Intuition What this page is
The parent note told you the machine: ( a + b ) n = ∑ k = 0 n ( k n ) a n − k b k . This page drives that machine through every kind of question an exam can invent . We first draw a map of all the situations, then solve one example per situation so you never meet a case you haven't seen.
Before symbols run wild, three reminders in plain words:
( k n ) ("n choose k ") = the number of ways to pick k things out of n . See Combinations — nCr .
n ! ("n factorial") = n ⋅ ( n − 1 ) ⋯ 2 ⋅ 1 , and 0 ! = 1 . See Factorials .
T k + 1 = ( k n ) a n − k b k is the ( k + 1 ) -th term : the counter k starts at 0 , so the first term has k = 0 .
n " is allowed to be on this page
Throughout, we take n to be a non-negative whole number : n ∈ { 0 , 1 , 2 , 3 , … } . In this note we write N = { 0 , 1 , 2 , … } — so N includes 0 here . The finite binomial theorem holds for every such n , including the trivial n = 0 (worked as Ex 0 below). Everything except that one degenerate case also has n ≥ 1 , but the sum ∑ k = 0 n that STOPS at k = n is valid for all n ≥ 0 .
Common mistake Read this before you use ANY example below
The finite binomial theorem — the neat sum ∑ k = 0 n that STOPS at k = n — is only valid when n is a non-negative integer . If n is negative or a fraction (e.g. ( 1 + x ) − 1 or ( 1 + x ) 1/2 ), the expansion becomes an infinite series that only converges for ∣ x ∣ < 1 ; that is a different tool, the Binomial series for any index . So: whenever you see "binomial theorem", first check the exponent is a whole number ≥ 0 . If it isn't, none of the "finite" moves here (fixed n + 1 terms, a last term b n , ( k n ) with n an integer) apply.
Intuition The counting picture behind every
( k n )
Before we grind algebra, hold onto the geometric/probabilistic meaning. ( k n ) is the number of paths down a grid of choices: at each of n steps you go "left = pick a " or "right = pick b ", and ( k n ) counts how many distinct routes end with exactly k right-turns. The figure below shows those paths piling up — the tall middle bars are why middle coefficients are the biggest.
How to read it: each dot is a partial choice; the number in it counts the paths reaching it (that is exactly ( k n ) , the same numbers as a row of Pascal's Triangle ). Follow any downward zig-zag: it grabs one letter per bracket. The red highlighted route grabs b three times out of six, landing in the ( 3 6 ) = 20 pile — visually the fattest column. This is why the coefficient of a n − k b k equals ( k n ) : it literally counts routes. (And with n = 0 there are no steps at all: a single dot, one empty path — that's Ex 0.)
Every binomial-theorem question falls into one of these boxes. Each row is a "case class"; the last column names the example that clears it.
#
Case class
What is tricky about it
Example
Z
Degenerate index n = 0 ( a + b ) 0 = 1
zero steps, one term
Ex 0
A
Plain positive terms ( a + b ) n
just turn the crank
Ex 1
B
A minus sign ( a − b ) n
the ( − 1 ) k sign pattern
Ex 2
C
A specific term / power of x
solve for k , don't expand all
Ex 3
D
Term independent of x (power = 0 )
limiting/degenerate power case
Ex 4
E
Both terms carry x , fractional/negative powers
careful exponent bookkeeping
Ex 5
F
Middle term(s) — even vs odd n
two sub-cases n even / odd
Ex 6
G
Evaluating at special points (x = ± 1 )
sums of coefficients
Ex 7
H
Real-world / numerical approximation
using ( 1 + x ) n for small x
Ex 8
I
Exam twist : unknown in the expansion
reverse-engineer n or a constant
Ex 9
We will visit Z→I in order. (In every one, n ∈ N = { 0 , 1 , 2 , … } — recall the definition above.)
Worked example Ex 0 (cell Z) — Expand
( a + b ) 0
Forecast: how many terms does n + 1 predict when n = 0 ? Guess the whole answer.
Count the terms. The expansion has n + 1 = 0 + 1 = 1 term — exactly one.
Why this step? The "number of terms = n + 1 " rule must survive its smallest input; plugging n = 0 tests it.
Write the single term. Only k = 0 is allowed (the sum ∑ k = 0 0 has one index): T 1 = ( 0 0 ) a 0 − 0 b 0 = 1 ⋅ a 0 ⋅ b 0 = 1 .
Why this step? ( 0 0 ) = 1 (one way to choose nothing from nothing), and a 0 = b 0 = 1 by the power rule — even the empty product is 1 . So there are zero choice-steps and one empty path (the lone dot in the figure).
Answer: ( a + b ) 0 = 1 for any a , b (this is the standard convention for any base, including the harmless case where a + b = 0 ).
Why this step? This is the base-of-the-base case: the theorem doesn't break at n = 0 , it just returns the constant 1 , matching ordinary algebra where anything to the power 0 is 1 .
Verify: ( 0 0 ) = 1 , and ∑ k = 0 0 ( k 0 ) a − k b k = 1 . Also ( 2 + 5 ) 0 = 7 0 = 1 . ✓
Worked example Ex 1 (cell A) — Expand
( 2 x + 3 ) 3 fully
Forecast: guess how many terms, and whether the middle coefficients are bigger than the ends.
Write the general term. T k + 1 = ( k 3 ) ( 2 x ) 3 − k ( 3 ) k .
Why this step? Everything is one substitution into T k + 1 ; here a = 2 x , b = 3 , n = 3 .
Run k = 0 , 1 , 2 , 3 (there are n + 1 = 4 terms).
k = 0 : ( 0 3 ) ( 2 x ) 3 = 8 x 3
k = 1 : ( 1 3 ) ( 2 x ) 2 ( 3 ) = 3 ⋅ 4 x 2 ⋅ 3 = 36 x 2
k = 2 : ( 2 3 ) ( 2 x ) ( 3 ) 2 = 3 ⋅ 2 x ⋅ 9 = 54 x
k = 3 : ( 3 3 ) ( 3 ) 3 = 27
Why this step? Each term multiplies the binomial coefficient by both powers — keep the numbers separate from the x 's.
Assemble: ( 2 x + 3 ) 3 = 8 x 3 + 36 x 2 + 54 x + 27 .
Why this step? Since each k produced a different power of x , no two terms combine — we just line them up in descending powers, so the expansion is simply the four terms added.
Verify: set x = 1 : LHS = ( 2 + 3 ) 3 = 125 ; RHS = 8 + 36 + 54 + 27 = 125 . ✓
Worked example Ex 2 (cell B) — Expand
( x − 2 ) 4
Forecast: which terms come out negative? (Guess before reading.)
Substitute the whole second term including its sign: b = − 2 . So T k + 1 = ( k 4 ) x 4 − k ( − 2 ) k .
Why this step? The commonest mistake is to forget the sign. Writing ( − 2 ) k bakes the alternation in automatically: ( − 2 ) k = ( − 1 ) k 2 k .
Run k = 0 … 4 :
k = 0 : x 4
k = 1 : ( 1 4 ) x 3 ( − 2 ) = − 8 x 3
k = 2 : ( 2 4 ) x 2 ( 4 ) = 24 x 2
k = 3 : ( 3 4 ) x ( − 8 ) = − 32 x
k = 4 : ( 16 ) = 16
Why this step? Odd k gives a negative sign (because ( − 1 ) k = − 1 ), even k positive — that's the signs alternating.
Assemble: ( x − 2 ) 4 = x 4 − 8 x 3 + 24 x 2 − 32 x + 16 .
Why this step? Again each k owns a distinct power of x , so we merely add the five terms; the alternating signs come straight from step 1's ( − 2 ) k .
Verify: set x = 1 : LHS = ( 1 − 2 ) 4 = 1 ; RHS = 1 − 8 + 24 − 32 + 16 = 1 . ✓
Worked example Ex 3 (cell C) — Coefficient of
x 5 in ( 3 + 2 x ) 8
Forecast: you should NOT expand all 9 terms — guess which single k gives x 5 .
General term: T k + 1 = ( k 8 ) ( 3 ) 8 − k ( 2 x ) k = ( k 8 ) 3 8 − k 2 k x k .
Why this step? Only the factor ( 2 x ) k carries x ; its power of x is exactly k .
Match the power: we want x k = x 5 ⇒ k = 5 .
Why this step? Solving one equation beats expanding — this is the whole point of the general term .
Plug k = 5 : coefficient = ( 5 8 ) 3 3 2 5 = 56 ⋅ 27 ⋅ 32 .
Why this step? Because only k = 5 produces x 5 , its single term's numerical part IS the coefficient — no other term contributes. Then 56 ⋅ 27 = 1512 , 1512 ⋅ 32 = 48384 .
Answer: 48384 .
Verify: ( 5 8 ) = ( 3 8 ) = 56 (symmetry), and 3 3 ⋅ 2 5 = 27 ⋅ 32 = 864 , 56 ⋅ 864 = 48384 . ✓
Worked example Ex 4 (cell D) — Term independent of
x in ( x 2 − x 2 ) 6
Forecast: "independent of x " means the total power of x equals 0 . Guess whether such a k even exists.
General term: T k + 1 = ( k 6 ) ( x 2 ) 6 − k ( − x 2 ) k = ( k 6 ) ( − 2 ) k x 2 ( 6 − k ) x − k .
Why this step? x 1 = x − 1 , so the second factor contributes a negative power. We must combine both powers.
Collect the exponent of x : 2 ( 6 − k ) − k = 12 − 3 k .
Why this step? One symbol (x ) can only have one power per term; add the exponents.
Set it to zero (that's what "independent of x " means): 12 − 3 k = 0 ⇒ k = 4 .
Why this step? If this equation had no whole-number solution in 0 ≤ k ≤ 6 , there would be no constant term — a real possibility you must always check. Here k = 4 is valid.
Evaluate: ( 4 6 ) ( − 2 ) 4 = 15 ⋅ 16 = 240 .
Why this step? Only k = 4 hits power 0 , so this one term IS the constant term — evaluate its number and you are done.
Answer: 240 .
Verify: power at k = 4 is 12 − 12 = 0 ✓; value ( 4 6 ) ⋅ ( − 2 ) 4 = 15 ⋅ 16 = 240 ✓.
Worked example Ex 5 (cell E) — Coefficient of
x 7 in ( x + x 2 3 ) 10
Forecast: x = x 1/2 and x 2 1 = x − 2 . Guess: will the required k be a whole number?
General term: T k + 1 = ( k 10 ) ( x 1/2 ) 10 − k ( 3 x − 2 ) k = ( k 10 ) 3 k x 2 10 − k x − 2 k .
Why this step? Rewrite every root and reciprocal as a power of x first — never mix notations.
Collect the exponent: 2 10 − k − 2 k = 2 10 − k − 4 k = 2 10 − 5 k .
Why this step? Combine into a single power so we can match it to x 7 .
Match: 2 10 − 5 k = 7 ⇒ 10 − 5 k = 14 ⇒ k = − 5 4 .
Why this step? k must be an integer with 0 ≤ k ≤ 10 . Here k = − 0.8 is not a valid k .
Conclusion: there is no x 7 term — its coefficient is 0 .
Why this step? Covering the degenerate outcome: not every power you ask for appears. Always report 0 when no valid k exists.
Verify: check a power that DOES appear — k = 2 gives 2 10 − 10 = 0 , so a constant term exists; k = 0 gives x 5 . The reachable powers are 5 , 2 5 , 0 , … stepping by − 2 5 ; x 7 is not in that list. ✓
Worked example Ex 6 (cell F) — Middle term(s) of (i)
( x + y ) 6 and (ii) ( x + y ) 7
Forecast: an even power gives ONE middle term, an odd power gives TWO. Guess which position(s). (The path picture at the top of this page shows why the middle piles up highest.)
(i) n = 6 (even). There are n + 1 = 7 terms, so the single middle one is term number 2 n + 1 = 4 .
Middle term = T 4 , i.e. k = 3 .
Why this step? Term index = k + 1 , so the 4 th term uses k = 3 — exactly the centre of 0 … 6 .
T 4 = ( 3 6 ) x 3 y 3 = 20 x 3 y 3 .
Why this step? Substituting k = 3 into T k + 1 with a = x , b = y gives equal powers x 3 y 3 — the balanced centre, which the figure shows as the tallest single column.
Answer (i): 20 x 3 y 3 .
(ii) n = 7 (odd). There are 8 terms; the two middle ones are terms 2 n + 1 = 4 and 2 n + 3 = 5 , i.e. k = 3 and k = 4 .
3. T 4 = ( 3 7 ) x 4 y 3 = 35 x 4 y 3 , and T 5 = ( 4 7 ) x 3 y 4 = 35 x 3 y 4 .
Why this step? ( 3 7 ) = ( 4 7 ) = 35 — the coefficients are mirror-symmetric about the centre (in the path picture, the two central columns have identical height).
Answer (ii): 35 x 4 y 3 and 35 x 3 y 4 .
Verify: (i) ( 3 6 ) = 20 ✓. (ii) ( 3 7 ) = 35 = ( 4 7 ) ✓; both middle coefficients equal, as Pascal's Triangle predicts.
Worked example Ex 7 (cell G) — Find
k = 0 ∑ n ( k n ) and k = 0 ∑ n 2 k ( k n )
Forecast: these are the same theorem, just fed clever inputs. Guess the first answer (2 n ?).
Start from ( 1 + x ) n = ∑ k = 0 n ( k n ) x k (put a = 1 , b = x ).
Why this step? Sums of binomial coefficients are just the expansion read at a chosen x .
Put x = 1 : ( 1 + 1 ) n = 2 n = ∑ k = 0 n ( k n ) .
Why this step? At x = 1 every x k becomes 1 , so the expansion collapses to the bare sum of coefficients — instantly 2 n .
So k = 0 ∑ n ( k n ) = 2 n — also the number of subsets of an n -element set.
Put x = 2 : ( 1 + 2 ) n = 3 n = ∑ k = 0 n ( k n ) 2 k .
Why this step? Choosing x = 2 makes each x k = 2 k , exactly the weight we want on ( k n ) ; the left side evaluates in one shot.
So k = 0 ∑ n 2 k ( k n ) = 3 n .
Verify (n=4): ∑ ( k 4 ) = 1 + 4 + 6 + 4 + 1 = 16 = 2 4 ✓; ∑ 2 k ( k 4 ) = 1 + 8 + 24 + 32 + 16 = 81 = 3 4 ✓.
Worked example Ex 8 (cell H) — Estimate
( 1.02 ) 5 using the binomial expansion
Forecast: since 0.02 is tiny, high powers of it are negligible. Guess ( 1.02 ) 5 ≈ 1.1 ?
Write 1.02 = 1 + x with x = 0.02 : ( 1 + x ) 5 = ∑ k = 0 5 ( k 5 ) x k .
Why this step? The ( 1 + x ) n form is built for approximations — the terms shrink fast when x is small. This is the finite cousin of the Binomial series for any index (and here n = 5 ∈ N , so the sum is exact and finite).
Keep only the first three terms (units: pure number, x dimensionless):
( 1 + x ) 5 ≈ 1 + 5 x + ( 2 5 ) x 2 = 1 + 5 x + 10 x 2 .
Why this step? x 3 = 8 × 1 0 − 6 is far below our target precision — dropping it is the whole art of approximation.
Substitute x = 0.02 : 1 + 5 ( 0.02 ) + 10 ( 0.0004 ) = 1 + 0.10 + 0.004 = 1.104 .
Why this step? We evaluate the kept terms and add; the discarded x 3 and higher would only nudge the 4th decimal, so 1.104 is safe to 3 places.
Answer: ( 1.02 ) 5 ≈ 1.104 .
Verify: true value 1.0 2 5 = 1.1040808 … ; our estimate 1.104 is correct to 3 decimals. ✓
Worked example Ex 9 (cell I) — Find
n if the coefficients of x 2 and x 3 in ( 1 + x ) n are equal
Forecast: the coefficient of x k in ( 1 + x ) n is just ( k n ) . Guess: does equal-coefficients force a specific n ?
Identify the coefficients: coefficient of x 2 is ( 2 n ) , of x 3 is ( 3 n ) .
Why this step? In ( 1 + x ) n the power of x equals k , so "coefficient of x k " literally means ( k n ) .
Set them equal: ( 2 n ) = ( 3 n ) .
Why this step? Translate the word "equal" into an equation before touching factorials.
Expand and simplify. 2 ! ( n − 2 )! n ! = 3 ! ( n − 3 )! n ! . Cancel n ! and cross-multiply:
3 ! ( n − 3 )! = 2 ! ( n − 2 )! ⇒ 6 ( n − 3 )! = 2 ( n − 2 )! .
Why this step? We clear the fractions so the unknown n is no longer trapped inside a denominator.
Cancel the factorials. Since ( n − 2 )! = ( n − 2 ) ( n − 3 )! , divide both sides by ( n − 3 )! :
6 = 2 ( n − 2 ) .
Why this step? ( n − 3 )! is common to both sides, so cancelling it collapses the factorial equation into ordinary algebra — one linear equation in n .
Solve for n . 6 = 2 ( n − 2 ) ⇒ 3 = n − 2 ⇒ n = 5 .
Why this step? Divide by 2 , then add 2 ; n = 5 is a valid non-negative integer, so the finite theorem genuinely applies and the answer is admissible.
Verify: in ( 1 + x ) 5 , ( 2 5 ) = 10 and ( 3 5 ) = 10 — equal. ✓
Common mistake The three traps this page trains against
Case D/E: if the exponent equation gives a non-integer or out-of-range k , the answer is 0 (the term doesn't exist) — don't force it.
Case B: substitute the sign into b ; write ( − 2 ) k , not 2 k .
Case F: even n → one middle term at k = 2 n ; odd n → two middle terms at k = 2 n − 1 , 2 n + 1 .
Mnemonic Which case am I in? — each rule with its worked anchor
"Power zero? Answer is just 1." → like Ex 0: ( a + b ) 0 = 1 , one term, no work.
"Whole thing? Crank it." → like Ex 1: small n , list all n + 1 terms, e.g. ( 2 x + 3 ) 3 gives 4 terms.
"One term? Match the power." → like Ex 3: set the x -power equal to what's asked, solve for k (there k = 5 ), evaluate that single term.
"No x ? Set power to zero." → like Ex 4: write the total exponent 12 − 3 k , put = 0 , get k = 4 , evaluate.
"Small number? Keep three terms." → like Ex 8: ( 1 + x ) n with tiny x , keep 1 + n x + ( 2 n ) x 2 , e.g. ( 1.02 ) 5 ≈ 1.104 .
"Non-integer power? STOP — that's the infinite series, not this page." → recall the warning box; use Binomial series for any index .
Recall Quick self-test
What is ( a + b ) 0 and how many terms does it have? ::: 1 ; exactly one term (n + 1 = 1 ).
Coefficient of x 5 in ( 3 + 2 x ) 8 ? ::: 48384 (from k = 5 ).
Constant term of ( x 2 − 2/ x ) 6 ? ::: 240 (from k = 4 ).
Coefficient of x 7 in ( x + 3/ x 2 ) 10 ? ::: 0 — no integer k works.
∑ k = 0 n 2 k ( k n ) = ? ::: 3 n (put x = 2 in ( 1 + x ) n ).
Middle term of ( x + y ) 6 ? ::: 20 x 3 y 3 (T 4 , k = 3 ).
Estimate ( 1.02 ) 5 to 3 places? ::: 1.104 .
If ( 2 n ) = ( 3 n ) then n = ? ::: 5 .
Does the finite theorem apply to ( 1 + x ) − 2 ? ::: No — negative index gives an infinite series (∣ x ∣ < 1 ).