = x³ + x²y + xy² − x²y − xy² − y³ = x³ − y³
And similarly:
x³ + y³ = (x + y)(x² − xy + y²)
These two are gold for factorising any difference or sum of cubes. (Notice the sign rule: the middle term of the long bracket is minus for the plus cube identity and plus for the minus one.)
Factorise 27b³ - 1.
- Write each part as a cube. 27b³ = (3b)³ and 1 = 1³. So this is a difference of cubes with x = 3b and y = 1.
- Use x³ − y³ = (x − y)(x² + xy + y²).
- Here x² = (3b)² = 9b², xy = (3b)(1) = 3b, and y² = 1.
- So 27b³ − 1 = (3b − 1)(9b² + 3b + 1).
One more identity: a³ + b³ + c³ − 3abc
Multiplying (x + y + z)(x² + y² + z² − xy − yz − zx) and simplifying gives a neat result used in many tricky problems:
x³ + y³ + z³ − 3xyz = (x + y + z)(x² + y² + z² − xy − yz − zx)
A handy fact falls straight out of it: if x + y + z = 0, then the right side is 0, so x³ + y³ + z³ = 3xyz.
Simplifying Rational Expressions
A rational expression is just one algebraic expression divided by another — a fraction with algebra on top and bottom. The trick to simplify it is the same as with number fractions: factorise top and bottom, then cancel the common factor.
Simplify (x² - 7x + 12) / (5x² + 5x - 100), given the denominator is not zero.
- Factorise the top, x² − 7x + 12. Need two numbers with sum −7 and product 12: they are −3 and −4. So the top is (x − 3)(x − 4).
- Factorise the bottom, 5x² + 5x − 100. First take out 5: 5(x² + x − 20).
- Factorise x² + x − 20. Need sum 1 and product −20: they are 5 and −4. So x² + x − 20 = (x − 4)(x + 5), and the bottom is 5(x − 4)(x + 5).
- Now cancel the common bracket (x − 4): the fraction becomes (x − 3)(x − 4) divided by 5(x − 4)(x + 5) = (x − 3) / [5(x + 5)].
Common Mistakes
Identities are easy to use but also easy to misuse. Here are the slips that trip students most.
(a + b)² = a² + b²
The square sign sits right outside the bracket, so it feels natural to 'send it in' to each term separately, just like you spread a multiplication over a sum.
Squaring a bracket means multiplying it by itself: (a + b)(a + b). That gives the extra middle term, so (a + b)² = a² + 2ab + b², not a² + b².
(a − b)² = a² − b²
A minus inside the bracket makes you expect a minus in the answer, and a² − b² looks like a clean, balanced result.
(a − b)² = a² − 2ab + b². The last term is +b² because (−b)² is positive. The expression a² − b² is a totally different identity — it equals (a + b)(a − b).
When factorising x² − 5x + 6, the two numbers are 2 and 3 because they multiply to 6.
You correctly spot that the numbers must multiply to 6, and 2 and 3 are the obvious pair. The sign is easy to forget once the product checks out.
They must also add to the middle coefficient, −5. So you need −2 and −3: they multiply to +6 and add to −5. The factors are (x − 2)(x − 3).
(a + b)³ = a³ + b³
It looks like the square pattern, and dropping the middle terms feels tidy and symmetrical.
(a + b)³ = a³ + 3a²b + 3ab² + b³. There are two middle terms, 3a²b and 3ab². The expression a³ + b³ is a separate identity that factorises as (a + b)(a² − ab + b²).
You can cancel x in (x² − 7x + 12)/(5x² + 5x − 100) because x appears on top and bottom.
In number fractions you cancel common factors, so it is tempting to cancel any letter that shows up in both the top and the bottom.
You can only cancel a whole factor that multiplies the entire top and the entire bottom. Here x is only part of each term, not a factor of the whole. You must factorise first, then cancel the common bracket (x − 4).
Quick Check
Test yourself before the practice set.
Which of these is an identity (true for all values)?
In the square diagram for (a + b)², what do the two equal rectangles together represent?
Factorise x² + 8x + 15.
Practice Problems
Easy
Expand (7x + 4y)² using a suitable identity.
Use (a + b)² = a² + 2ab + b² with a = 7x and b = 4y.
(7x)² + 2(7x)(4y) + (4y)² = 49x² + 56xy + 16y².
Answer: 49x² + 56xy + 16y².
Find 105² using an identity.
Write 105 = 100 + 5 and use (a + b)² with a = 100, b = 5.
100² + 2 × 100 × 5 + 5² = 10000 + 1000 + 25 = 11025.
Factorise 9x² + 24xy + 16y².
Find the squares: 9x² = (3x)² and 16y² = (4y)². Check the middle term with a = 3x, b = 4y: 2ab = 2(3x)(4y) = 24xy ✓.
So 9x² + 24xy + 16y² = (3x + 4y)².
Medium
Find the product (2s + 7)(2s − 7) using an identity.
This is the difference-of-squares pattern (a + b)(a − b) = a² − b², with a = 2s and b = 7.
(2s)² − 7² = 4s² − 49.
Factorise 16y² − 24y + 9.
The middle sign is minus, so try (a − b)² = a² − 2ab + b². Find squares: 16y² = (4y)² and 9 = 3².
Check the middle: with a = 4y, b = 3, 2ab = 2(4y)(3) = 24y ✓.
So 16y² − 24y + 9 = (4y − 3)².
Expand (3x − 2y + 4z)² using the three-term identity.
Use (a + b + c)² = a² + b² + c² + 2ab + 2bc + 2ca with a = 3x, b = −2y, c = 4z.
Squares: (3x)² = 9x², (−2y)² = 4y², (4z)² = 16z².
Pair terms: 2(3x)(−2y) = −12xy, 2(−2y)(4z) = −16yz, 2(4z)(3x) = 24zx.
Answer: 9x² + 4y² + 16z² − 12xy − 16yz + 24zx.
Factorise 64y³ + 125z³.
This is a sum of cubes: 64y³ = (4y)³ and 125z³ = (5z)³.
Use the identity p³ + q³ = (p + q)(p² − pq + q²) with p = 4y and q = 5z.
(4y + 5z)[(4y)² − (4y)(5z) + (5z)²] = (4y + 5z)(16y² − 20yz + 25z²).
Challenge
If a + b + c = 5 and ab + bc + ca = 10, prove that a³ + b³ + c³ − 3abc = −25.
There is a known form of the cube identity:
a³ + b³ + c³ − 3abc = (a + b + c)(a² + b² + c² − ab − bc − ca).
We need a² + b² + c². Use (a + b + c)² = a² + b² + c² + 2(ab + bc + ca).
So 5² = a² + b² + c² + 2(10), giving 25 = a² + b² + c² + 20, so a² + b² + c² = 5.
Now substitute: a³ + b³ + c³ − 3abc = (5)(5 − 10) = (5)(−5) = −25. Proved.
By factorising, show that n³ − n is always divisible by 6 for every natural number n.
Factorise: n³ − n = n(n² − 1) = n(n − 1)(n + 1) = (n − 1) · n · (n + 1).
So n³ − n is the product of three consecutive natural numbers.
Among any three consecutive numbers, at least one is even (so the product is divisible by 2) and exactly one is a multiple of 3 (so it is divisible by 3).
Divisible by both 2 and 3 means divisible by 6. Hence n³ − n is always divisible by 6.
A rectangular pool has area 2x² + 7x + 3 square hastas. If its width is (2x + 1) hastas, find its length.
Length = Area ÷ width, so we factorise 2x² + 7x + 3 and pull out the (2x + 1) factor.
Split the middle term 7x. We need two numbers that multiply to 2 × 3 = 6 and add to 7: they are 6 and 1.
2x² + 7x + 3 = 2x² + 6x + x + 3 = 2x(x + 3) + 1(x + 3) = (2x + 1)(x + 3).
Since the width is (2x + 1), the length is (x + 3) hastas.
Summary
You can now explain and use the following:
- An identity is an equation that is true for all values of its variables, unlike an ordinary equation which holds only for special values.
- The square identities (a + b)² = a² + 2ab + b² and (a − b)² = a² − 2ab + b², including their area proofs by cutting a square into pieces.
- The difference of squares (a + b)(a − b) = a² − b², and the three-term square (a + b + c)² = a² + b² + c² + 2ab + 2bc + 2ca with its grid picture.
- Multiplying (x + a)(x + b) = x² + (a + b)x + ab, and using it backwards to factorise by splitting the middle term.
- The cube identities (a + b)³ and (a − b)³ (with the volume proof from a cube), plus x³ + y³, x³ − y³, and x³ + y³ + z³ − 3xyz.
- Using identities to calculate fast (like 43², 29², 55², 119²) and to simplify rational expressions by factorising and cancelling common factors.
What’s Next
Next we leave algebra for a while and turn to a shape that is everywhere — the circle. You will study its parts, the special angles inside it, and the rules that make wheels, clocks and arches work. Head to Chapter 5 — I’m Up and Down, and Round and Round (Circles).
Frequently Asked Questions
What is the difference between an equation and an identity?
An equation is true only for some special values of the variable. For example x² - 1 = 24 is true only when x = 5 or x = -5. An identity is an equation that is true for every value of the variable. For example (a + b)² = a² + 2ab + b² works for all a and b, so it is an identity.
Why is (a + b)² equal to a² + 2ab + b² and not a² + b²?
When you multiply (a + b)(a + b) you get four products: a times a, a times b, b times a, and b times b. That gives a² + ab + ba + b², which is a² + 2ab + b². The two middle terms ab and ba are the part students forget, so (a + b)² is always 2ab bigger than a² + b² when a and b are both positive.
How do you prove (a + b)² = a² + 2ab + b² using area?
Draw a square whose side is (a + b). Split each side into a part of length a and a part of length b. This cuts the big square into four pieces: a square of area a², a square of area b², and two equal rectangles each of area ab. Adding the pieces gives a² + 2ab + b², which must equal the whole square (a + b)².
How do I use identities to find squares like 43² quickly?
Break the number into an easy sum or difference. Write 43² as (40 + 3)² and use (a + b)² = a² + 2ab + b². That is 1600 + 240 + 9 = 1849. For numbers near a round value, such as 29², use (30 - 1)² with the (a - b)² identity.
How do you factorise x² + 7x + 12 without algebra tiles?
Compare it with x² + (a + b)x + ab. You need two numbers that add to 7 and multiply to 12. Those numbers are 3 and 4. So x² + 7x + 12 = (x + 3)(x + 4). This 'splitting the middle term' trick works for any expression of the form x² + (sum)x + (product).
What is the identity for (a + b)³ and how is it shown with a cube?
(a + b)³ = a³ + 3a²b + 3ab² + b³. To see it, take a cube of edge (a + b) and split it. You get one small cube of volume a³, one of volume b³, three cuboids each of volume a²b, and three cuboids each of volume ab². Adding all eight pieces gives a³ + 3a²b + 3ab² + b³.