In short
The binomial theorem states that for any positive integer n, (a + b)^n = \displaystyle\sum_{r=0}^{n} \binom{n}{r}\, a^{n-r}\, b^r. Each term has the form \binom{n}{r}\, a^{n-r}\, b^r, where \binom{n}{r} is the binomial coefficient from combinations. The coefficients form the rows of Pascal's triangle, and the theorem turns the tedious task of multiplying n brackets into a single formula.
Suppose you are computing compound interest. A principal of P rupees grows to P(1 + r)^n after n years at rate r. For n = 2 or n = 3, you can expand (1 + r)^n by hand using algebraic identities. But what about n = 10? Or n = 20? Multiplying twenty brackets of (1 + r) is not a task anyone wants to do manually.
The binomial theorem gives you a direct formula for the expansion of (a + b)^n for any positive integer n. No repeated multiplication, no guesswork about coefficients. Each coefficient turns out to be a combination \binom{n}{r} — the same numbers that count selections of r objects from n.
Why are combinations hiding inside an algebra problem?
Start with something small. Expand (a + b)^3 by writing it as three factors:
When you multiply these out, every term in the final expansion comes from choosing either a or b from each of the three brackets and multiplying those choices together. For example:
- Choosing a from all three brackets gives a \cdot a \cdot a = a^3.
- Choosing b from the first bracket and a from the other two gives b \cdot a \cdot a = a^2b.
- Choosing b from the first and second brackets and a from the third gives b \cdot b \cdot a = ab^2.
Each term has the form a^{3-r}b^r, where r is the number of brackets from which you chose b. The coefficient of a^{3-r}b^r is the number of ways to choose which r of the 3 brackets contribute a b — that is \binom{3}{r}.
This pattern works for every positive integer n. Each term of (a+b)^n comes from a binary choice across n brackets, and the coefficient counts those choices.
The theorem
Binomial Theorem (positive integer exponent)
For every positive integer n and any a, b:
Written out:
There are n + 1 terms. In each term, the exponents of a and b add up to n.
The expansion of (a + b)^4, for example, has 5 terms:
with coefficients \binom{4}{0}, \binom{4}{1}, \binom{4}{2}, \binom{4}{3}, \binom{4}{4} = 1, 4, 6, 4, 1.
Proof by induction
The combinatorial argument above (choosing a or b from each bracket) is convincing, but a formal proof by mathematical induction makes it watertight.
Base case. For n = 1: (a + b)^1 = a + b = \binom{1}{0}a + \binom{1}{1}b. The formula holds.
Inductive step. Assume the theorem holds for some positive integer k:
Multiply both sides by (a + b):
In the second sum, substitute s = r + 1 (so r = s - 1, and as r runs from 0 to k, s runs from 1 to k + 1):
The r = 0 term of the first sum gives \binom{k}{0}a^{k+1} = a^{k+1}. The s = k+1 term of the second sum gives \binom{k}{k}b^{k+1} = b^{k+1}. For 1 \leq r \leq k, the two sums contribute terms with the same powers a^{k+1-r}b^r, and their coefficients add:
This is Pascal's identity — proven in properties of combinations. Collecting all terms:
The formula holds for k + 1. By induction, it holds for all positive integers n.
The general term
General term of the binomial expansion
The (r+1)-th term of the expansion of (a + b)^n is:
The index r starts at 0, so T_1 = \binom{n}{0}a^n, T_2 = \binom{n}{1}a^{n-1}b, and so on up to T_{n+1} = \binom{n}{n}b^n.
The general term formula is the workhorse. To find any specific term, plug in n, a, b, and r.
Finding a specific term. The 5th term of (2x + 3)^7:
Here n = 7, a = 2x, b = 3, and the 5th term has r = 4 (since the (r+1)-th term corresponds to r):
Pascal's triangle and the binomial coefficients
The coefficients \binom{n}{r} arrange themselves into Pascal's triangle — a triangular array where each entry is the sum of the two entries directly above it:
Each row of the triangle gives you the full expansion of (a + b)^n immediately. To expand (a + b)^6, read row 6: coefficients 1, 6, 15, 20, 15, 6, 1.
Notice the symmetry: the triangle reads the same left-to-right as right-to-left. This reflects \binom{n}{r} = \binom{n}{n-r}. The row also sums to 2^n — set a = b = 1 in the binomial theorem: (1+1)^n = \sum \binom{n}{r} = 2^n. These properties are explored in detail in binomial coefficients.
Interactive: build the expansion term by term
Drag the red point along the horizontal axis to set n from 1 to 6. The figure shows the terms of (1 + x)^n evaluated at x = 0.5, stacked as a bar chart. Each bar represents one term \binom{n}{r}(0.5)^r, and the total height is (1.5)^n.
Some useful special cases
Setting b = -b. Replace b with -b in the theorem:
The terms alternate in sign: +, -, +, -, \dots
Setting a = 1, b = x. This gives the often-used form:
Setting a = b = 1.
The sum of all the binomial coefficients in row n is 2^n. This makes sense: 2^n is the total number of subsets of an n-element set, and \binom{n}{r} counts the subsets of size r.
Setting a = 1, b = -1.
The alternating sum of binomial coefficients is always 0. More on both of these in binomial coefficients.
Two worked examples
Example 1: Expand $(2x + 3)^5$ and find the coefficient of $x^3$
Step 1. Identify a, b, and n. Here a = 2x, b = 3, n = 5.
Why: the binomial theorem requires two terms raised to a power. Writing (2x + 3)^5 in the form (a + b)^n pins down all three ingredients.
Step 2. Write the general term.
Why: (2x)^{5-r} = 2^{5-r} \cdot x^{5-r} separates the numerical coefficient from the power of x.
Step 3. Find the term with x^3. Set 5 - r = 3, giving r = 2.
Why: r = 2 means the third term. The coefficient \binom{5}{2} = 10, and the numerical part 2^3 \cdot 3^2 = 72 multiplied by 10 gives 720.
Step 4. Write the full expansion for verification. Using r = 0, 1, 2, 3, 4, 5:
Spot-check: r = 0 gives \binom{5}{0} \cdot 2^5 \cdot 3^0 = 32. r = 5 gives \binom{5}{5} \cdot 2^0 \cdot 3^5 = 243. The first term is (2 \cdot 1)^5 \cdot 3^0 \cdot x^5 — consistent. The last term is 3^5 = 243 — also consistent with setting x = 0: (0 + 3)^5 = 243.
Result. The coefficient of x^3 in (2x + 3)^5 is 720.
The coefficient 720 emerges from three multiplicative pieces: the combinatorial count \binom{5}{2}, the power of 2 from the a-part, and the power of 3 from the b-part. Missing any one of these three is the most common calculation error.
Example 2: Find the $4$th term and the last term of $\left(x - \dfrac{2}{x}\right)^9$
Step 1. Identify a, b, and n. Here a = x, b = -\dfrac{2}{x}, n = 9.
Why: the negative sign belongs to b. Writing b = -2/x ensures the alternating signs appear automatically through (-1)^r.
Step 2. Write the general term.
Why: \left(-\frac{2}{x}\right)^r = (-2)^r \cdot x^{-r}. Combining x^{9-r} \cdot x^{-r} = x^{9-2r} simplifies the exponent.
Step 3. Find the 4th term (r = 3).
Why: \binom{9}{3} = \frac{9 \cdot 8 \cdot 7}{3!} = 84, and (-2)^3 = -8. The negative sign is automatic.
Step 4. Find the last term (r = 9).
Spot-check: the expansion has 9 + 1 = 10 terms, so T_{10} is indeed the last. Its sign is negative because (-2)^9 = -512 (odd power).
Result. The 4th term is -672\,x^3 and the last term is -\dfrac{512}{x^9}.
The power of x in the general term is 9 - 2r, which decreases as r increases: x^9, x^7, x^5, x^3, x^1, x^{-1}, x^{-3}, x^{-5}, x^{-7}, x^{-9}. The general and middle terms article takes this further — finding the term independent of x, the middle term, and the term with the greatest coefficient.
Common confusions
-
"The r-th term has \binom{n}{r}." The (r+1)-th term has \binom{n}{r}. Indexing starts at r = 0, so T_1 corresponds to r = 0, T_2 to r = 1, and so on. The most common error is an off-by-one mistake: looking for the 5th term but using r = 5 instead of r = 4.
-
"The exponents of a and b both start from n." The exponent of a starts at n and decreases: n, n-1, \dots, 0. The exponent of b starts at 0 and increases: 0, 1, \dots, n. At every step, the two exponents sum to n.
-
"(a - b)^n has all negative signs." The signs alternate. The term with b^r carries a factor (-1)^r: positive when r is even, negative when r is odd.
-
"There are n terms in the expansion." There are n + 1 terms, because r ranges from 0 to n inclusive.
-
"The binomial theorem works for any exponent." The version stated here requires n to be a positive integer. For fractional or negative exponents, the expansion becomes an infinite series (the generalised binomial theorem), which is a different result with different convergence requirements.
Going deeper
If you can expand (a + b)^n, find any specific term, and read coefficients off Pascal's triangle, you have the full toolkit for problems on the binomial theorem. The rest of this section is for readers who want to see a few more connections.
The binomial theorem for (1+x)^n and approximations
When x is small compared to 1, the higher-power terms of (1 + x)^n become negligible. The first two terms give the linear approximation:
For instance, (1.01)^{10} \approx 1 + 10(0.01) = 1.1. The exact value is 1.10462..., so the approximation is accurate to about 4\%. Including the x^2 term improves it: 1 + 0.1 + \binom{10}{2}(0.01)^2 = 1 + 0.1 + 0.0045 = 1.1045, matching four decimal places.
This approximation is used throughout physics and engineering for quick estimates — and it is the seed of the general binomial series for non-integer exponents.
Connection to the multinomial theorem
The binomial theorem expands (a + b)^n — two terms raised to a power. The natural generalisation is the multinomial theorem, which expands (a_1 + a_2 + \cdots + a_k)^n. The coefficients become multinomial coefficients \frac{n!}{r_1!\, r_2! \cdots r_k!}, and the sum runs over all non-negative integer sequences (r_1, r_2, \dots, r_k) with r_1 + r_2 + \cdots + r_k = n.
Pingala and the Indian roots of Pascal's triangle
The triangular array of binomial coefficients is often called Pascal's triangle in textbooks, but the pattern was known in India long before. Pingala's Chandahshastra (around 200 BCE) described the systematic enumeration of binary patterns in Sanskrit prosody — effectively constructing the rows of the triangle. Halayudha's 10th-century commentary made the additive rule (\binom{n}{r} = \binom{n-1}{r-1} + \binom{n-1}{r}) explicit. The triangle is sometimes called the Meru-prastara (staircase of Mount Meru) in Indian mathematical tradition.
Where this leads next
- Binomial Coefficients — properties of \binom{n}{r}: the sum 2^n, the alternating sum 0, and other coefficient identities.
- General and Middle Terms — finding specific terms, the middle term, the term independent of x, and the greatest coefficient.
- Combinations — Basics — the foundation: why \binom{n}{r} = \frac{n!}{r!(n-r)!} and how combinations relate to permutations.
- Properties of Combinations — Pascal's identity, symmetry, and Vandermonde's identity.
- Algebraic Identities — the square and cube identities that are special cases of the binomial theorem for n = 2 and n = 3.