9th Lord of the Math

Solution Booklet

Saint Stephen’s High School

December 3, 2016
9th Lord of the Math
Contributors
Solution Booklet Balete Immanuel Josiah
Balete Nathanael Joshua
Saint Stephen’s High School Chuatak John Thomas
Co Randall Lewis
Sison Richie Rainier
Sy Julius Vincent
Tan Hans Markson
Tiu Benedict Ryan
Yu Marc Adrian

December 3, 2016

DISCLAIMER: Not all of the problems here are original. Some are lifted or edited from previous competi-

tions or textbooks. All information provided here is for educational purposes only.
Team Finals
TF1 Problem The Russian Cyrillic alphabet has 33 letters, 21 of which are
consonants and 10 are vowels. The remaining two letters do not fit in
either category, and are called “signs”. The two signs cannot appear at
the beginning of a word and can only follow a consonant. How many
three-letter Russian “words” (strings of letters) with at least one vowel
satisfy the above condition?
Answer 21 370
Solution Denote by C a consonant, V a vowel, and S a sign. Then
only strings with length three that are of the following types satisfy
the conditions: V V V , V VC and permutations, VCC and permu-
tations, VCS, and CSV . There are 103 = 1000 strings of the form
V V V , 21 × 102 × 3 = 6300 strings of the form V VC (including per-
mutations), 212 × 10 × 3 = 13 230 strings of the form VCC (including
permutations), and 21 × 10 × 2 × 2 = 840 strings of the form VCS or
CSV , giving a total of 21 370 possible strings.
TF2 Problem Taxicab Geometry is one kind of non-Euclidean geometry
where all points are in the x y-plane and the distance function is de-
fined as the sum of the positive differences of their corresponding
1 2
x-, and y-coordinates. For example, the distance between (1, 2) and
(.5, −6) is 8.5. Find the area of the circle in a taxicab geometry cen-
tered at the origin with radius 50 units. A circle is defined as the set
of points that are equidistant from its center.
Answer 5000 square units
Solution It can be easily verified that this circle has the shape of a
square, with vertices (±50, 0) and (0, ±50). Therefore, the area is
502 ⋅ 2 = 5000 square units.
TF3 Problem ISBNs are numbers that are used to identify most published
books. It consists of ten digits, where the first nine digits a1 to a9
range from 0 to 9, and the tenth digit a10 from 0 to 10. The first nine
digits give information about the book for identification, and the
tenth digit is a “check digit” to check if the first nine digits might be
wrongly encoded. The check digit a10 is chosen given the first nine
digits such that
10
∑ ia i ≡ 0 (mod 11).
i=1
If the first nine digits are 0 − 825417 − 39, find the check digit.
Answer 8
 Solution Expanding the given condition, we have
0(1) + 8(2) + 2(3) + 5(4) + 1(5) + 4(6) + 7(7)
+ 8(3) + 9(9) + 10 (a10 ) = 217 + 10a10 ≡ 0 (mod 11).
Since 217 ≡ 8 (mod 11), then a10 = 8 because 80 + 8 ≡ 0 (mod 11).
TF4 Problem For a positive integer n, let r n be a positive integer from 1
to 9 inclusive such that n ≡ r n (mod 9). Define
A = {n ≤ 1000 ∣ n ≡ 0 (mod r n )} .
Find the cardinality of A.
Answer 527
Solution It is easy to see that for all positive integers x ≤ 1000 such
that r x = 1, 3, or 9, then x ∈ A. There are 112 numbers that are 1
(mod 9), 111 that are 3 (mod 9), and 111 that are 9 (mod 9).
If x ≡ 2 (mod 9), then x = 18a + 2, a a nonnegative integer. Since
x ≤ 1000, a ≤ 55. Thus there are 55 + 1 = 56 numbers here. (Note
that a = 0 is a valid case.)
If x ≡ 4 (mod 9), then x = 36a + 4. The possible values for a here
are {0, 1, . . . , 27}, giving 28 numbers.
3 4
If x ≡ 5 (mod 9), then x = 45a + 5, and a ≤ 22. Thus there are 23
numbers for this case.
If x ≡ 6 (mod 9), then x = 18a + 6, and like in the case where x ≡ 2
(mod 9), there are 56 possible numbers.
If x ≡ 7 (mod 9), then x = 63a + 7. The largest value of a such that
x ≤ 1000 is a = 15, so there are 16 numbers for this case.
Finally, if x ≡ 8 (mod 9), then x = 72a + 8, and there are 14 possible
numbers for x.
In total, the cardinality of A is 112 + 111 + 111 + 56 + 28 + 23 + 56 +
16 + 14 = 527.
TF5 Problem The two front tires of a new four-wheeled car will wear
out after 38 400 km, whereas the two rear tires will wear out after
51 600 km. Also, suppose that five identical tires, including one spare
tire, come with the car. If you can easily change the tires whenever
you want, what is the maximum distance that can be driven?
Answer 55 040 km
Solution If we assume that tires wear out at a constant rate, then the
total wear (i.e., if the wear on a tire is 1 then the tire is unusable) on
 2 2 1
√
the four tires for every kilometer is . Thus r 3
+ =
38 400 51 600 11 008 O1 O2 O3 is equilateral, O1 ( , r). Since O4 lies on the altitudes of
2 2
the maximum number of kilometers obviously happens if the wear △O1 O2 O3 , it is the orthocenter
√ and thus the centroid, as the triangle
is spread evenly among the five tires. Thus the total distance that can r 3
is equilateral. Then O4 ( , r).
2 6
be traveled is 5 × 11 008 = 55 040 km.
Note that A, O1 , O4 and B are collinear and the line that contains
TF6 Problem Three circles with equal radii Γ1 , Γ2 , Γ3 are on a plane such them is perpendicular to the segment √ O2 O3 . Since AO1 is a radius,
that each of the three circles passes through the other two circles’ r 3+2
then A is r units above O1 , i.e., A ( , r); similarly BO1 is a
2 2
centers. A smaller circle Γ4 is internally tangent to all three circles, √
r 3−2
and the three circles are all internally tangent to a larger circle Γ5 . If radius thus B ( , r). A radius of Γ4 is O4 B, which has length
√ √ 2 2 √
the product of the lengths of the radii of the five circles is equal to 162, 3 3−2 3− 3
r− r= r. On the other hand, a radius of Γ5 is O4 A,
6 2 √ 3
find the radius of Γ1 . 3+2
√
3 3+ 3
√
which has length r− r= r.
Answer 3 2 6 3
√
Solution 3
3− 3
Therefore the product of the lengths of the radii is r ⋅ r⋅
√ 3
A 3+ 3 2
O1 r = 162, or r 5 = 162. Solving for r, we have r = 3.
3 3
O4 √
TF7 Problem The volume of a right circular cylinder is 6 3 cm3 . What
O2 O3
B is its minimum possible total surface area?
√
Answer 18 3 π cm2
Let O n be the center of circle Γn . (In this case O3 = O4 .) We as- Solution Let r and h be the radius and the height of the cylinder,
sign O2 (0, 0), O3 (r, 0), such that r is the radius of Γ1 . Then, since respectively. Also let T SA be the total surface area, and V the vol-

5 6
 √
ume. Then from AM-GM, T SA = 2πr 2 + 2πrh ≥ 3 2πr 2 (πrh)2 =
3 Therefore either y − x is −2 or 5. The sum is 5 − 2 = 3.
√ √
3 2π (πr 2 h)2 = 3 2πV 2 = 18 3 π cm2 .
3 3 √
TF10 Problem The sum of the first ten terms of an arithmetic sequence

TF8 Problem Find cot (cot−1 2 + cot−1 3 + cot−1 5 + cot−1 7). is 155 and the sum of the first two terms of a geometric sequence is
11 9. Find all possible ordered pairs of the common difference and the
Answer
23
Solution Note that Arg(2 + i) = cot−1 2, Arg(3 + i) = cot−1 3, Arg(5 + common ratio (d, r) if the common difference is the first term of the

i) = cot−1 5, and Arg(7 + i) = cot−1 7. Then, by de Moivre’s formula, geometric sequence and the common ratio is the first term of the

we have arithmetic sequence.

2 25
Answer (3, 2) and ( , )
3 2
cot−1 2 + cot−1 3 + cot−1 5 + cot−1 7
Solution As from the problem let the common difference, and the
= Arg(2 + i) + Arg(3 + i) + Arg(5 + i) + Arg(7 + i) first term of the geometric sequence, be d; and let the common ratio,
= Arg ((2 + i)(3 + i)(5 + i)(7 + i)) and the first term of the arithmetic sequence, be r. Then we know
23 11 that r + (r + d) + (r + 2d) + ⋯ + (r + 9d) = 10r + 45d = 155, and
= Arg(110 + 230i) = tan−1 = cot−1
11 23 9
d + dr = 9. Solving for r in the second equation gives r = − 1 (since
d
90
d =/ 0). Then, −10+45d = 155. This simplifies to 90+45d 2 = 165d,
d
TF9 Problem x34y73 is divisible by 7. Find the sum of all possible values 2
2
or 3d − 11d + 6 = 0 ⇒ (d − 3)(3d − 2) = 0. Therefore d = 3 or d = .
of y − x. 3
2 25
If d = 3 then r = 2; if d = then r = . Thus the ordered pairs are
Answer 3 3 2
2 25
Solution Since 34 073 ≡ 4 (mod 7), 100 000 ≡ 5 (mod 7), and 100 ≡ (3, 2) and ( , ).
3 2
2 (mod 7) then x and y satisfy the equivalence (5x+2y) ≡ 3 (mod 7). TF11 Problem Suppose you have a circular pizza divided into six equal
But (5x + 2y) ≡ (−2x + 2y) ≡ 10 (mod 7), so (y − x) ≡ 5 (mod 7).

7 8
slices, and you have to choose one flavor for each slice. If there are third. If we first denote the flavor with three slices as A, the one with
three flavors to choose from, and adjacent slices have to have different two as B, and the last one C our only permutation is ABABAC. Now
flavors, how many ways are there to flavor the pizza? there are three choices for A, two for B and one for C. Thus we have
Answer 14 6 ways.
Solution Three cases have to be considered: one where there are three
Therefore we have a total of 3 + 5 + 6 = 14 ways.
slices of one flavor and three of another; one where there are two slices
TF12 Problem In a diving competition, 5 judges score each dive on a scale
for each of the three flavors; and another where there is one slice of
from 1 to 10. The point value of the dive is obtained by dropping the
the first flavor, two of the second, and three of the third. One cannot
highest and lowest scores and multiplying the sum of the remaining
have 4 or more slices for one flavor as it will require some adjacent
scores by the degree of difficulty. If a dive with a degree of difficulty
slices to have the same flavor.
of 3.2 received scores of 7.5, 8.0, 9.0, 6.0 and 8.5, what was the point
Case 1. Three slices for each of two flavors. If the flavors are A and B
value of the dive?
then one can only have ABABAB as the order of the flavors. Note that
Answer 76.8
ABABAB = BABABA since this is just rotation by one slice. Therefore
Solution The sum of the scores excluding the highest and lowest
3
there are ( ) = 3 ways in this case.
2 scores is 24. Multiplying by 3.2 we get 76.8.
Case 2. Two slices for each of three flavors. The only possible permuta-
TF13 Problem For what real values of k will the function f (x) = (k −2)x +
tions are ACBCAB, ABCBAC, ABCABC, ACBABC, and ACBACB.
3k − 4, x ∈ R be even, and for what values of k will the function be
All other permutations are rotations of any of the above. Thus there
odd?
are five ways here. 4
Answer Even: k = 2; Odd: k =
3
Case 3. One slice for the first flavor, two for the second, three for the Solution An even function f satisfies f (−x) = f (x). Therefore we

9 10
 have −(k − 2)x + 3k − 4 = (k − 2)x + 3k − 4 for all real x, thus k − 2 = 0. 4
cards being adjacent is and 1 flip is needed. The probability that
32
Thus k = 2 will make the function even. 6
the two are not adjacent is and 2 flips are needed.
32
An odd function, meanwhile, satisfies f (−x) = − f (x). Therefore we
Case 4. There are exactly 3 white cards. The probability that the
have −(k − 2)x + 3k − 4 = −((k − 2)x + 3k − 4) for all real x, which 3
4 three cards are adjacent is and 1 flip is needed. The probability
implies 3k − 4 = 0. Thus k = will make the function odd. 32
3 6
that exactly two of the three cards are adjacent is and 2 flips are
32
TF14 Problem Lewis has five cards. Each card has one black and one white 1
needed. The probability that none are adjacent is and 3 flips are
face. He shuffles the five cards and puts them in a row. If Lewis can 32
needed.
flip consecutive cards with the same face to the other face, what is the
Case 5. There are exactly 4 white cards. The probability that the four
expected value of the minimum number of flips needed to make all
2
cards are adjacent is and 1 flip is needed. The probability that
the cards black face up? 32
3 3
Answer exactly three of the four are adjacent is and 2 flips are needed.
2 32
Solution We work by cases. For brevity a “white card” means a card Case 6. There are exactly 5 white cards. The probability that this
whose white face is face up. 1
happens is and 1 flip is needed.
32
Case 1. There are exactly 0 white cards. The probability of this hap- 1 5 4 6 3
1 Therefore, the expected value is ⋅0+ ⋅1+ ⋅1+ ⋅2+ ⋅
pening is and 0 flips are needed. 32 32 32 32 32
32 6 1 2 3 1 48 3
1+ ⋅2+ ⋅3= ⋅1+ ⋅2+ ⋅1= = .
32 32 32 32 32 32 2
Case 2. There is exactly 1 white card. The probability of this happening
5 TF15 Problem In the cryptarithm APAT + APAT = WALO, each letter
is and 1 flip is needed.
32 consistently represents one digit from 0 to 9. Two letters cannot rep-
Case 3. There are exactly 2 white cards. The probability of the two resent the same digit, and a number cannot start with the digit zero.

11 12
What is the sum of all possible values of the 4-digit number WALO? to the thousands place and none to the tens place. If T = 0, O = 0,
The letter O does not necessarily represent the digit zero. contradicting the fact that all letters must represent different digits.
Answer 40 226 If T = 1, O = 2; but A = 2. Similarly T =/ 2. If T = 3, O = 6, but P = 6.
Solution First, A can only be 1, 2, 3 or 4. Also, either W is one more T =/ 4 since L = 4. Therefore there are no solutions for this case.
than L or one less. In fact, A cannot be 1 or 3 because there are no
2 6 2 ❙
0❙
✓✓ 2 6 2 1 2 6 2 ✓
2
❙
integer solutions for 2P = 1, 2P = 11, 2P = 3 or 2P = 13.
+ 2 6 2 ❙
0✓
✓❙ + 2 6 2 1 + 2 6 2 ❙2✓
Case 1. A = 2. First consider the case where W = 4 and L = 5: 5 2 4 ❙
0❙
✓✓ 5 2 4 ✓
2
❙ 5 2 4 ✓
4
❙

2P2T + 2P2T = 425O. Here, P = 1 as there is no carry-over to the

2 6 2 3 2 6 2 ❙4✓
thousands place. On the other hand, there is a carry-over to the tens
+ 2 6 2 3 + 2 6 2 ✓
4
place; thus T ≥ 5. T cannot be 5 as it has already been used. If T = 6,
❙

5 2 4 ❙
6❙ 5 2 4 8
then O = 2; but A = 2. If T = 7, O = 4; if T = 8, O = 6; and if T = 9,
✓✓

O = 8. These are the three solutions. Case 2. A = 4. First consider the case where W = 8 and L = 9. Then
2P = 4 or P = 2. Since there is a carry-over to the tens digit, T ≥ 5. If
2 1 2 ❙
5❙
✓✓ 2 1 2 6 2 1 2 7
T = 6, O = 2 but already P = 3. If T = 7, O = 4 but L = 4. T =/ 8 as
+ 2 1 2 ❙
5❙
✓✓ + 2 1 2 6 + 2 1 2 7
W = 8 and T =/ 9 as L = 9. Thus, T can only be 5, and O = 0.
4 2 5 0 4 2 5 ✓
2
❙ 4 2 5 4
4 2 4 5 4 2 4 6 4 2 4 7
2 1 2 8 2 1 2 9
+ 4 2 4 5 + 4 2 4 6 + 4 2 4 7
+ 2 1 2 8 + 2 1 2 9
8 4 9 0 8 4 9 ✓
2
❙ 8 4 9 ✓
4
❙
4 2 5 6 4 2 5 8
If W = 5 and L = 4, then P = 6 and T ≤ 4 since there is a carry-over

13 14
 4 2 4 ❙
8❙
✓✓ 4 2 4 ❙
9❙
✓✓ Individual Semifinals

+ 4 2 4 ❙
8❙
✓✓ + 4 2 4 ❙
9❙
✓✓ Easy Round

8 4 9 6 8 4 9 ❙
8✓
✓❙ IS-E1 Problem Bulbasaur, Charmander, and Squirtle have some berries
If W = 9 and L = 8, then P = 7. Now T ≤ 4 as there is no carry-over to eat. All three are in a generous mood, so Bulbasaur gives Char-
to the tens digit. If T = 0, O = 0 which cannot be; if T = 2, O = 4 but mander as many berries as Charmander has and Squirtle as many
already A = 4. Similarly T =/ 4. If T = 1, O = 2; if T = 3; O = 6. These Squirtle has. Then, Charmander does the same, giving Bulbasaur
are the two solutions. and Squirtle as many berries as they each have. Finally, Squirtle

4 7 4 ❙
0❙
✓✓ 4 7 4 1 4 7 4 2 gives Bulbasaur and Charmander as many berries as each have. If

+ 4 7 4 ❙
0✓
✓❙ + 4 7 4 1 + 4 7 4 2 after this each has 16 berries, how many berries did Bulbasaur have

9 4 8 ❙
0❙
✓✓ 9 4 8 2 9 4 8 ✓
4
❙ at first?
Answer 26
4 7 4 3 4 7 4 ❙4✓ Solution We work backwards. Before Squirtle shared its berries
+ 4 7 4 3 + 4 7 4 ✓
4 16
❙
we know that Bulbasaur and Charmander each have = 8 berries.
2
9 4 8 6 9 4 8 ❙
8❙
✓✓
Since there is a total of 48 berries, Squirtle has 48−8−8 = 32 berries.
Our solutions are 4254, 4256, 4258, 8490, 9482, 9486. Their sum is 8
Before Charmander shared, Bulbasaur has = 4 berries and Squir-
2
40 226. 32
tle, = 16 berries, leaving Charmander with 48 − 4 − 16 = 28
2
berries. Thus, at the beginning of the game, before Bulbasaur shared,
28 16
Charmander has = 14 berries and Squirtle, = 8 berries. Thus,
2 2
Bulbasaur has 48 − 14 − 8 = 26 berries.

15 16
IS-E2 Problem Find the last digit of 711 + 810 − 912 . Solution 2 log x = log(2x) implies that x 2 = 2x, since 2 log x =
Answer 4 log x 2 . This gives us x = 0 or x = 2. If x = 0, there is no value for
Solution Note that 711 < 911 and 810 < 911 . Thus 711 +810 < 911 +911 < y as log 0 is undefined; if x = 2, y = 2 log 2. Therefore there is only
912 , and the given expression is negative. From modulo arithmetic one intersection.
it is known that the last digit of 711 is 3, the last digit of 810 is 4, and
IS-E5 Problem How many permutations of five distinct letters are there
the last digit of 912 is 1. Therefore the last digit of 711 + 810 − 912 is
if one letter has to be in front of another?
not 3 + 4 − 1 = 6, but 10 − 6 = 4.
Answer 60
IS-E3 Problem A convex polyhedron consists solely of hexagonal and Solution Without the restriction, there are 5! = 120 permutations.
quadrilateral faces. If for all vertices three faces meet at a vertex, Now there is a bijection between permutations where one letter, say
how many quadrilateral faces are there? A, is always in front of another, say B, and permutations where B is
Answer 6 ahead of A. Thus exactly half of the permutations satisfy the given
120
Solution Let there be m hexagonal faces and n quadrilateral faces. condition, or = 60 permutations.
6m + 4n 6m + 4n 2
We have m + n faces, vertices, and edges, since
3 2 Average Round
two adjacent faces of convex polyhedra always meet at an edge. By
6m + 4n 6m + 4n IS-A1 Problem At most how many circles of radius 1 cm can fit inside a
Euler’s polyhedron formula we have +(m+n) = +
3 2 square with side 20 cm such that the circles do not overlap?
2. Simplifying the equation, the m’s cancel out, and n = 6.
Answer 105
IS-E4 Problem At how many points do the graphs of y = 2 log x and
Solution The circle have to be packed such that adjacent circles’
y = log(2x) intersect in the x y-plane?
centers form equilateral triangles. Refer to the following figure.
Answer 1

17 18
 ((4n + 1) − (4n + 3))i = 2 − 2i. This happens 504 times, thus the
answer is 1008 − 1008i.

IS-A3 Problem An unfair coin is tossed n times, for some positive integer
√
√ n. If the variance of the distribution of the number of heads is 99,
3
find the minimum possible value for n.
Answer 40
Since the diameter of each circle is 2 cm, then ten of them can fit Solution The variance of the binomial distribution Bi(n, p) is np(1−
√
exactly on the bottom row. Thus the second bottom row will have p) = 99, where p is the probability of heads. By AM-GM, p(1 −
1 √ n √ √
one less (9 circles). The rows will alternate having 10 and 9 circles. p) < (as p =/ 1 − p), so 99 < , or n > 4 99 = 1584. Since
√4 √ √ 4
Now we have to find how many rows there will be. 39 = 1521 < 1584 < 1600 = 40, the minimum possible value
√ of n is 40.
If there are n rows, then the height will be 2 + (n − 1) 3. Now
√ √
2 + (n − 1) 3 ≤ 20, or n ≤ 6 3 + 1. But 100 < 108 < 121 ⇒ 10 < IS-A4 Problem A turtle born on January 1 in the first half of the nineteenth
√ √ √
6 3 < 11 ⇒ 11 < 6 3+1 < 12. Therefore there are ⌊6 3 + 1⌋ = 11 century was x years old in the year x 2 . How old is it now, in years?
rows. There are six rows with ten circles and five rows with nine Answer 210 years old
√ √ √
circles, thus we can fit at most 60 + 45 = 105 circles. Solution We see that 42 = 1764 < 1800 < 1849 = 43 <
√
2016 √ 1850. Thus x = 43, and the turtle was born in 432 − 43 = 1806. It
IS-A2 Problem Evaluate the series ∑ kik , where i = −1.
k=0 is now 210 years old.
Answer 1008 − 1008i
2016 4n+4
Solution The sum is equal to ∑ kik . We consider the sum ∑ kik , IS-A5 Problem Find the sum of all values of x that satisfy the equation
k=1 k=4n+1
where n is a nonnegative integer. This sum is (4n + 4) − (4n + 2) + 2⌊x⌋ = x + 2{x}, where ⌊x⌋ is defined as the least integer greater

19 20
 than or equal to x, and {x} = x − ⌊x⌋. Currently our function is f (x) = x 4 + 8x 3 + 4x 2 + 0x + 7, and to
Answer 4 make the sum of the coefficients 31, 1 has to be decreased from a
Solution Substituting x = ⌊x⌋ + {x} into the original equation gives coefficient and 12 added to the coefficient of the lower power of x.
us 2⌊x⌋ = ⌊x⌋ + 3{x}, or ⌊x⌋ = 3{x}. Since 0 ≤ {x} < 1, 0 ≤ ⌊x⌋ < 3. We can’t remove 1 from 0, as all coefficients are positive (having a
Thus ⌊x⌋ can only be 0, 1, or 2. coefficient as zero just means the term won’t exist, and so does not
1 violate the conditions). Thus the possible functions are f1 (x) = x 4 +
If ⌊x⌋ = 0, then {x} = 0, and x = 0; if ⌊x⌋ = 1, then {x} = and
3 8x 3 +3x 2 +12x +7, f2 (x) = x 4 +7x 3 +16x 2 +7, f3 (x) = 20x 3 +4x 2 +7.
4 2 8
x = ; if ⌊x⌋ = 2, then {x} = and x = . The sum of all possible
3 3 3 Then f1 (10) + f2 (10) + f3 (10) = 18 427 + 18 607 + 20 407 = 57 034.
values, then, is 4. ¿ √
Á √ √
3 3 √
Á
À 3 3 3
√
Difficult Round IS-D2 Problem Evaluate 21 3
2 2 2 25 29 ⋯, where the expo-
1 1 3

IS-D1 Problem A polynomial f with positive integer coefficients satisfies nents of 2 are from the successive terms of the sequence 1, 1, 1, 3, 5, 9, . . .,

f (12) = 311×113. If the sum of its coefficients is 31, find all possible where the first three terms are 1 and succeeding terms are generated

values of f (10). by getting the sum of the last three terms.

√
Answer 57 034 Answer 7 16

Solution Let f (x) = a n x n + a n−1 x n−1 + ⋯ + a1 x + a0 . Then f (12) = Solution Denote by a the given expression. Then

a n 12n + a n−1 12n−1 + ⋯ + a1 ⋅ 12 + a0 . A possible set of values for the 1 1 1 3 5 9 1 1 1 3 5 9

a = 2 3 2 9 2 27 2 81 2 243 2 729 ⋯ = 2 3 + 9 + 27 + 81 + 243 + 729 +⋯ .
coefficients are the consecutive digits of the base-12 representation
of 311 × 113 = 35 14310 = 18 40712 . Note that the sum of the digits
here is 20, and that 20 + 12 − 1 = 31.

21 22
 Let the exponent of a be x. Then b n = 3b n−1 + a n−1 . Also, the initial values are a1 = 3, b1 = 1.
1 1 1 3 5 9 Adding the two equations together gives us a n +b n = 4(a n−1 +b n−1 ),
x= + + + + + +⋯ (1)
3 9 27 81 243 729
1 1 3 5 9 a1 + b1 = 4. Therefore, a n + b n = 4n .
3x = 1+ + + + + +⋯ (2)
3 9 27 81 243
1 3 5 9 Similarly, subtracting one from the other results in a n −b n = 2(a n−1 −
9x = 3 + 1+ + + + +⋯ (3)
3 9 27 81 b n−1 , a1 − b1 = 2. Therefore, a n − b n = 2n . This means that a n =
3 5 9
27x = 9 + 3 + 1+ + + +⋯ (4) 4n + 2n 4n − 2n
3 9 27 and b n = .
2 2
8 4 42016 + 22016
(4) − (3) − (2) − (1) gives us (27 − 9 − 3 − 1)x = 8 or x = = . The question is asking for a2016 , which is = 24031 +22015 .
4 √ √ 14 7 2
7
Therefore a = 2 = 2 = 16.
7 4 7

IS-D4 Problem For each of the about 7 billion people in the world, com-
IS-D3 Problem How many 2016-digit positive integers that only have dig- pute the product of the number of fingers in his/her right hand, left
its 1, 2, 3, and 4 are there such that the number has an even number hand, right foot and left foot. Suppose the about 7 billion products
of 2’s? are also multiplied together. Give a reasonable estimate, within 5%
Answer 24031 + 22015 of the exact answer, of this value. You may leave your answer in
Solution Let a n be the number of n-digit integers that satisfy the exponential form.
above condition. Also, let b n be the number of n-digit integers Answer 0
with only 1, 2, 3, and 4 but with an odd number of 2’s. Then a n = Solution An amputee will have zero as the product, and thus the
3a n−1 + b n−1 , because a n consists of all numbers with either (a) 1, product of all the numbers is zero. Only zero is accepted since
3, or 4 as the first digit and with an even number of 2’s among the 0 ± 5% = 0.
succeeding (n − 1) digits, or (b) 2 as the first digit and with an
IS-D5 Problem The angle of elevation of a building is observed from a
odd number of 2’s among the succeeding (n − 1) digits. Similarly

23 24
point on the horizontal plane on which it stands. At a point x feet Individual Finals
√
nearer the angle of elevation is the complement of the original angle −1 + 5
E Problem What is the value of φ12 + φ8 + φ5 + 160φ, where φ = .
observed. At another point y feet nearer (from the second point) 2
Answer 99
the angle of elevation is now double the first. Express the height of
Solution Note that φ2 + φ − 1 = 0.
the building
√ in terms of x and y.
x 2 First, we show by induction that for all positive integers n,
2
Answer (x + y) − ( )
2
Solution Let the original angle of elevation observed have measure φ n = (−1)n+1 Fn φ + (−1)n Fn−1 ,

A, the height of the building be h, and the distance between the where F0 = 0, F1 = 1, Fi = Fi−1 + Fi−2 for integers i ≥ 2 are the Fibonacci
building and the third point be z. Then from the given tan A = numbers.
h y+z h
= , and tan 2A = . Now from the double-angle
x+y+z h z Now for n = 1, LHS = (−1)2 F1 φ + (−1)1 F0 = φ + 0 = φ = RHS.
y+z
2(
2(y + z)h
)
h h Assume that the statement is true for n = k. Then φ k = (−1)k+1 Fk φ +
formula = 2 = 2 , or 2z(y + z) = h2 − (y +
z y+z h (y + z)2
1−( ) (−1)k Fk−1 . Now for n = k + 1:
h
z)2 ⇒ h2 = (y + z)(y + 3z).
φ k+1 = φ k φ = ((−1)k+1 Fk φ + (−1)k Fk−1 ) φ
h y+z
But ⇒ h2 = (y + z)(x + y + z). Equating the
= (−1)k+1 Fk φ2 + (−1)k Fk−1 φ
=
x+y+z h
x
two implies x + y + z = y + 3z as y + z =/ 0. Therefore, z = . Now = (−1)k+1 Fk (−φ + 1) + (−1)k Fk−1 φ
2
2 2
3 2 2
x 2
h = (y + z)(x + y + z) = y + 2x y + x = (x + y) − ( ) , or = (−1)k φ (Fk + Fk−1 ) + (−1)k+1 Fk
√ 4 2
2
x = (−1)k φFk+1 + (−1)k+1 Fk
h = (x + y)2 − ( ) .
2
= (−1)k+2 φFk+1 + (−1)k+1 Fk

25 26
 Thus the statement is true for n = k + 1, and by induction true for all k votes and B has k + 1 votes. The smallest k is the first time A trails.
positive integers n. This means that For every bad list, after the first instance A trails, we swap all remaining
votes of A to B and vice versa. Therefore after this point A gets 17 − k
φ12 = (−1)13 F12 φ + (−1)12 F11 = −144φ + 89
votes and B gets 26 − k votes, giving us A with k + 17 − k = 17 votes and
8 9 8
φ = (−1) F8 φ + (−1) F7 = −21φ + 13
B with k + 1 + 26 − k = 27 votes.
5 6 5
φ = (−1) F5 φ + (−1) F4 = 5φ − 3
After the manipulation, B will always win. Therefore any list with 17 A’s
Adding these three equations gives us φ12 + φ8 + φ5 = −160φ + 99, or and 27 B’s can be reverted to a bad list by selecting the first time B has
φ12 + φ8 + φ5 + 160φ = 99. more votes than A (which is guaranteed) then swapping A’s votes with
B’s votes afterwards. Thus there exists a bijection between bad lists and
A Problem In a class election for class head, 44 students write their choice
44
lists with 17 A’s and 27 B’s, and there are ( ) bad lists.
on a slip of paper. Now the teacher counts the votes in a uniformly 17
random order. If Student A gets 26 votes and Student B gets 18 votes, Thus, the probability is
what is the probability that Student A never trailed during the counting (44
# of bad lists ) 44!26!18! 2 1
) = 1 − 17 =1− =1− = .
process?
(1 − 44
# of lists (18) 27!17!44! 3 3
1
Answer
3
Solution We define a list like ABAABAB . . ., which tells us who was 2064
∏ k!
selected for the ith slip. Now, since the votes for A and votes for B can D Problem Find all nonnegative integers N ≤ 2064 such that
k=1
is a
44 N!
be rearranged we have ( ) lists with 26 A’s and 18 B’s. perfect square.
18
We call a list with 26 A’s and 18 B’s good if it satisfies the given condition Answer 1032

and bad otherwise. A list is bad when there exists a k such that A has

27 28
Solution First we show that 1032 is a solution.
2064 1032
∏ k! = ∏ (2k)!(2k − 1)!
k=1 k=1
1032
= ∏ (2k) ((2k − 1)!)2
k=1
1032 1032
2 1032
= ( ∏ ((2k − 1)!) ) (2 ∏ k)
k=1 k=1
1032 2
= (2516 ∏ (2k − 1)!) (1032!)
k=1

Therefore N = 1032 is a possible solution. Now assume for the sake

of contradiction that there exists another nonnegative integer M that
M!
satisfies this condition. Then it follows that either (if M > 1032)
1032!
2064 2064
∏ k! ∏ k!
1032!
or (if M < 1032) is a perfect square, since both and
k=1 k=1
M! 1032! M!
are perfect squares.
M!
Note that if M > 1032, then will always contain only one copy of
1032!
M!
1033, since 1033 is prime. Thus will never be a perfect square.
1032!
1032!
Similarly, if M < 1032, then will always contain only one copy of
M!
1032!
1031, since 1031 is prime. Thus will never be a perfect square.
M!
Therefore only N = 1032 satisfies the given condition.

29
10th Lord of the Math
Solution Booklet

Saint Stephen’s High School

January 13, 2018


10th Lord of the Math
Solution Booklet
Saint Stephen’s High School
January 13, 2018
JHS Individual Finals
JIF-1 Problem What is the coefficient of the term containing x 4 y 3 in the expansion (x +
y)7 ?
Answer 35
7
Solution Using the binomial theorem, the coefficient is ( ) = 35.
3
JIF-2 Problem Triangle ABC has lengths AB = 6, BC = 7, and AC = 8. Find the length
of the angle bisector from A to BC.
Answer 6
Solution Let AD (D ∈ BC) be the angle bisector. Let BD = x, which implies that
DC = 7 − x.
Then, from the angle bisector theorem, we have 8x = 6(7 − x) ⇒ 8x = 42 − 6x ⇒
x = 3. Therefore, BD = 3 and DC = 4.
The length of AD can now be determined using Stewart’s theorem:
AC 2 ⋅ BD + AB2 ⋅ DC = BC ⋅ (BD ⋅ DC + AD2 )
64 ⋅ 3 + 36 ⋅ 4 = 7(3 ⋅ 4 + AD2 )
192 + 144 = 7(12 + AD 2 )
AD = 6
JIF-3 Problem Convert the octal number 20178 to base ten.
4
 6
Answer 1039 2nπ
complex numbers are also equal; thus ∑ cos = 0. However, from trigono-
n=0 7
Solution In base ten 20178 = 2 × 512 + 1 × 8 + 7 = 1039. 2π 12π 4π 3π 10π 6π
metric identities, cos = cos , cos = − cos = cos and cos =
7 7 7 7 7 7
JIF-4 Problem If real numbers x and y satisfy (x+5)2 +(y−12)2 = 142 , find the minimum π 8π
− cos = cos .
7 7
value of x 2 + y 2 . 3π 2π π
Substituting these results in, we have −2 cos + 2 cos − 2 cos + 1 = 0 ⇒
7 7 7
Answer 1 π 2π 3π 1
cos − cos + cos = .
7 7 7 2
Solution The line connecting the centers of circles C1 ∶ (x + 5)2 + (y − 12)2 = 142
JIF-7 Problem For how many integers x less than or equal to 2017 is the expression
5 12
and C2 ∶ x 2 + y 2 = c 2 has equation 12x + 5y = 0. The line intersects C1 at ( , − ). √ √ √
13 13 x x x an integer?
If c 2 is chosen to minimize x 2 + y 2 subject to C1 , then C1 , C2 and the line connecting
the centers of the two circles should concur at the point of tangency of C1 and C2 , Answer 3
5 12 5 2 12 2 √ 7 √
i.e., ( , − ) ∈ C2 . Therefore, c 2 = ( ) + (− ) = 1.
7
Solution The expression is equivalent to x 8 = ( 8 x) . Since this is an integer, 8 x
13 13 13 13
JIF-5 Problem The mean of three numbers is 7 and the mode is 5. Find the range. also has to be an integer. Since 38 = 6561 > 2017, and x ≥ 0, we have three possible
values for x: 08 , 18 , and 28 = 256.
Answer 6
JIF-8 Problem For odd integers n, define the double factorial to be n!! = 1⋅3⋅5⋯(n−2)⋅n.
Solution Since 5 is the mode and the mean is greater than 5, the two smaller num-
Find the last three digits of 55!!.
bers are 5. Then the third number is 7 × 3 − 5 − 5 = 11. The range is 11 − 5 = 6.
π 2π 3π Answer 625
JIF-6 Problem Evaluate cos − cos + cos .
7 7 7
Solution It is easy to see that 53 ∣ 55!!, and that 55!! is odd. Thus, it only suffices to
1
Answer find 55!! mod 8 to get 55!! mod 1000 due to the Chinese remainder theorem, since
2
2π 8 and 125 are two coprime numbers that multiply to 1000.
Solution From De Moivre’s theorem, x = cis is a seventh root of unity. Thus
7
x 7 − 1 = 0. Since x ≠ 1, x 6 + x 5 + x 4 + x 3 + x 2 + x + 1 = 0. Note that the product of any four consecutive odd integers is congruent to (−3) ⋅
6
2π 2nπ
Substituting x = cis into the RHS gives ∑ cis = 0. The real parts of equal
7 n=0 7

5 6
 (−1) ⋅ 1 ⋅ 3 ≡ 9 ≡ 1 (mod 8). Thus are bounded by the segments of the triangles?

28 Answer 3003
55!! ≡ ∏(2i − 1) ≡ 17 ≡ 1 (mod 8).
i=1
Solution Consider first the parallelograms whose sides are not parallel to the bot-
tom side of the triangle. To find the number of such parallelograms, we construct
Since 55!! ≡ 1 (mod 8) and 55!! ≡ 0 (mod 125), then 55!! ≡ 625 (mod 1000).
a bijection. First draw one extra row at the bottom of the triangle, as shown. If
JIF-9 Problem Real numbers a and b satisfy 3a = 7 and 7b = 27. Find the value of ab.
we extend the four segments to intersect this extra row, four points are obtained.
Answer 3 It is easy to see that these four points uniquely determine a parallelogram in this
14
case. Thus the number of parallelograms here is ( ) = 1001, as there are fourteen
Solution By the properties of exponents, 3ab = 7b = 27 = 33 . Therefore, ab = 3. 4
points on the extra row.
JIF-10 Problem Real numbers m and n satisfy m2 − 2m − 5 = 0 and n2 − 2n − 5 = 0. Find
Similarly we have 1001 parallelograms whose sides are not parallel with the left side
all possible values of m3 − n3 .
√ and 1001 whose sides are not parallel with the right side of the triangle. Thus there
Answer 0, ±18 6
are 3003 parallelograms.
Solution A quadratic polynomial has at most two roots. We consider two cases.

Case 1. m = n. Then m3 − n3 = 0.

Case 2. m ≠ n. Then from Vieta’s formulas, m + n = 2 and mn = −5. Now m3 − n3 =

√ √
(m − n)(m2 + mn + n2 ) = ± (m + n)2 − 4mn ⋅ ((m + n)2 − mn) = ± 4 − (−5) ⋅
√ √
(22 − 4(−5)) = ±2 6 ⋅ 9 = ±18 6. Note that the sign of m3 − n3 varies if m > n or
m < n.
√
Thus there are three possible values for m3 − n3 : 0, ±18 6.

JIF-11 Problem An equilateral triangle with side length 12 is divided into 144 congruent JIF-12 Problem An infinite sequence of inscribed semicircles with decreasing radii is drawn
non-overlapping equilateral triangles with side length 1. How many parallelograms such that their straight edges are all parallel. Find the area of all the regions inside

7 8
 an odd number of semicircles, if the radius of the largest semicircle is 100 units? JIF-14 Problem An ordered list of 100 real numbers is such that the sum of any 9 consec-
9
10 000π utive numbers on the list are all equal, i.e., if a n is the nth term, then ∑ a i = ⋯ =
Answer i=1
3 100

Solution Consider two semicircles constructed in this fashion. If the larger semi- ∑ a i , and that the sum of any 11 consecutive numbers are also all equal. Find the
i=92
largest possible value of the range of the numbers.
√ has radius r, then by the Pythagorean theorem, the smaller one has radius
circle
r 2
. Answer 0
2
Solution From the given it is implied that if x ≡ y (mod 9) or x ≡ y (mod 11),
√
r 2 then a x = a y . But the list has more than 9 × 11 elements, implying that all the
2 √
r 2 numbers must be equal. Thus the range is zero.
r 2

JIF-15 Problem The sum of n consecutive integers is 192. Find the greatest possible value
for n.
Therefore, the area of the shaded region is
Answer 384
10 000π
10 000π 5000π 2500π 1250π 2 10 000π 2 10 000π
− + − +⋯= = ⋅ = Solution Note that (−191)+(−190)+⋯+191+192 = 192, which has 384 consecutive
2 2 2 2 1 2 3 3
1 − (− )
2 integers. It is clear that a longer string off consecutive integers cannot sum up to
192. Let x be the smallest number among the consecutive integers. Then x + (x +
JIF-13 Problem How many rectangles can be formed by a 4×101 grid of squares such that n(n − 1)
1) + ⋯ + (x + n − 1) = 192 ⇒ nx + = 192 ⇒ n(n + x − 1) = 384. Since
2
the rectangles contain the square on the second column, second row?
both n and n + x − 1 have to be integers, n ≤ 384. We have shown that n = 384 is
Answer 1200 possible; thus it is the maximum value.

Solution Each rectangle can be determined by a pair of horizontal lines and a pair
of vertical lines. A rectangle that contains the square on the second column, second
row would have one line to the left, one to the right, one above, and one below the
said square. Thus the number of rectangles is 2 × 100 × 2 × 3 = 1200.

9 10
JHS Team Finals not break. What is the minimum number of drops John needs to guarantee that
he finds the highest "safe" floor?
JTF-1 Problem Evaluate sin(55○ ) sin(65○ ) sin(175○ ) + sin(125○ ) sin(130○ ) sin(240○ ) +
Answer 64
sin(35○ ) sin(95○ ) sin(115○ ) + sin(50○ ) sin(145○ ) sin(150○ ).
√ √ Solution We denote a + ⋯ + b as the sum of all integers between a and b, inclusive.
6− 2
Answer
4 n
n(n + 1)
First, note that 64 is the smallest integer n that satisfies ∑ i = ≥ 2017, as
Solution i=1 2
63(64)
= 2016. This is how the drops will work:
2
sin(55○ ) sin(65○ ) sin(175○ ) + sin(125○ ) sin(130○ ) sin(240○ ) We drop the first tablet on the following floors, until it breaks: floors 64, 64 + 63,
+ sin(35○ ) sin(95○ ) sin(115○ ) + sin(50○ ) sin(145○ ) sin(150○ ) 64 + 63 + 62, . . ., 64 + ⋯ + 12 = 2014, 2015, and 2016. (We know the tablet will break
= cos(35○ ) cos(25○ ) sin(5○ ) − cos(35○ ) cos(40○ ) cos(30○ ) on the top floor.) There are 55 floors in this sequence.

+ sin(35○ ) cos(5○ ) cos(25○ ) + cos(40○ ) sin(35○ ) sin(30○ ) If the first tablet breaks on the 64th floor, then the maximum number of drops
= cos(25 )(cos(35 ) sin(5 ) + cos(5 ) sin(35 ))
○ ○ ○ ○ ○
needed to know the highest floor is 63 for the second tablet, as John would need
− cos(40○ )(cos(35○ ) cos(30○ ) − sin(35○ ) sin(30○ )) to drop his second (and last) tablet from the first floor going up, as there are no

= cos(25○ ) sin(40○ ) − cos(40○ ) cos(65○ ) replacements. Thus at most 64 drops are needed.

= cos(25○ ) sin(40○ ) − cos(40○ ) sin(25○ ) Now, if the first tablet breaks on floor 64 + ⋯ + n, 12 ≤ n ≤ 63, then we know that
√ √
6− 2 the floor we want is between floors 64 + ⋯ + (n + 1) and 64 + ⋯ + n. To get to floor
= sin(15 ) =
○
4 64 + ⋯ + n the first tablet would have been dropped 65 − n times; the second tablet
will require a maximum of n − 1 drops from (64 + ⋯ + (n + 1)) + 1 to 64 + ⋯ + n.
JTF-2 Problem Rugged Corp is making a new tablet that can survive falls from heights
We still need a maximum of 64 drops.
(but not from 2018 floors high). To test this the company gives John, a quality
assurance tester, two units of the said tablet on a 2017-floor skyscraper. John will If the first tablet doesn’t break from the 2014th floor, we proceed with the 2015th

try to determine the highest floor from which a dropped tablet will survive the fall. and 2016th floors, requiring a maximum of 55 drops.

The company will not give John extra tablets but John can reuse a tablet that did Thus John needs to do a minimum of max(64, 64, 55) = 64 drops.

11 12
 Note that if the initial floor is not 64, it will not be optimal. If it is < 64, then if the JTF-5 Problem How many integers x satisfy (x − 20)17 (x − 17)20 (x − 2017)2017 ≤ 0?
tablet will only break at the 2017th floor, more than 64 drops are needed. On the
Answer 1999
other hand, if it is > 64, then if the tablet breaks at the 64th floor, more than 64
Solution Obviously, x = 17, 20, 2017 are solutions. Note that (x − 17)20 is always
drops are needed.
non-negative. We consider two cases:
JTF-3 Problem How many permutations of STEPHEN IAN are there such that the vow-
Case 1. (x − 20)17 < 0 and (x − 2017)2017 > 0. Then x < 20 and x > 2017, which is
els are in alphabetical order?
impossible.
Answer 75 600
Case 2. (x − 20)17 > 0 and (x − 2017)2017 < 0. Then 20 < x < 2017. There are
10!
Solution There are a total of = 907 200 ways to rearrange the letters without
2!2! 2016 − 21 + 1 = 1996 integers that satisfy this.
any restrictions, since there are two E’s and two N’s.
Adding x = 17, 20, 2017, there are a total of 1999 integers.
4!
Consider the rearrangements of the vowels only. There are = 12 ways to arrange
2!
JTF-6 Problem Let a, b, c, d be four positive integers. If the LCM of a and b is 60, the
the vowels. This means that the 907 200 ways can be grouped into groups of 12,
LCM of a, b, and c is 120, and the LCM of b, c, and d is 24. What is the least
such that within each group S, T, P, H and the two N’s are in the same positions.
possible value of the LCM of c and d?
Only one in each group have the vowels in alphabetical order.
Answer 8
Therefore, the number of permutations whose vowels are in alphabetical order is
907 200
= 75 600. Solution The LCM of c and d is at least the bigger of the two, so minimizing c and
12
d is a good tactic if c and d on a large scale. From the given at least one of a and
JTF-4 Problem What is the largest factor of 11! that is one more than a multiple of 6?
b is divisible by 4 and both are not divisible by 8. Since the LCM of a, b, and c is
Answer 385
120, then c is divisible by 8. This means that the LCM is at least 8.
Solution The prime factorization of 11! is 28 ⋅ 34 ⋅ 52 ⋅ 7 ⋅ 11. Since the factor we
Say c = 8. Since the LCM of b, c and d is 24, then we can choose b as a factor of 3
are finding is not divisible by 6, then it can be expressed as 5a1 7a2 11a3 , where a1 ∈
and d = 1, so that the LCM is equal to 8. Since this is possible with a = 20, b = 3,
{0, 1, 2}, and a2 , a3 ∈ {0, 1}. Since 5 ≡ 11 ≡ (−1) (mod 6) and 7 ≡ 1 (mod 6),
c = 8, d = 1, the minimum LCM of c and d is 8.
the largest factor that is congruent to 1 (mod 6) is 5 × 7 × 11 = 385.

13 14
JTF-7 Problem How many integers evenly divide 999 999? JTF-10 Problem There are 1007 points in the interior of a convex pentagon such that no
three of 1012 points, including the vertices of the pentagon, are collinear. The pen-
Answer 128
tagon is partitioned into several triangles. Each vertex of each of these triangles
Solution 999 999 = 33 ⋅7⋅11⋅13⋅37. Therefore we have (3+1)(1+1)4 = 64 positive
is either a vertex of the pentagon or one of the 1007 points. How many triangles
factors. If x evenly divides 999 999, −x does too; thus the total number of integers
result?
that evenly divide 999 999 is 128.
Answer 2017
1027 + 2
JTF-8 Problem Find the sum of the digits of the decimal expansion of .
6
Solution Consider the interior angles formed in the triangles. If x triangles are
Answer 158
formed, then the sum of the interior angles of all the triangles, in degrees, is 180x.
1027 + 2 1 1027 − 1 1
Solution Note that = ⋅( + 1) = ⋅ 3⋯3 4 = 1 6⋯6 7. The sum
6 2 3 2 ± ± However, this sum is also equal to the sum of the internal angles of the pentagon
26 3’s 25 6’s
of the digits is 1 + 6 × 25 + 7 = 158. and 360○ for each point, since the sum of the internal angles that have a specific

JTF-9 Problem The answers to each of the five statements below is either true or false. point as vertex is 360○ . This is equal to 540 + 360 ⋅ 1007 in degrees.

1. The answers of Statement 3 and Statement 4 are different. Therefore, 180x = 540 + 360 ⋅ 1007 ⇒ x = 3 + 2 ⋅ 1007 = 2017.

2. Statement 1 is true.
3. There are more False answers than True answers.
4. There are more False answers than True answers in the above statements.
5. There is an unequal amount of True and False answers in the above statements.
What are the answers to each of the five statements?
Answer False, False, True, True, False
JTF-11 Problem How many pounds of H2 O must be evaporated from 50 pounds of a 3%
salt solution so that the remaining solution will be 5% salt?
Answer 20
Solution There are 50(0.03) = 1.5 pounds of salt in the original solution. If x
is the number of pounds of water that must be evaporated, then the remaining
solution contains 50 − x pounds. Since the amount of salt is unchanged, then (50 −

Solution If #2 is true, then #1 is true. This means that #4 is false, in turn implying x)(0.05) = 1.5 ⇒ 50 − x = 30 ⇒ x = 20.
4034
that #3 is true. This contradicts #1 being true. JTF-12 Problem Find the remainder when ( ) is divided by 20172 .
2017
Therefore, #2 is false. Then, #1 is false, #4 is true, #3 is true, and #5 is false. Answer 2

15 16
 Solution From the Vandermonde identity, we have

2017
4034 2017 2017 OA
( )= ∑( )( ). 1
2017 i=0 i 2017 −i OC r
1
r
2017 D
Note that the first and last terms in the expansion of the right-hand side, ( )⋅
2017 1 1
2017 2017
( ), each equal 1. Since 2017 is a prime, then it divides ( ) for all integers i
0 i 1 1
2017 2017 OB OX
from 1 to 2016. Thus for all integers i from 1 to 2016, ( )( ) is divisible
i 2017 − i
by 20172 .
4034
Thus, the remainder when ( ) is divided by 20172 is 2.
2017 Note that quadrilateral O A OC O B O X is a kite, as O X O A = O X O B = 2 and O B OC =
√
JTF-13 Problem Circle X with radius 3 contains a concentric circle Y with radius 1. Two OC O A = r + 1. Furthermore, O A O B = 2 2 and O X OC = 3 − r. Therefore,
other circles, A and B, are congruent to Y and are tangent to both X and Y such √
2 2(3 − r) √ √
that the centers of circles A, B and Y form a right triangle. A smaller circle C is [O A OC O B O X ] = = 3 2 − 2r,
2
tangent to circles A, B, and X. Find the radius of circle C.
√ and [O A O X O B ] = 2.
9−3 2
Answer
7
Let D be the midpoint of O A O B . Since △O A O B OC is isosceles, we know that
√
Solution Let O Z be the center of circle Z, and r be the radius of circle C.
OC O X ⊥ O A O B . From the Pythagorean theorem, OC D = (OC O B ) − (O B D) =
2 2
¿ √ 2
Á
À(r + 1)2 − ( 2 ) = √r 2 + 2r − 1.
Á
2
√ √
2 2 r 2 + 2r − 1 √ 2
Then [O A OC O B ] = = 2r + 4r − 2. This means that
2
√ √ √
[O A OC O B O X ] = [O A O X O B ] + [O A OC O B ] = 3 2 − 2r = 2 + 2r 2 + 4r − 2
√
9±3 2
. This equation simplifies to 7r 2 − 18r + 9 = 0, with roots r = .
7

17 18
 Note that from the question
√ circle C√with radius r is smaller
√ than the circles with SHS Elims, Part 1
9−3 2 9+3 2 9−3 2
radius 1. Since <1< , we have r = .
7 7 7
SE1-E1 Problem Suppose a statistician wants to test if two standard dice are loaded or
JTF-14 Problem How many right triangles can be formed using the vertices of a cube?
not, using the sum of the numbers on top as a one-tailed test statistic. If he rolls
Answer 48 the dice and both turn up five, what is the p-value?

Solution There are two kinds of right triangles that can be formed: one isosceles 1
Answer
√ √ √ 6
with sides 1, 1, 2 and the other with sides 1, 2, 3, as shown. There are 24 for
Solution Since the test statistic X is considered to be one-tailed, the p-value is
each, so there are 48 right triangles in all. 3
equal to P(X = 10) + P(X = 11) + P(X = 12). But P(X = 10) = , P(X = 11) =
36
2 1 1
, and P(X = 12) = . Therefore, the p-value is .
36 36 6
SE1-E2 Problem A 100×100 grid (with ten thousand cells) is drawn. The positive integers
from 1 to 10 000 are written, one each, in each cell such that the sum of the num-

Ð→ bers on each row, column, or main diagonal are all equal to a certain constant.
JTF-15 Problem AB is a diameter of circle O. P is a point on AB outside the circle and C
Find this constant.
is a point on the circle such that CP is a tangent. If AP = 8 and CP = 4, find the
length of AB. Answer 500 050
1
Answer 6 ⋅ 10 000 ⋅
Solution The sum of all positive integers less than or equal to 10 000 is
2
10 001 = 50 005 000. Since the sum of the numbers on each row should be equal
Solution Let the radius of the circle be x. Since CP is a tangent, OCP is a right
to k, then 100k = 50 005 000. Therefore, k = 500 050.
triangle with OP as hypotenuse. We have CP = 4, OC = x and OP = 8 − x. From
the Pythagorean theorem, we have x 2 + 16 = x 2 − 16x + 64 or x = 3. Therefore the SE1-E3 Problem Suppose f is a real function satisfying f (x + f (x)) = 4 f (x) and f (1) =

diameter AB has length 6. 4. Find f (21).

Answer 64

Solution Substitute x = 1: f (5) = f (1 + f (1)) = 4 f (1) = 16.

19 20
 Substitute x = 5: f (21) = f (5 + f (5)) = 4 f (5) = 64. mn = 5 ∶ 5 = f (5) = f (5 ⋅ 1) = f (5 + 1) = f (6) = 5

SE1-E4 Problem Find the probability of drawing, without replacement, two aces from a mn = 6 ∶ 5 = f (6) = f (2 ⋅ 3) = f (2 + 3) = f (5) = 5
deck of 52 playing cards.
1 Therefore, g(1) = 4, g(2) = 5, g(3) = 6; g(1) + g(2) + g(3) = 4 + 5 + 6 = 15.
Answer
221 7
SE1-E7 Problem If sin θ = , find sin 2θ.
4 3 25
Solution We multiply the probabilities for each of the two draws: ⋅ =
52 51 336
1 1 Answer ±
= . 625
13 × 17 221
24
SE1-E5 Problem Find the product of all real zeros of log x + log(x + 2) − 3. Solution From the Pythagorean identity, cos θ = ± , the sign dependent on the
25
7 24 336
√ quadrant where θ belongs. Thus sin 2θ = 2 sin θ cos θ = ±2 ⋅ ⋅ =± .
Answer 1001 − 1 25 25 625
SE1-E8 Problem There are one thousand lockers and one thousand students in a school.
Solution The equation log x + log(x + 2) = 3 is equivalent √to x + 2x = 10 , or
2 3

−2 ± 4004 √ The principal asks the first student to go to every locker and open it. Then he has
x 2 + 2x − 1000 = 0. This quadratic equation has roots = −1 ± 1001.
√ 2 √ the second student go to every second locker and close it. The third goes to every
Since log(−1 − 1001) is undefined, the only real zero is −1 + 1001.
third locker and, if it is closed, he opens it, and if it is open, he closes it. The fourth
SE1-E6 Problem Let f (n) denote the sum of the distinct prime factors of positive integer student does this to every fourth locker, and so on. After the process is completed
n. Let g(k) be the kth smallest integer mn (m, n ∈ Z+ ), such that f (m + n) = with the thousandth student, how many lockers are closed?
f (mn). What is g(1) + g(2) + g(3)?
Answer 969
Answer 15
Solution A locker is closed if it is handled by an even number of students. We
Solution We check for smaller numbers. know that only perfect squares have an odd number of factors. Since there are 31
perfect squares from 1 to 1000, there are 1000 − 31 = 969 closed lockers.
mn = 2 ∶ 2 = f (2) = f (2 ⋅ 1) ≠ f (2 + 1) = f (3) = 3
SE1-E9 Problem Find the amplitude of the graph of the function f (x) = 3 sin x + cos x.
mn = 3 ∶ 3 = f (3) = f (3 ⋅ 1) ≠ f (3 + 1) = f (4) = 2
√
mn = 4 ∶ 4 = f (4) = f (2 ⋅ 2) = f (2 + 2) = f (4) = 4 Answer 10

21 22
 √ √ √
f (x) 3 1 Answer 12 + 4 2 + 4 5 + π
Solution Divide by 10 on both sides to get √ = √ sin x + √ cos x. Now
10 10 10
1 3 f (x) 1 Solution
cos sin−1 √ = √ , so we can rewrite this as √ = sin (x + sin−1 √ ).
10 10 √ 10 10
Therefore the amplitude of the function is 10. 90○○
A
1 15 90○○
90
SE1-E10 Problem Positive integers a, b, and c satisfy a+ 1 = . Find (a−2b−4c)2017 .
b+ c 2
Answer 1
α D(180−α) ○
(180−α)○○○ (180−α)○○○○
(180−α)
Bα
15 1 1
Solution Since =7+ =7+ , a = 7 and b = c = 1. Therefore, (a − 2b −
2 2 1 + 11
4c)2017 = 12017 = 1.
99
k+1
SE1-A1 Problem Evaluate the sum ∑
k=1 (k − 1)! + k! + (k + 1)!
(180−β)○○○○
(180−β)
1
Answer 1 −
100! CC
ββ
Solution Simply the summand first:

k+1 k+1
= Note that 90 + 2(180 − α) + (180 − β) = 360 ⇒ 2α + β = 270. Thus, the area
(k − 1)! + k! + (k + 1)! (k − 1)!(1 + k + k(k + 1))
k+1 1 inside the rubber band is equal to the sum of the areas of kite ABCD, the four
= =
(k − 1)!(k + 1)2 (k − 1)!(k + 1) rectangles (which share a side with kite ABCD), and a circle with radius 1, since
k (k + 1) − 1 1 1
= = = − the central angles 90○ , α, α and β add up to 360○ .
(k + 1)! (k + 1)! k! (k + 1)!
The kite has diagonals 4 and 6; thus it has area 12. Also, from the Pythagorean
1 1 1 √ √
Thus, the sum telescopes to − =1− theorem, AB = AD = 2 2, and CD = CB = 2 5. Therefore, the four rectangles
1! 100! 100! √ √
have a total area of 4 2 + 4 5. Finally, a circle with radius 1 has area π.
SE1-A2 Problem Six circles of radius 1 unit are centered at the points (0, 0), (0, 2), (0, −2),
√ √
(0, −4), (2, 0), (−2, 0). If a rubber band is wrapped around this figure, find the Thus, the area of the interior is 12 + 4 2 + 4 5 + π.

area of the region inside the rubber band. SE1-A3 Problem If for some acute angle θ, tan θ + cot θ = 4, what is sin θ?

23 24
 √ √
6± 2
Answer
4
Solution From the original equation multiply both sides by sin θ cos θ (since
sin θ =/ 0 and cos θ =/ 0) to get sin2 θ + cos2 θ = 4 sin θ cos θ. But the left-hand
side of this equation is 1 and the right hand side is precisely
√ 2 sin√ 2θ. Thus we
1 6 − 2
have sin 2θ = or 2θ = 30○ , 150○ . If θ = 15○ , sin θ = ; if θ = 75○ ,
√ √ 2 √ √ 4
6+ 2 6± 2
sin θ = . Thus the values for sin θ are . First we have to choose a point among the eight. This accounts for 8 ways. Let’s say
4 4
SE1-A4 Problem Eight equally-spaced points are on the perimeter of the circle. What is we chose the bigger point shown in the figure. Then, we draw a line to the point

the probability that if three of the points are connected, it forms an acute triangle? 3 points away from the first point, moving clockwise (to avoid double-counting).
1 If we choose any two of the three points, the resulting triangle is always obtuse.
Answer 3
7 Therefore there are ( ) ways to choose these two points. In general, there are
2
Solution We first calculate the number of triangles that can be chosen from these 3
8 8( ) = 24 obtuse triangles.
eight points. This is equal to ( ) = 56 triangles. 2
3
This means that there are 56 − 24 − 24 = 8 acute triangles. The probability, then,
A right triangle is always formed by choosing two diametrically opposite points 8 1
is = .
and one of the six remaining points. There are four pairs of diametrically opposite 56 7

points; therefore, there are 24 right triangles that can be formed. SE1-A5 Problem Find the greatest common factor of F2017 and F2016 , where Fn is defined
as F1 = F2 = 1, Fn = Fn−1 + Fn−2 for integers n > 2.
Now we have to calculate the number of obtuse triangles. Refer to the figure
below. Answer 1

Solution For two integers a > b, the GCF of a and b is equal to the GCF of a − b
and b. Therefore gcf(F2017 , F2016 ) = gcf(F2017 − F2016 , F2016 ) = gcf(F2015 , F2016 ) =
⋯ = gcf(F2 , F1 ) = gcf(1, 1) = 1.

SE1-A6 Problem The value of the function f (x) = ax + b for all integers x is a positive
integer that leaves a remainder of 3 when divided by 4. Find all possible values

25 26
 for a. or the seat of the first moviegoer. If any other seat is vacant when the last movie-
goer enters, then it was also vacant when the moviegoer who is supposed to sit
Answer 0
there entered the cinema — a contradiction.
Solution Note that f (x) > 0 for all integers x. The only linear function that
But since it is equally likely that the last moviegoer goes to either of the two possi-
satisfies this is the constant function. Thus a = 0 and b is any positive integer that
ble seats (the first moviegoer’s seat and the last moviegoer’s seat), the probability
leaves a remainder of 3 when divided by 4.
1
is .
2
SE1-A7 Problem Let −1 < x < 1. Find the value of the infinite series S = 1 + 22 x + 32 x 2 +
10
20
⋯ + n2 x n−1 + ⋯. SE1-D2 Problem The sum ∑ ( )92i−1 721−2i can be expressed in the form 2x + 2 y ,
i=1 2i − 1
1+x where x and y are distinct positive integers. Find x + y.
Answer
(1 − x)3
Answer 98
n(n + 1)
Solution Since n2 = 2 ⋅ − n, then 10
20
2 Solution Let S1 be the above sum. Consider S2 = ∑ ( )92i 720−2i .
i=0 2i
n(n + 1) n−1
S = 2 (1 + 3x + 6x 2 + ⋯ + x + ⋯ ) − (1 + 2x + 3x 2 + ⋯ + nx n−1 + ⋯ ) Then, from the binomial theorem,
2
2 1 1+x
= − = . 20
20
(1 − x)3 (1 − x)2 (1 − x)3 1620 = (9 + 7)20 = ∑ ( )9i 720−i = S1 + S2
i=0 i
20
20
SE1-D1 Problem There are 432 seats in a cinema and 432 moviegoers with tickets mark- 220 = (9 − 7)20 = ∑ ( )9i (−7)20−i = S1 − S2 ,
i=0 i
ing their respective assigned seats. The moviegoers line up to enter the cinema.
The first moviegoer did not know that the ticket shows an assigned seat, and thus 20 20
since if i is even then ( )9i 720−i = ( )9i (−7)20−i = S1 + S2 .
i i
chooses a random seat and sits there. The remaining moviegoers sit on their seat
1620 + 220
if it is empty — otherwise he/she chooses a random open seat to sit. What is the Adding the two equations together and dividing by two results in S1 = =
2
280 + 220
probability that the 432nd moviegoer sits at his/her assigned seat? = 279 + 219 . The required answer is 79 + 19 = 98.
2
Answer 50% SE1-D3 Problem Define the omega function ω(x) as the real number that satisfies x =

Solution The last unoccupied seat is either the seat meant for the last moviegoer

27 28
 ω(x ⋅ e x ), for real numbers x. Simplify There are 34 × 2 = 162 terms in this expansion; it is easy to see that these terms
are not like terms.
√
⋯ √
√
2
2
2
SE1-D5 Problem Find
iπ
sin
in terms of the omega function.
∞
(i + 1)!
∑ π π .
i=3 cos cos
Answer e ω(ln 2) i! (i + 1)!
√
√ 3
Solution Let x be the given quantity. Then x 2 = x ⇒ 21/x = x ⇒ x x = 2 ⇒ Answer
3
x ln x = ln 2. Let a = ln x, or e x = a. The equation becomes e a ⋅ a = ln 2 ⇒
Solution
ω(e a ⋅ a) = ω(ln 2) ⇒ a = ω(ln 2). Substituting back the expression for x, we get
iπ π π
ln x = ω(ln 2) ⇒ x = e ω(ln 2) . sin sin ( − )
∞
(i + 1)! ∞ i! (i + 1)!
∑ π π =∑ π π
SE1-D4 Problem How many terms on the 2017th row of the Pascal triangle, whose first i=3 cos cos i=3 cos cos
i! (i + 1)! i! (i + 1)!
row has the terms 1 and 1, are not divisible by 3? π π π π
∞ sin cos − cos sin
i! (i + 1)! i! (i + 1)!
Answer 162 =∑ π π
i=3 cos cos
i! (i + 1)!
Solution This question is equivalent to finding the number of coefficients in the ∞
π π
= ∑ (tan − tan )
expansion of (1 + x)2017 that are not divisible by 3. Note that (1 + x)3 ≡ 1 + x 3 i=3 i! (i + 1)!
(mod 3). Similarly, for k a power of 3, (1 + x)k ≡ 1 + x k (mod 3). √
π 3
Therefore, the series telescopes to tan = .
But 2017 = 729 × 2 + 243 × 2 + 27 × 2 + 9 × 2 + 1. Thus 3! 3

2 2 2
(1 + x)2017 ≡ ((1 + x)729 ) ((1 + x)243 ) ((1 + x)27 ) ⋅
2
((1 + x)9 ) (1 + x) (mod 3)

≡ (x 1458 + 2x 729 + 1)(x 486 + 2x 243 + 1)⋅

(x 54 + 2x 27 + 1)(x 18 + 2x 9 + 1)(x + 1) (mod 3)

29 30
SHS Finals Solution Either a is odd and b is even, or a is even and b is odd. Thus the sum is

π π π
sin + sin + sin 1 1 1
= ∑ a⋅ ∑ b+ ∑ a⋅ ∑ b
1 1
12 6 4 . ∑ a ⋅ 4b
SF-1 Problem Evaluate π π π a+b odd 3 a even 3 b odd 4 a odd 3 b even 4
cos + cos + cos 1 14 1 1
12 6 4 = 9 ⋅ + 3 ⋅ 16
√ 1 − 19 1 − 16 1 1 − 19 1 − 16 1
3
Answer 1 4 3 1 7
3 = ⋅ + ⋅ =
8 15 8 15 120
Solution

π 3π π 3π SF-3 Problem Let f ∶ Z+ ↦ Z+ be a piecewise function such that

π 2π 3π 2π + −
sin + sin + sin sin + 2 sin 12 12 cos 12 12 ⎧
12 12 12 = 12 2 2 ⎪
⎪
⎪
π 2π 3π π 3π π 3π ⎪
⎪ 1 n=1
cos + cos + cos + − ⎪
⎪
⎪
12 12 12 cos 2π + 2 cos 12 12 cos 12 12 ⎪ √
f (n) = ⎨ f ( n) n a perfect square .
12 2 2 ⎪
⎪
⎪
2π 2π −π ⎪
⎪
⎪
sin + 2 sin cos ⎪
⎪
⎪ f (n − 1) + 1 otherwise
= 12 12 12 ⎩
2π 2π −π
cos + 2 cos cos
12 12 12 What is the smallest n such that f (n) = 100?
2π −π √
sin (1 + 2 cos )
12 12 π 3 Answer 2024
= = tan =
2π −π 6 3
cos (1 + 2 cos )
12 12 Solution We start with a few lemmas.
Lemma 1. For all integers n ≥ 4, f (n) < n.
SF-2 Problem Find the sum of the reciprocals of all positive integers that can be ex-
pressed in the form 3a ⋅ 4b , where a and b are positive integers such that a + b is Proof. We proceed by strong induction. Note that f (4) = f (2) = 2 < 4. Assume
odd. for the sake of induction that f (k) < k is true for all integers k from 4 to an integer
7 m ≥ 4. By definition, if m+1 is not a perfect square, then f (m+1) = f (m)+1 < m+1,
Answer
120
and we are done.
√
On the other hand, if m + 1 is a perfect square, then f (m + 1) = f ( m + 1). How-

31 32
 √
ever, since m ≥ 4, m2 − m − 1 > 0. This means that m > m + 1. Therefore,
√ √
f (m + 1) = f ( m + 1) < m + 1 < m < m + 1, and we are done. The inductive
step is complete.
Thus, for all integers n ≥ 4, f (n) < n.
Lemma 2. For any positive integer n that can be expressed in the form
x2 ≤ n < (x + 1)2 , x∈ Z+ , and 0 ≤ y ≤ 2x, y ∈ Z≥ , f (n) = f (x) + y.
Proof. We proceed by induction on y. Note that if y = 0 f (n) = f (x 2 ) = f (x) =
f (x) + 0. Assume for the sake of induction that if n = x 2 + k for some integer k
between 0 and 2x − 1, then f (n) = f (x) + k. Then, since n + 1 is not a perfect
square, f (n + 1) = f (n) + 1 = f (x) + (k + 1), which completes the proof.
Lemma 3. For any positive integer n that can be expressed in the form x 2 + y, where
x 2 ≤ n < (x + 1)2 , x ∈ Z+ , and 0 ≤ y ≤ 2x, y ∈ Z≥ , f (n) ≤ 3x.
Proof. From Lemma 1 and Lemma 2, f (n) = f (x) + y ≤ x + 2x = 3x.
From Lemma 3, we see that for all n ≤ 332 + 2 ⋅ 33 = 1155, f (n) ≤ 3 ⋅ 33 = 99. Thus,
if f (n) = 100, and n = x 2 + y (with x and y as in Lemma 2), then x ≥ 34.
Consider all n = 492 + y = 2401 + y, where y is an integer from 0 to 98, inclusive.
Then f (n) = f (49) + y = f (7) + y = 5 + y = 100 ⇒ y = 95. This means that x = 49
is possible. Thus, if f (n) = 100, then 34 ≤ x ≤ 49.
However, if x = 34 or 35, then f (n) = f (x)+y ≤ 13+70 = 83 < 100, since 0 ≤ y ≤ 70,
f (34) = f (5) + 9 = 3 + 9 = 12 and f (35) = f (34) + 1 = 13. Thus no n exists for this
33
case.
If x ≥ 36, let x = 36 + p where p ∈ Z≥ . To minimize n, we need to minimize x, and
in turn, p.
We have f (n) = f (x) + y = f (36 + p) + y = f (6) + p + y = 4 + p + y = 100, due to
x2 + y, where Lemma 2. To minimize p, we maximize y. (Note that increasing x by a unit amount
leads to a greater increase in n than increasing y by a unit amount.) The maximum
possible y is y = 2x = 2(36 + p). Then,
4 + p + 2(36 + p) = 100 ⇒ 3p + 76 = 100 ⇒ p = 8
This implies that x = 36 + p = 36 + 8 = 44, y = 2x = 88, and n = x 2 + y = 442 + 88 =
2024.
i−1
SF-4 Problem Define {S i }∞
i=0 to be a sequence of sets such that S0 = {0}, and S i = ⋃ {S k },
k=0
the union of the set containing S0 , the set containing S1 , until the set containing
S i−1 . How many opening braces { are needed to write down the power set of S6 in
expanded form, if the empty set ∅ is to be written as { }?
Answer 2081
Solution Note that S0 = {0}, S1 = {S0 } = {{0}}, S2 = {S0 , S1 } = {{0}, {{0}}}, and
so on.
First, we show by strong induction that S i has exactly 2i opening braces. It is easy to
see that S0 has 1 opening brace and that S1 has 2. Now say that for all i between 0 to n,
S i uses 2i opening braces. Then for S n+1 = S0 , S1 , S2 , . . . , S n , we need one outermost
34
 opening brace, then add the number of opening braces for S0 up to S n . Thus the Say for some positive integer k, Fa+x = Fa Fx+1 + Fa−1 Fx is true for all x ≤ k. Then
number of opening braces needed for S n+1 is 1+(1+2+. . .+2n ) = 1+2n+1 −1 = 2n+1 .
Thus, we conclude that for all non-negative integers i, S i has exactly 2i opening Fa+x+1 = Fa+x + Fa+x−1

braces. = Fa Fx+1 + Fa−1 Fx + Fa Fx + Fa−1 Fx−1

The power set of S6 , denoted as 2S6 , is the set of all subsets of S6 . Thus = Fa (Fx + Fx+1 ) + Fa−1 (Fx−1 + Fx )

= Fa Fx+2 + Fa−1 Fx+1

2 = {{}, {S0 }, . . . , {S5 }, {S0 , S1 }, . . . , {S4 , S5 }, . . . , {S0 , . . . , S5 }}.
S6

Thus Fa+b = Fa Fb+1 + Fa−1 Fb for all positive integers a and b.

2S 6 has 26 elements, since each element of S6 can either be in a subset or not. First,
Next, we show by induction on n that Fm divides Fmn .
there are 1 + 26 = 65 opening braces used before expanding the S i ’s, since one out-
ermost opening brace is used, and for every subset of S6 , another opening brace is It is clear that Fm divides Fm .

used. For the inductive step, say for some positive integer i, Fi divides Fim . We first sub-

Since each S i are either in a subset or not, there is an equal number of subsets with stitute a = m and b = im in the previous identity. Then we have

S i and of subsets without S i , for all i. Thus, each S i appears 25 = 32 times. Since S i
F(i+1)m = Fim+m = Fm Fim+1 + Fm−1 Fim .
has exactly 2i+1 opening braces, the total number of opening braces in 2S6 is 65 +
32(1 + 2 + 4 + 8 + 16 + 32) = 65 + 32(63) = 65 + 2048 − 32 = 2081.
Thus, F(i+1)m is also divisible by Fm . Thus for all positive integers m and n, Fmn is
SF-5 Problem Prove that for all positive integers n and m, Fmn is divisible by Fm . Fk is the divisible by Fm .
kth Fibonacci number, defined by F1 = F2 = 1, Fk = Fk−1 + Fk−2 for integers k ≥ 3.

Solution

Proof. Extending the sequence, we see that F0 = 0. We first show, by inducting on

b, that Fa+b = Fa Fb+1 + Fa−1 Fb for positive integers a and b.

It is easy to see that Fa+1 = Fa + Fa−1 = Fa F2 + Fa−1 F1 .

35 36
Mathematical Results Pythagorean Theorem

△ABC is a right triangle with right angle at B iff AB2 + BC 2 = AC 2 . (p 9, 18, 19, 24)
Angle Bisector Theorem

If D is a point on BC of △ABC such that AD is an angle bisector of the triangle, Stewart’s Theorem

then AC ⋅ BD = AB ⋅ CD. (p 4) If D is a point on side BC in △ABC, then

Binomial Theorem AC 2 ⋅ BD + AB2 ⋅ DC = BC ⋅ (BD ⋅ DC + AD2 ).

For integer n,
n
n (p 4)
(x + y)n = ∑ ( )x k y n−k .
k=0 k

(p 4, 28) Vandermonde Identity

For non-negative integers m, n, and r,

Chinese Remainder Theorem
n+m r
m n
If r and s are coprime positive integers, then for all integers x and y, there exists an ( ) = ∑ ( )( ).
r k=0 k r−k
integer N such that N ≡ x (mod r) and N ≡ y (mod s). Furthermore there exists
only one N that satisfies this and is between 1 and rs. (p 6) (p 16)

Vieta’s Formulas
De Moivre’s Theorem
Let x1 , ⋯, x n be the n roots of the polynomial a n x n + a n−1 x n−1 + a n−2 x n−2 +⋯+ a1 x +
For any complex number z = r cis θ and integer n, z n = r n cis(nθ). (p 5)
a0 = 0. Also, let s i be the sum of all possible products of i x k terms where the k’s are
distinct. This means that s1 = x1 + ⋯ + x n , s2 = x1 x2 + x1 x3 + ⋯ + x1 x n + ⋯ + x n−1 x n ,
Pythagorean Identity
and so on until s n = x1 x2 ⋯x n . Then for all i = 1, 2, . . . , n,
For all θ, sin2 θ + cos2 θ = 1. (p 22)
(−1)i a n−i
si = .
an

37 38
 (p 7)

Disclaimer: Not all of the problems here are original. Some are lifted from, or based on,
other material. All information provided here is for educational purposes only.

39
 Contents
Junior Division 3
Junior Division Eliminations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

11th St. Stephen’s Lord of the Math

Easy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Average . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Difficult . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Solution Booklet Junior Division Team Finals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Junior Division Individual Finals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Senior Division 29
St. Stephen’s High School Senior Division Eliminations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Easy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Average . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Difficult . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Senior Division Semifinals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Senior Division Finals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

January 12, 2019

2
Junior Division Solution This is simply the number of seven-digit numbers, as a seven digit number cannot
have more than seven 7’s.
Junior Division Eliminations
JE-E4 Problem What is the probability that the number of letters (written in English) of a positive
Easy integer less than or equal to 20 is prime?
1 1
JE-E1 Problem If = 4, find the value of . 9
x +5 x +6 Answer
20
4 9
Answer Solution From the table, the probability is .
5 20
1 1 5 1 4
Solution = 4 ⇒ (x + 6) − 1 = ⇒ x + 6 = ⇒ = . Number # of Letters Number # of Letters
(x + 6) − 1 4 4 x +6 5
one 3 eleven 6
JE-E2 Problem Find the largest possible area of a quadrilateral with perimeter 8.
two 3 twelve 6
Answer 4 square units three 5 thirteen 8
Solution Let the four sides be a, b, c, d. From Bretschneider’s formula, the area of the four 4 fourteen 8
quadrilateral is five 4 fifteen 7
√ six 3 sixteen 7
α+β seven 5 seventeen 9
A= (4 − a)(4 − b)(4 − c)(4 − d) − abcd cos2 ,
2 eight 5 eighteen 8
α+β nine 4 nineteen 8
where α and β are opposite angles. Since abcd cos2 ≥ 0, then by AM-GM,
2 ten 3 twenty 6
4 JE-E5 Problem If A lies in the second quadrant and 3 tan A + 4 = 0, what is the value of 2 cot A −
(4 − a) + (4 − b) + (4 − c) + (4 − d) 16 − 8 4
A2 ≤ (4−a)(4−b)(4−c)(4−d) ≤ ( ) =( ) = 16. 5 cos A + sin A?
4 4
23
√ Answer
Therefore the maximum area is 16 = 4, which can be attained when the quadrilateral is a 10
4
square. Solution We know that tan A = − . Since A ∈ QII, sin A > 0 > cos A. Thus,
3
JE-E3 Problem How many seven-digit numbers have at most seven 7’s?
tan A 4
sin A = − √ =
Answer 9 000 000 1 + tan A 5
2

3 4
 √ √ √
1 3 has length 402 − 142 = 1404 = 6 39, so the distance between the two flagpoles is
cos A = − √ =− √ √
1 + tan A
2 5
2 × 6 39 = 12 39.
1 3
cot A = =−
tan A 4 JE-E8 Problem Find all ordered pairs of integers (a, b) satisfying (a 3 + a 2 − 1) − (a − 1)b = 0.
3 3 4 23
Therefore, 2 cot A − 5 cos A + sin A = 2 (− ) − 5 (− ) + = . Answer (0, 1) and (2, 11)
4 5 5 10
a3 + a2 − 1 1
Solution Note that a cannot be 1. Solving for b, we get b = = a 2 +2a+2+ .
JE-E6 Problem Find all real numbers a such that ∣x + ∣x∣ + a∣ + ∣x − ∣x∣ − a∣ = 2 has exactly three a−1 a−1
1
real solutions in x. Since has to be an integer, then a is either 0 or 2. If a = 0 then b = 1; if a = 2, then
a−1
b = 11.
Answer −1
JE-E9 Problem Suppose a soccer game ends with a score of 7-5. How many possible half-time
Solution Note that if x is a solution, then so is −x. Thus, in order to have an odd number
scores are there? (In soccer, the score is the number of goals each team scored.)
of solutions, x = −x ⇒ x = 0 has to be a solution.
Answer 48
Substituting x = 0 results in ∣a∣ + ∣ − a∣ = 2 ⇒ 2∣a∣ ⇒ ∣a∣ = ±1.
Solution There are 8 possible half-time scores for the first team, from 0 to 7, and 6 for the
If a = 1 and x ≥ 0 then ∣x + x + 1∣ + ∣x − x − 1∣ = ∣2x + 1∣ + 1 = 2 ⇒ x = 0, so there is only
second team, from 0 to 5. Therefore there are 8 × 6 possible half-time scores.
one solution in x.
JE-E10 Problem Let x = log17 tan 1○ + log17 tan 2○ + ⋯ + log17 tan 45○ and y = log17 tan 46○ +
If a = −1 and x ≥ 0, then ∣x + x − 1∣ + ∣x − x + 1∣ = ∣2x − 1∣ + 1 = 2 ⇒ ∣2x − 1∣ = 1 ⇒ x = 0, 1. x
log17 tan 47○ + ⋯ + log17 tan 89○ . What is ?
Then the three solutions of the original equation are x = 0, ±1. y

This means that a = −1 is the only possible value of a. Answer −1

JE-E7 Problem An 80 m rope is suspended at its two ends from the tops of two 50 m-tall flagpoles. Solution Note that log17 tan 45○ = log17 1 = 0. Also,

If the lowest point to which the midpoint of the rope can be pulled is 36 m from the ground,
log17 tan θ + log17 tan(90○ − θ) = log17 (tan θ tan(90○ − θ)) = log17 1 = 0.
then what is the distance between the flagpoles?
√
Answer 12 39 m Then log17 tan θ = − log17 tan(90○ − θ). This means that x = −y, leading to the desired
answer of −1.
Solution In order for the rope to be at the lowest possible point, that point must be the
middle of the rope. Thus, we are faced with solving a right-angled triangle with hypotenuse
40 m and one side of length 50 − 36 = 14 m. By the Pythagorean theorem, the third side

5 6
Average JE-A4 Problem Find k in k sin 18○ sin 42○ sin 78○ = cos 36○ .

JE-A1 Problem The angles A, B, and C of △ABC, where side x is opposite angle X, are in arith- Answer 4
metic progression. If 2b = 3c , determine the angle A.
2 2
Solution
○ 5π
Answer 75 or
12 sin 18○ sin 42○ sin 78○ = sin 18○ sin(60○ − 18○ ) sin(60○ + 18○ )
Solution Since A, B, and C are in arithmetic progression, A+B+C = 3B = 180○ ⇒ B = 60○ .
= sin 18○ ((sin 60○ cos 18○ )2 − (cos 60○ sin 18○ )2 )
Since 2b2 = 3c 2 , by the sine law,
3 1
√ = sin 18○ ( cos2 18○ − sin2 18○ )
b 3 sin 60○ sin B 4 4
=√ = √ = . 1
c 2 2/2 sin C = (3 sin 18 cos 18 − sin3 18○ )
○ 2 ○
4
1
Therefore, C has measure 45○ or 135○ . But 60○ + 135○ = 195○ > 180○ , so C has to be 45○ . = (3 sin 18○ (1 − sin2 18○ ) − sin3 17○ )
4
This means that A = 180○ − 60○ − 45○ = 75○ . 1
= (3 sin 18○ − 4 sin3 18○ )
4
JE-A2 Problem What is the largest prime number that divides (19! − 17!)? 1 1
= sin 54○ = cos 36○ .
Answer 31 4 4

Thus, k = 4.
Solution 19! − 17! = 17!(18 ⋅ 19 − 1) = 17!(341) = 17!(31)(11). Since the largest prime
factor of 17! is 17, the largest prime factor of 19! − 17! is 31. JE-A5 Problem For what values of x does

JE-A3 Problem What is the probability of getting a suit full house (three of a suit and two of 2+ 2
2+ x+1
another suit) when one draws a five-card hand from a standard deck of 52 playing cards? 2+ x+1
2+ 2
429 2+ x+1
Answer 2+ 2
4165 2+ 2
x+1
13 13
Solution There are ( ) ways to choose three cards of one suit and ( ) ways to choose
3 2 not have a real value?
two cards of two suit, both taking into account only the rank of the card. √
Answer −5, −2, −1, −7 ± 2 7
There are 4 × 3 ways to choose the suits of the three of a kind and the pair. Therefore, the
26
12 ⋅ (13 ) (13)  *⋅ 11 * 12 11 ⋅ 13 ⋅ 3
3 ⋅ 2 ⋅
12 286
 >⋅ 
78
 120 429
= = =

probability is 4 17 ⋅ 5 ⋅ 49
.
(52
5
) >

2
52 ⋅ >

17 5
51 ⋅ 
50 ⋅ 49 ⋅ 
>
 48
>
 4165

7 8
 Solution The denominator of a fraction cannot be zero. Thus we have value of her final score?
20
Answer
x + 1 ≠ 0 ⇒ x ≠ −1 91
2 Solution Let k be the expected value of her final score. If she rolls a 1 the game ends;
2+ ≠ 0 ⇒ 2x + 4 ≠ 0 ⇒ x ≠ −2
x +1 1
otherwise, she continues on, and since her score after the first roll will become (where
2+ x+1 √ x
2+ 2
≠ 0 ⇒ x 2 + 14x + 21 ≠ 0 ⇒ x ≠ −7 ± 2 7 k
2+ 2 x ≠ 1 is the number rolled), then the expected value from the second roll onward is .
x+1 x
x +1 1 1
2+ ≠ 0 ⇒ x + 5 ≠ 0 ⇒ x ≠ −5 Since the probability of rolling a specific number from 1 to 6 is , we have k = +
2 6 6
k 1 1 1 1 1 29 20
( + + + + ) ⇒ 6k = 1 + k ⇒ k = .
6 2 3 4 5 6 20 91
JE-A6 Problem Three congruent circles, centered at (0, 0), (1, 1), and (2, 1), have a common
tangent. Find the radius of the circles. Difficult
√
5 JE-D1 Problem Find all real numbers x, y, z satisfying the system of equations
Answer
10
⎧
⎪
Solution From an illustration it is clear that the common tangent is a common external ⎪
⎪
⎪ x + ⌊y⌋ + {z} = 1.1
⎪
⎪
⎪
tangent of the leftmost and rightmost circles, and a common internal tangent of the middle
⎨⌊x⌋ + {y} + z = 2.2
and rightmost circles. Then, the common tangent is parallel to the line passing through the ⎪
⎪
⎪
⎪
⎪
⎪
⎪{x} + y + ⌊z⌋ = 3.3
centers of the leftmost and rightmost circles, and passes through the midpoint of the line ⎩
segment connecting the middle and rightmost circles.
where ⌊k⌋ is the greatest integer less than or equal to k and {k} = k − ⌊k⌋.
The slope of the line passing through the centers of the leftmost and rightmost circles is 0.5.
Answer (x, y, z) = (0.1, 1.2, 2)
Thus, let the line be y = 0.5x +b, for some b. Since the line passes through (1.5, 1), we have
1 = 0.75 + b ⇒ b = 0.25. Therefore, the equation of the common tangent is 2x − 4y + 1 = 0. Solution Using the fact that {k} = k − ⌊k⌋, we can re-express the system as

∣2(0) − 4(0) + 1∣ ⎧
⎪
The radius of the circles is the distance of any of the centers to this line is √ = ⎪
⎪ x + ⌊y⌋ + z − ⌊z⌋ = 1.1 . . . (1)
22 + (−4)2 ⎪
⎪
⎪
√ ⎪
⎨⌊x⌋ + y − ⌊y⌋ + z = 2.2 . . . (2)
1
√ =
5
. ⎪
⎪
⎪
⎪
⎪
⎪
20 10 ⎪ x − ⌊x⌋ + y + ⌊z⌋ = 3.3 . . . (3)
⎩
JE-A7 Problem Jane rolls a fair, standard six-sided die repeatedly until she rolls a 1. She begins
with a score of 1, and each time she rolls x, her score is divided by x. What is the expected (1) + (2) − (3) results in 2⌊x⌋ + 2z − 2⌊z⌋ = 0 ⇒ ⌊x⌋ + z − ⌊z⌋ = 0. This implies that z is an

9 10
 integer, or z = ⌊z⌋ ⇒ ⌊x⌋ = 0 ⇒ 0 ≤ x < 1. in the region [−π, 2π], where ⌊x⌋ is the greatest integer less than or equal to x.
−π −π
Equation (1) then becomes x + ⌊y⌋ = 1.1. In other words, 0 ≤ 1.1 − ⌊y⌋ < 1. The only Answer (−π, ) ∪ ( , 0)
2 2
integer ⌊y⌋ that satisfies this is ⌊y⌋ = 1. This implies that x = 1.1 − 1 = 0.1, and 1 ≤ y < 2.
Solution The function f is defined if f1 (x) = (∣x − 1∣ − ⌊x⌋)−1/2 , f2 (x) = csc−1 ⌊sin x⌋ and
√
Equation (2) now becomes 0 + y − 1 + ⌊z⌋ = y + z − 1 = 2.2 ⇒ y = 3.2 − z ⇒ 1 ≤ 3.2 − z < 2. f3 (x) = sin−1 (1 + ⌊∣x∣⌋) are all defined.
The only integer z that satisfies this is z = 2. Then y = 3.2 − z = 1.2.
Now, f1 is defined when ∣x − 1∣ > ⌊x⌋. This is true for all x < 1. Therefore, f1 is defined for
Therefore, the only solution is (x, y, z) = (0.1, 1.2, 2). x ∈ [−π, 1).
JE-D2 Problem According to Benford’s Law, for a set of numbers chosen in some specified pro- We know that csc−1 θ is defined when ∣θ∣ ≥ 1. Thus, csc−1 ⌊sin x⌋ exists when either ⌊sin x⌋ ≥
cedure, the event that the leftmost digits of a number written in base 10 (removing leading π
1 or ⌊sin x⌋ ≤ −1. ⌊sin x⌋ ≥ 1 ⇒ x = ; meanwhile, ⌊sin x⌋ ≤ −1 implies x ∈ (π, 2π) ∪
n+1 2
zeros) is n ∈ Z+ has probability log10 ( ). If the probability that the second digit from (−π, 0).
n
the left (removing leading zeros) of a set of numbers that satisfy Benford’s Law is 2 can be √
Finally, for any x ∈ R, ∣ sin x∣ ≤ 1. Now, when ∣ sin x∣ < 1, then ⌊∣ sin x∣⌋ = 0 ⇒ ⌊∣ sin x∣⌋ =
written in the form log10 a − log10 b, where a and b are positive coprime integers, express π √
0 ⇒ f3 = sin−1 1 = . On the other hand, when ∣ sin x∣ = 1, then sin−1 (1 + ⌊∣ sin x∣⌋) =
ab as a product of powers of prime numbers. 2
π 3π
sin−1 2, which does not exist. Therefore, f3 is defined for x ∈ [−π, 2π] ∖ {± , }.
Answer 218 ⋅ 41 ⋅ 43 ⋅ 53 ⋅ 73 ⋅ 83 2 2
We get the intersection of the three sets to get the domain of definition:
Solution The probability that the second digit is 2 can be computed by adding the proba-
bilities that the first two digits are 10x + 2, from x = 1 to x = 9. Thus, π π 3π
[−π, 1) ∩ ({ } ∪ (−π, 0) ∪ (π, 2π)) ∩ ([−π, 2π] ∖ {± , })
2 2 2
⎛⋅
13 ⋅
23  ⋅ 43 ⋅ 53 ⋅ 
33 63  ⎞
 ⋅ 73 ⋅ 83 ⋅ 
93 π 3π
= (−π, 0) ∩ ([−π, 2π] ∖ {± , })
P(second digit is 2) = log10
⎝ 12

4
>⋅ 22
>
2
⋅ 32 ⋅ 42

2
>⋅ 52

4
>⋅ 
2
>⋅ 
62

8
>⋅ 82 ⋅ 
72
 > ⎠
92

4 2 2
 π
43 ⋅ 53 ⋅ 73 ⋅ 83 = (−π, 0) ∖ {− }
= log10 ( ) 2
218 ⋅ 41

Therefore, a = 43 ⋅ 53 ⋅ 73 ⋅ 83 and b = 218 ⋅ 41. Their product, then, is 218 ⋅ 41 ⋅ 43 ⋅ 53 ⋅ 73 ⋅ 83. JE-D4 Problem If
log2 log1/2 log2 x = log5 log1/5 log5 y = log7 log1/7 log7 z,
JE-D3 Problem Find the domain of definition of the function
√ arrange x, y, z in increasing order.
f (x) = (∣x − 1∣ − ⌊x⌋) −1/2
+ csc ⌊sin x⌋ + sin (1 + ⌊∣ sin x∣⌋)
−1 −1
Answer z < y < x

11 12
 Solution Without loss of generality set them all to 0. Then log1/2 log2 x = 1 ⇒ log2 x = Solution Let m2 = x 2 + 19x + 88. Then the equation x 2 + 19x + 88 − m2 = 0 must have
1
⇒ x = 21/2 , and similarly, y = 51/5 and z = 71/7 . But x 10 = 32 > 25 = y10 and integral solutions in x, that is, the determinant 192 − 4(88 − m2 ) must be a perfect square,
2
y35 = 78 125 > 16 807 = z 35 . Therefore, z < y < x. say n2 . The equation 192 − 4(88 − m2 ) = n2 simplifies to (n − 2m)(n + 2m) = 9. Consider
the following cases:
JE-D5 Problem Triangle ABC is an equilateral triangle. A point D is randomly chosen in the
triangle. What is the probability that triangle ABD is obtuse? Case 1. n − 2m = 3 and n + 2m = 3. Then n = 3 and m = 0.
√
9 + 3π Case 2. n − 2m = −3 and n + 2m = −3. Then n = −3 and m = 0.
Answer
18
Case 3. n − 2m = 9 and n + 2m = 1. Then n = 5 and m = −2.
Solution Triangle ABD is obtuse if and only if it is in the interior of the semicircle shown
in the figure. Case 4. n − 2m = 1 and n + 2m = 9. Then n = 5 and m = 2.

C Case 5. n − 2m = −9 and n + 2m = −1. Then n = −5 and m = 2.

Case 6. n − 2m = −1 and n + 2m = −9. Then n = −5 and m = −2.

G F
Either m2 = 0 or m2 = 4. If m2 = 0, x 2 +19x+88 = 0 simplifies to x = −8 or −11. Meanwhile,
if m2 = 4, x 2 + 19x + 84 = 0 simplifies to x = −12 or −7. Therefore, the possible values of x
B A are −7, −8, −11, −12.
E
JTF-2 Problem Four nickels and six dimes are tossed, and the total number N of heads is ob-
Thus, the desired probability is the area of the intersection of the semicircle
√ and △ABC, served. If N = 4, what is the probability that exactly two nickels showed up heads?
3 2 √
divided by the area of △ABC. Let AE = r. Then the area of △ABC is (4r ) = 3r 2 . 3
4 Answer
7
The area of the intersection is equal to the area of equilateral triangles CGE, AFE, and the
4 1 6 1
[△ABC], and
minor sector GF. The total area of the equilateral triangles is one-half of√ ( ) 4( ) 6
√ the 3
π 1 π 1 π/24 9 3π 9 + 3π Solution The probability is 2 2 2 2 = .
area of the sector is ⋅ = . The probability is then + √ = + = 10 1 7
6 4 24 2
. ( ) 10
3/4 18 18 18 4 2
JTF-3 Problem What is the value of tan 101○ + tan 124○ + tan 101○ tan 124○ ?
Junior Division Team Finals
Answer 1
JTF-1 Problem Find all integer values of x such that x 2 + 19x + 88 is a perfect square.

Answer −7, −8, −11, −12

13 14
 Solution We know that tan 225○ = 1. Thus, C

tan 101○ + tan 124○

1 = tan 225○ =
1 − tan 101○ tan 124○
1 − tan 100 tan 125 = tan 100○ + tan 125○
○ ○
A B
M
tan 100 + tan 125 + tan 100 tan 125 = 1.
○ ○ ○ ○

JTF-4 Problem Let ω be a non-real cube root of unity. Express (1 + ω)19 as a linear function of The area of the desired region is the difference between the area of △APB, which is right-
ω. angled since PB is a tangent of circle A, and the area of the√
minor sector AMP. Now,
√ √ √ 1 √ 3
Answer 1 + ω PB = AB2 − AP 2 = 4 − 1 = 3, so [△APB] = ⋅ 1 ⋅ 3 = . Also, the area of sector
2 2 √
Solution Since ω3 = 1 and ω ≠ 1, we have 1 2 π π
AMP is ⋅ 1 ⋅ = . Therefore, the area of the desired region is
3 π
− .
2 3 6 2 6
ω3 − 1 = 0 ⇒ (ω − 1)(ω2 + ω + 1) = 0 ⇒ 1 + ω = −ω2 . JTF-6 Problem How many kilograms of water must be evaporated from 50 kilograms of a 3% salt
solution so that the remaining solution will be 5% salt?
Therefore, (1 + ω)19 = −ω38 = −ω36 ⋅ ω2 = −ω2 = 1 + ω.
Answer 20 kilograms
JTF-5 Problem Each side of equilateral △ABC has length 2 units. A unit circle centered at A cuts
Solution The salt in the solution is 1.5 kg. This is 5% of the new solution, which will be
AB at M. A tangent to the circle from B and lying outside the triangle meets the circle at 1.5 kg
= 30 kg. Thus 20 kg shall be evaporated.
P. Find the area of the region bounded by BP, BM, and minor arc MP. .05
√ 1
Answer
3 π
− JTF-7 Problem What is the value of − 2 sin 70○ ?
2 6 2 sin 10○
Solution Refer to the following figure: Answer 1

Solution

1 1 1 − 4 cos 20○ sin 10○

− 2 sin 70○ = − 2 cos 20○ =
2 sin 10 ○ 2 sin 10○ 2 sin 10○
1 − 2 sin 30 + 2 sin 10
○ ○
2 sin 10○
= = =1
2 sin 10○ 2 sin 10○

15 16
JTF-8 Problem A convex equilateral heptagon has angles that measure 168○ , 108○ , 108○ , 168○ , Charges Combinations Rearrangements Product
x ○ , y○ , and z ○ , in clockwise order. What is y? ---- 2 ⋅ 2 ⋅ 2 ⋅ 2 = 16 1 16
---0 2 ⋅ 2 ⋅ 2 ⋅ 15 = 120 4 480
Answer 132
---+ 2 ⋅ 2 ⋅ 2 ⋅ 3 = 24 4 96
Solution Let P be a point in the heptagon such that ABCDP is a regular pentagon. Refer --00 2 ⋅ 2 ⋅ 15 ⋅ 15 = 900 6 5400
to the following figure: --0+ 2 ⋅ 2 ⋅ 15 ⋅ 3 = 180 12 2160
F -000 2 ⋅ 15 ⋅ 15 ⋅ 15 = 6750 4 27 000

y ○ The sum, then, is 16 + 480 + 96 + 5400 + 2160 + 27 000 = 33 352.

E G
P JTF-10 Problem Solve the equation sec x cos 5x + 1 = 0, 0 < x < 2π.
60 ○ ○ 60 ○ π π 3π 5π 5π 7π 7π 11π
108 Answer { , , , , , , , }
D 108○ 6 4 4 6 4 6 4 6
108○ A
Solution The equation is equivalent to cos 5x = − cos x = cos(π − x), cos x ≠ 0. Thus
108○ 108○ 5x = 2nπ ± (π − x), where n is an integer.
(2n + 1)π (2n − 1)π
C B If 5x = 2nπ + π − x, then x = ; if 5x = 2nπ − π + x, then x = . The
6 4
π π 3π 5π 5π 7π 7π 11π
Since ∠DEP = ∠GAP = 60○ , it follows that △DEP and △GAP are equilateral. Then values of x in the interval (0, 2π) are { , , , , , , , }.
6 4 4 6 4 6 4 6
EP = PG = EF = FG, that is, EFGP is a rhombus. This means that y○ = ∠EPG = JTF-11 Problem A space diagonal of a polyhedron is a line segment connecting two vertices of
360○ − 108○ − 2(60○ ) = 132○ . the polyhedron and is in the interior of the polyhedron. A dodecahedron is a polyhedron
JTF-9 Problem There are 20 different amino acids in the human body, three of which have a consisting of twelve pentagons such that three pentagons meet at a vertex. How many space
positive charge (+1), 2 have a negative charge (−1) and the rest have no charge (0). A protein diagonals does it have?
is a ordered sequence of amino acids whose charge is equal to the sum of the charges of its Answer 100
amino acids. How many proteins with negative charge are there that are four amino acids
Solution The 12 pentagons give 60 sides and 60 vertices. Since three faces meet at a vertex,
long?
the dodecahedron has 60 ÷ 3 = 20 vertices. Furthermore, two faces meet at an edge, so
Answer 33 352 20
the dodecahedron has 60 ÷ 2 = 30 edges. Among the 20 vertices, there are ( ) = 190
2
Solution We tabulate the combinations as follows: line segments that can be formed. For each pentagonal face, there are 5 diagonals, so the
dodecahedron has 5 × 12 = 60 face diagonals. Any line segment that is not an edge or a face

17 18
 diagonal has to be a space diagonal; thus, there are 190 − 60 − 30 = 100 space diagonals. sum of the positive factors of the square of the integer is 2821. Find the square of the sum
of the positive factors of the integer.
JTF-12 Problem What is the value of the nonnegative integer n that satisfies n! = 112 2962 −
79 8962 ? Answer 5184

Answer 13 Solution Note that 1300 = 5 × 10 × 26 = (20 + 22 )(30 + 32 )(50 + 52 ) and 2821 = 7 × 13 × 31 =
(20 + 21 + 22 )(30 + 31 + 32 )(50 + 51 + 52 ). Thus the integer is 2 × 3 × 5 = 30. The sum of the
Solution
positive factors of 30 is (20 + 21 )(30 + 31 )(50 + 51 ) = 3 × 4 × 6 = 72, and the square of that
112 2962 − 79 8962 = (112 296 − 79 896)(112 296 + 79 896) is 722 = 5184.

= (32 400)(192 192) = 182 102 ⋅ 192 ⋅ 1001 JTF-15 Problem Two chimpanzees are playing a variation of tic-tac-toe. Instead of stopping when
= 2 ⋅ 9 ⋅ 3 ⋅ 6 ⋅ 2 ⋅ 5 ⋅ 10 ⋅ 2 ⋅ 8 ⋅ 12 ⋅ 7 ⋅ 11 ⋅ 13 someone has formed a line, they continue and fill up the whole 3 × 3 grid. A chimpanzee
wins if and only if it is able to form a line and the other is unable to. The game ends in a
= 2 ⋅ 3 ⋅ 2 ⋅ 2 ⋅ 5 ⋅ 6 ⋅ 7 ⋅ 8 ⋅ 9 ⋅ 10 ⋅ 11 ⋅ 12 ⋅ 13
draw if either both or none of them form a line. Assuming these two chimpanzees have
= 13!
an equal chance of picking any of the empty squares available, and that a chimpanzee won,
what is the probability that the first chimpanzee won?
JTF-13 Problem Solve for real numbers x and y in 4x 3 + 3x 2 y + y3 = 8, 2x 3 − 2x 2 y + x y 2 = 1. 31
√
3
√ √ √ Answer
25 3 3 25 3
100 2 3 100 37
Answer (1, 1), ( , ),( , )
5 5 10 5 Solution Denote the first player’s symbol as O and the second player’s symbol as X . Con-
Solution Note that neither x = 0 nor y = 0 are solutions. Let y = mx. Then we have sider the following sets of outcomes. Note that the second player cannot form two lines
x (4 + 3m + m ) = 8 and x (2 − 2m + m ) = 1. Eliminating x from this system results in
3 3 3 2 3 since the second player only has four turns.
4 + 3m + m2 = 8(2 − 2m + m2 ) ⇒ m3 − 8m2 + 19m − 12 = 0 ⇒ (m − 1)(m − 3)(m − 4) = set of outcomes where...
0 ⇒ m = 1, 3, 4.
A nobody forms a line
If m = 1, then x (2 − 2m + m ) = x = 1 ⇒ x = 1 ⇒ y = 1.
3 2 3
B the first player forms two lines
√ √
3
25 3 3 25 C the first player forms exactly one line and the second player does not form a line
If m = 3, then x (2 − 2m + m ) = 5x = 1 ⇒ x =
3 2 3
⇒y= .
5 5 D the second player forms exactly one line and the first player does not form a line
√3
√
100 2 3 100 E both players form one line
If m = 4, then x 3 (2 − 2m + m2 ) = 10x 3 = 1 ⇒ x = ⇒y= .
10 5
JTF-14 Problem The sum of the squares of the positive factors of a positive integer is 1300, and the

19 20
 ∣B∣ + ∣C∣ 22 + 40 62 31
The probability, then, is = = = .
A ∣B∣ + ∣C∣ + ∣D∣ 22 + 40 + 12 74 37
B
E D
C

∣B∣ + ∣C∣
The desired answer is .
∣B∣ + ∣C∣ + ∣D∣
First, we find ∣D∣. The only way for the second player to win is if it forms a diagonal line.
There are two ways to form a diagonal, and there are six remaining spots to select the last
X . All the other spots will be filled with O ’s. This means that ∣D∣ = 2 × 6 = 12.

If the first player completes two lines in six different ways (not yet counting reflections or
rotations), as shown.

O O O O O O O O O O O O O
O O O O O O O O O O
O O O O O O O
With rotations or reflections, there are a total of ∣B∣ = 4 + 4 + 1 + 8 + 4 + 1 = 22 ways to place
the O ’s.

Next, consider the case where the first player makes exactly one line and the second player
does not form a line. The following are the ways (without accounting yet for reflections or
rotations) for the O ’s:

O O O O O O O O O O O O O
O O O O O O O O O O
O O O O O O O
Taking rotations and reflections into account, there are a total of ∣C∣ = 4+8+4+8+8+8 = 40
configurations.

21 22
Junior Division Individual Finals Solution The first equation implies c = 100 − a − b. Substituting into the second, we have

JIF-1 Problem Find the maximum value of 4x − x 4 , where x is a real number.

a2 + b2 + (100 − a − b)2 = 1917
Answer 3 a 2 + b2 + 10 000 + a 2 + b2 − 200a − 200b + 2ab = 1917
Solution Consider x 4 − 4x + 3 = (x 4 − 2x 2 + 1) + (2x 2 − 4x + 2) = (x 2 − 1)2 + 2(x − 1)2 ≥ 0. 2(a 2 + b2 + ab − 100a − 100b + 5000) = 1917
This implies 4x − x 4 ≤ 3.
Since the variables are all integers, the left-hand side is always even and thus cannot be equal
JIF-2 Problem Find the area of the convex quadrilateral whose vertices are A(1, 2), B(3, 0),
to 1917, an odd number.
C(2, 4), and D(5, 3).
17 JIF-4 Problem A and B are two events such that P(A∁ ) = 0.3, P(B) = 0.4, and P(A ∩ B∁ ) = 0.5.
Answer What is the value of P (B ∣ A ∪ B∁ )?
2
Solution Answer 0.25
W C(2, 4) X
Solution

P(B ∩ (A ∪ B∁ )) P(B ∩ A)
D(5, 3) P (B ∣ A ∪ B∁ ) = =
P(A ∪ B )∁ P(A) + P(B∁ ) − P(A ∩ B∁ )
A(1,
A(1,
A(1,
A(1, 2)
2)
2)
A(1,2)
2) P(A) − P(A ∩ B∁ )
A(1,
A(1, 2)
2) =
2 − P(A∁ ) − P(B) − P(A ∩ B∁ )
1 − P(A∁ ) − P(A ∩ B∁ )
=
2 − P(A∁ ) − P(B) − P(A ∩ B∁ )
1 − 0.3 − 0.5 0.2
Z B(3, 0) Y = = = 0.25
2 − 0.3 − 0.4 − 0.5 0.8
2
From the figure, [ABCD] = [W XY Z] − [WAC] − [CXD] − [DBY] − [BZ A] = 16 − −
2 JIF-5 Problem Which point on the circle (x − 11)2 + (y − 13)2 = 116 is farthest from the point
3 6 4 15 17
− − = 16 − = . (41, 25)?
2 2 2 2 2
JIF-3 Problem Find all ordered triples of positive integers (a, b, c), a ≤ b ≤ c, that satisfy a+b+c = Answer (1, 9)
100 and a 2 + b2 + c 2 = 1917.
Solution The farthest point lies on the intersection of the line passing through the center of
√
Answer {} the circle (11, 13) and (41, 25). Note that the distance from the center to (41, 25) is 6 29

23 24
 √
while the radius of the circle is 2 29. Since the difference between the x and y coordinates cients, excluding the constant, is 4096 − 581 = 3515.
30 12
are 30 and 12, respectively. Then the farther intersection is (11 − , 13 − ) = (1, 9).
3 3 JIF-9 Problem If the quadratic equation
√ √
JIF-6 Problem Simplify sin4 x + 4 cos2 x − cos4 x + 4 sin2 x. 2 1
4sec α x 2 + 2x + (β 2 − β + ) = 0
2
Answer cos 2x
has real roots in x, then what are the possible values of cos α + cos−1 β?
Solution Since sin2 x and cos2 x are both in the interval [0, 1], the expression simplifies to
π
√ √ √ √ Answer ±1
sin4 x + 4 cos2 x − cos4 x + 4 sin2 x = sin4 x − 4 sin2 x + 4 − cos4 x − 4 cos2 x + 4 3
2 1
= (2 − sin2 x) − (2 − cos2 x) Solution The discriminant of the equation is 4 − 4 ⋅ 4sec (β 2 − β + ). If the discriminant
α
2
1
= cos2 x − sin2 x = cos 2x sec2 α 2
is non-negative, then we have 4 (β − β + ) ≤ 1. However, sec2 α ≥ 1 ⇒ 4sec α ≥ 4
2
2
1 1 2 1 1 2 1
and β 2 − β + = (β − ) + ≥ . This means that 4sec α (β 2 − β + ) ≥ 1, implying
JIF-7 Problem Let n = 24 35 46 67 . How many positive integer factors does n have? 2 2 4 4 2
1 2
sec α = 1 and (β − ) = 0.
2
Answer 312 2
1 π π
Solution We have n = 223 312 . Thus it has (23 + 1)(12 + 1) = 24 ⋅ 13 = 312 factors. Then cos α = ±1 and cos−1 β = cos−1 = , so cos α + cos−1 β = ± 1.
2 3 3
JIF-8 Problem Find the sum of all the coefficients of the terms, excluding the constant, in the JIF-10 Problem If z1 , z2 , and z3 are three complex numbers such that
2 6
expansion of (1 + x + ) . 1 1 1
x ∣z1 ∣ = ∣z2 ∣ = ∣z3 ∣ = ∣ + + ∣ = 1,
z1 z2 z3
Answer 3515

Solution The constant term is find ∣z1 + z2 + z3 ∣.

6! 0 6! 1 6! 2 6! 3 Answer 1
2 + 2 + 2 + 2 1 1 1
6!0!0! 4!1!1! 2!2!2! 0!3!3! Solution We have ∣z1 ∣ = 1 ⇒ ∣z12 ∣ = 1 ⇒ z1 z1 = 1 ⇒ = z1 . Similarly = z2 and = z3 .
=1+6⋅5⋅2+6⋅5⋅3⋅4+5⋅4⋅8 z1 z2 z3
1 1 1
Thus ∣ + + ∣ = ∣z1 + z2 + z3 ∣ = ∣z1 + z2 + z3 ∣ = ∣z1 + z2 + z3 ∣ = 1.
= 1 + 60 + 360 + 160 = 581. z1 z2 z3
JIF-11 Problem What is the length of the longest median of a triangle whose sides have length 4,
2 6
The sum of all coefficients is (1 + 1 + ) = 46 = 4096. Therefore, the sum of all the coeffi- 7, and 9?
1

25 26
 √
Answer 61 JIF-14 Problem If the answer to this question is a real number x, find the value of

Solution The longest median is the one to the shortest side. Using Apollonius’s theorem, ∞ k k− j
k!x −k
we have 72 + 92 = 2(22 + x 2 ), where x is the length of the median. The equation simplifies ∑∑∑ .
√ k=0 j=0 i=0 i! j!(k − i − j)!
to x 2 = 61 ⇒ x = 61.

JIF-12 Problem If a and b are positive integers that leave respective remainders of 6 and 1 when Answer 4
divided by 14, and x is a positive integer solution to x − 2ax + b = 0, find the remainder
2 k k− j
k!
Solution We know that ∑ ∑ = 3k , as it is the number of ways to distribute
when x is divided by 14. j=0 i=0 i! j!(k − i − j)!
k items into three groups (any of which can be empty).
Answer 13
∞
3 k 1
Thus, the given equation is equivalent to x = ∑ ( ) = . The only real number x
Solution Let a = 14m + 6, b = 14n + 1, x = 14p + k for some integers m, n, p, k and 1 − x3
k=0 x
0 ≤ k ≤ 13. Then x 2 − 2ax + b = (14p + k)2 − 2(14m + 6)(14p + k) + (14n + 1) = that satisfies this is x = 4.
(196p2 + 28pk + k 2 ) − 2(196mp + 84p + 14mk + 6k) + (14n + 1) = 14q + k 2 − 12k + 1, for JIF-15 Problem Let D be a point inside acute △ABC such that ∠ADB = ∠ACB+90○ and AC⋅BD =
some integer q. Now k 2 − 12k + 1 and k 2 − 12k − 13 leave the same remainder when divided AB ⋅ CD
AD ⋅ BC. Find .
by 14, and k 2 − 12k − 13 = (k + 1)(k − 13) = 0. Then k = 13 so that x 2 − 2ax + b is 0 (a AC ⋅ BD
√
multiple of 14). Thus, the remainder when x = 14p + k = 14p + 13 is divided by 14 is 13. Answer 2

JIF-13 Problem For any positive integer n, let d(n) be the sum of its digits. Find n if n + d(n) = Solution Let E be the point on the plane such that DE is perpendicular to DB and intersects
1 000 000 000. AB, and DE and DB have the same length.

Answer 999 999 932 ∠ADB = ∠ACB + 90○ implies ∠ADE = ∠ACB. Also, since DE = DB, the given condition
AD BD DE
AC ⋅ BD = AD ⋅ BC implies = = . By SAS, △ADE ≈ △ACB, which means
Solution Since d(n) > 0, we have n ≤ 999 999 999, so n has at most 9 digits. This means DE BC CB
AE AB
that d(n) ≤ 9 × 9 = 81, or n ≥ 999 999 919 ⇒ n = 999 999 900 + 10x + y, for some digits = .
AD AC
x and y. In terms of x and y, d(n) = 63 + x + y. We now have n + d(n) = (999 999 900 +
AE AB
10x + y) + (63 + x + y) = 1 000 000 000, or 11x + 2y = 37. The only digits that satisfy this Now, ∠CAB = ∠DAE ⇒ ∠CAD = ∠BAE. Together with = , this implies
AD AC
are x = 3, y = 2, so we have n = 999 999 932. △AEB ∼ △ADC. Therefore, since △DEB is an isosceles right triangle,
√
AB BE 2BD AB ⋅ CD √
= = ⇒ = 2.
AC CD CD AC ⋅ BD

27 28
Senior Division Solution The desired points are of the form (x, x) or (x, −x), where x ∈ {−9, −8, . . . , 8, 9}∖
{0}. Thus, we have 18 × 2 = 36 points.
Senior Division Eliminations x +1 x −1
SE-E5 Problem Solve for real x: tan−1 ( ) + tan−1 ( ) = tan−1 (−7)
Easy x −1 x
Answer no solution
SE-E1 Problem How many 4-digit numbers with nonzero digits are divisible by 4 but not by 8?
Solution Take the tangent of both sides of the equation:
Answer 729
x +1 x −1
Solution For any three-digit number ABC with nonzero digits, exactly one of ABC2, +
x −1 x = −7
ABC4, ABC6, and ABC8 is divisible by 4 but not by 8. Therefore, there are 93 = 729 of x +1 x −1
1− ⋅
them. x −1 x
2x 2 − x + 1
= −7
SE-E2 Problem What is the probability that a positive integer less than 100 is prime? 1−x
25 2x 2 − x + 1 = −7(1 − x)
Answer
99 2x 2 − 8x + 8 = 0
Solution There are 99 positive integers less than 100, and there are 25 prime numbers from
2(x − 2)2 = 0 ⇒ x = 2
1 to 99: 2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, 71, 73, 79, 83, 89,
25
97. Thus the probability is . 1
99 However, the left-hand side when evaluated at x = 2 gives tan−1 3 + tan−1 , which is
2
SE-E3 Problem A car ran five full laps on a circular track whose radius is 20 km, for 1 hour at a positive. On the other hand, tan−1 (−7) < 0. Therefore x = 2 is a degenerate solution, and
uniform speed. Find the average velocity of the car. the given equation has no solutions.
Answer 0 kph ∣z∣2 − ∣z∣ + 1
SE-E6 Problem Find all possible values of z ∈ R that satisfy the inequality log√3 < 2.
2 + ∣z∣
Solution Since after five laps the car returned to its original position, the displacement is
Answer z ∈ (−5, 5)
zero and so is the velocity.
∣z∣2 − ∣z∣ + 1
SE-E4 Problem How many lattice points (x, y) are there such that both ∣x∣ and ∣y∣ are less than Solution The inequality is equivalent to 0 < < 3. The left-hand side is always
2 + ∣z∣
10, x is divisible by y, and y is divisible by x? true, so we focus on the right-hand side. Also, 2 + ∣z∣ > 0, so the inequality becomes

Answer 36 ∣z∣2 − ∣z∣ + 1 < 6 + 3∣z∣

29 30
 ∣z∣2 − 4∣z∣ − 5 < 0 3 21 AC
SE-E8 Problem In △ABC, tan A = and tan B = . Find the ratio .
4 20 BC
−1 < ∣z∣ ∧ ∣z∣ < 5
35
Answer
∣z∣ < 5 29
AC sin B tan A tan A 3/4
Solution From sine law, we have = . Then, sin A = =√ = =
1 1 BC sin A sec A 1 + tan A 5/4
2
SE-E7 Problem Suppose A1 A2 A3 ⋯A n is an n-sided regular polygon such that = + 3 tan B 21/20 21 AC 3/5 35
A1 A2 A1 A3 . Similarly, sin B = √ = = . Then = = .
1
. Find n.
5 1 + tan2 B 29/20 29 BC 21/29 29
A1 A4 SE-E9 Problem Find the coefficient of x 98 in the product (x + 1)(x + 2)(x + 3)⋯(x + 100).
Answer 7
Answer 12 582 075
Solution Let O be the circumcenter of the regular polygon. It is clear that ∠A1 OA2 =
360○ 2π Solution The coefficient of x 98 is the sum of the products of 1, 2, . . . , 100 taken two at a
. Let OA1 = r. Using the cosine law on △A1 OA2 , we have (A1 A2 )2 = 2r 2 (1 − cos ),
n n time. Therefore, the coefficient of x 98 is
π
or A1 A2 = 2r sin . Similarly, using the cosine law on △A1 OA3 and △A1 OA4 results in
n 2
2π 3π 1 ⎛ 100 100 ⎞
1 100 ⋅ 101 2 100 ⋅ 101 ⋅ 201
A1 A3 = 2r sin and A1 A4 = 2r sin , respectively. (∑ i) − ∑ i 2 = (( ) − )
n n 2 ⎝ i=1 i=1 ⎠ 2 2 6
Substituting into the given equation and removing common factors, we have 1
= (25 502 500 − 338 350) = 12 582 075.
2
1 1 1
= +
sin πn sin 2π sin 3π
n n SE-E10 Problem Square ABCD is inscribed in a unit circle. Let P be the intersection of line AB
1 1 1
− = and the tangent of the circle at C. Find the length of segment PD.
sin πn sin 3π sin 2π
n n √
π 3π Answer 10 units
3π π sin n sin n
sin − sin = 2π √
n n sin n
Solution Since the circle has radius 1, then the sides of the square have length 2. Since
π 2π 2π π 3π
2 sin cos sin = sin sin PC ⊥ AC, ∠PCB = 45○ ; also, PA ⊥ BC implies ∠CPB = 45○ . Therefore, △PBC is
n n n n n √
π 4π 3π = BC = 2. Furthermore, since △APD is a right triangle, we have PD =
isosceles, so PB √
sin (sin − sin ) = 0 √ √ √ √
n n n AP 2 + AD2 = (2 2)2 + ( 2)2 = 10.

4π 3π
The only integer n > 1 that satisfies this is n = 7, as =π− .
n n

31 32
Average SE-A3 Problem If no three diagonals of a convex decagon meet at a point, into how many line
segments are the diagonals divided by their intersections?
SE-A1 Problem Let x = sin2 A. Express sin A sin 2A sin 3A sin 4A as a polynomial in x, in general
form. Answer 455
10
Answer −64x 5 + 144x 4 − 104x 3 + 24x 2 Solution The total number of diagonals is ( )−10 = 35. For every four distinct vertices of
2
the decagon one can find exactly one intersection in the interior, as the decagon is convex.
Solution The given expression is equal to 10
Thus, the total number of intersections is ( ) = 210.
4
sin A sin 2A sin 3A sin 4A = sin A(2 sin A cos A)(3 sin A − 4 sin3 A)(2 sin 2A cos 2A) Since each intersection point produces two new segments, the 210 intersections add 420
= 2 sin2 A cos A sin A(3 − 4 sin2 A)4 sin A cos A(1 − 2 sin2 A) new segments. Adding the 35 original segments corresponding to the diagonals, there is a
= 8 sin A cos A(3 − 4 sin A)(1 − 2 sin A)
4 2 2 2 total of 420 + 35 = 455 segments.

= 8 sin4 A(1 − sin2 A)(3 − 4 sin2 A)(1 − 2 sin2 A) SE-A4 Problem Let a, b, and c be integers from 0 to 9, inclusive. How many triples (a, b, c) are
= 8x (1 − x)(3 − 4x)(1 − 2x) = 8x (3 − 13x + 18x − 8x )
2 2 2 3 there such that the three-digit number abc is a prime and the function f (x) = ax 2 + bx + c
= 24x 2 − 104x 3 + 144x 4 − 64x 5 has at least one rational zero?

Answer 21
SE-A2 Problem What is/are the real value(s) of x that satisfy the equation
Solution If a ≠ 0, then set f (10) = 100a + 10b + c. It is given that this prime, yet it can be
2 2 2 3 3 3 2018 2018 2018 2019 −2 factored into a product of two integers greater than 1, since f (x) has at least one rational
(x 2 π 2 + e 2 )(x 2 π 2 + e 2 )(x 2 π 2 + e 2 )⋯(x 2 π2 + e2 ) = e2 ?
zero. Therefore, there is a contradiction, and we only need to consider a = 0.
c
Answer 0 If a = 0, as long as b ≠ 0, we always have a rational zero − . Since there are 21 primes
b
between 10 and 99, there are also 21 triples (a, b, c) that satisfy both conditions.
Solution We multiply both sides by (x 2 π 2 − e 2 ) and simplify to get x 2
2019 2019 2019
π2 − e2 =
e 22019 −2
(x π − e ) = x π e
2 2 2 2 2 22019 −2
−e 22019 −2+2
. Then x 22019
π 22019
=x π e 2 2 22019 −2
, implying SE-A5 Problem Gretel has six paper clips, labeled 1 to 6, two cardboard boxes, and a fair die.
She first puts all six paper clips into the first cardboard box. She then rolls the die, and
2019 −2 2019 −2 2019 −2 2019 −2 2019 −2
x 2 π 2 (x 2 π2 − e2 ) = x 2 π 2 ((xπ)2 − e2 ) = 0. she moves the paper clip whose number is shown face-up on the die, from the box it is
currently in, to the other box. She rolls the die and moves paper clips repeatedly until both
e
Then either x = 0 or x = ± (as the exponent 22019 − 2 is even). It can be verified through boxes have exactly 3 paper clips. On average, how many times will she toss the die before
π
e
substitution that x = 0 is a solution, and x = ± are not. stopping?
π

33 34
 23 (25 + 10x + x 2 ) + 64(25 − 10x + x 2 ) = 65(25 − x 2 )
Answer
5
130x 2 − 630x = 0
Solution Let E i denoted the expected number of tosses before having both boxes with
equal paper clips given that the the first box originally had i paper clips, i ∈ {6, 5, 4, 3}. 63
Then x = 0 or .
The problem is finding E6 . We have the following system: 13

⎧ SE-A7 Problem Find the largest possible value of the five-digit number PU MaC in the cryptarithm
⎪
⎪
⎪ E6 =
23
⎧
⎪ ⎪
⎪ shown below. Here, identical letters represent the same digits and distinct letters represent
⎪
⎪
⎪ E6 = 1 + E5 ⎪
⎪
⎪
5
⎪
⎪
⎪ ⎪
⎪
⎪ distinct digits.
⎪
⎪
⎪
5
E5 = 1 + E4 +
1 ⎪
⎪
⎪
⎪
⎪
⎪
E6 ⎪
⎪
⎪ 18 N I M O
⎪ ⎪E5 = 5
⎪
6 6
⎪
⎨ ⇒⎨ + H M M T
⎪
⎪
⎪ ⎪
⎪
⎪
⎪
⎪
⎪
2 1 ⎪
⎪
⎪
⎪
⎪ E4 = 1 + E3 + E5 ⎪
⎪
P U M a C
⎪
⎪
⎪ 3 3 ⎪
⎪
⎪ E4 =
11
⎪
⎪
⎪ ⎪
⎪
⎪
⎪
⎩ E = 0 ⎪
⎪
5
3
⎪
⎪
⎪ E = 0
Answer 16 485
⎩ 3
Solution It is clear that P = 1. Now U cannot be 9 or 8; otherwise, U will be equal to at
23 least one of N or H. Let U = 7, so {N , H} = {8, 9}. Without loss of generality, let N = 9,
Therefore, the expected number of tosses is .
5
H = 8.
SE-A6 Problem Solve for real values of x: 9 I M O
√ √ √ + 8 M M T
(5 + x)2 + 4 3 (5 − x)2 = 5 25 − x 2
3 3

1 7 M a C

63 Since the addition in the hundreds place does not have a carry-over to the thousands place,
Answer 0 and
13 we conclude that I = 0. Then M is either 2 or 3.
Solution We take the cube of both sides to get
If M = 3, we have:
√ √ 9 0 3 O
(5 + x)2 + 64(5 − x)2 + 12(25 − x 2 )2/3 ( (5 + x)2 + 4 (5 − x)2 ) = 125(25 − x 2 )
3 3

√ + 8 3 3 T
(5 + x)2 + 64(5 − x)2 + 12(25 − x 2 )2/3 (5 25 − x 2 ) = 125(25 − x 2 )
3

1 7 3 6 C
(5 + x)2 + 64(5 − x)2 + 60(25 − x 2 ) = 125(25 − x 2 )
The remaining digits are 2, 4, and 5, but no two of the three add up to the third.
(5 + x)2 + 64(5 − x)2 = 65(25 − x 2 )

35 36
 If M = 2, we have: Solution The expression is equal to
9 0 2 O
23 23
+ 8 2 2 T cot ∑ cot−1 (n2 + n + 1) = cot ∑ tan−1 (
1
)
n=1 n=1 n2 + n + 1
1 7 2 4 C
23
(n + 1) − n
The remaining digits are 3, 6, and 5, but no two of the three add up to the third. = cot ∑ tan−1
n=1 n(n + 1) − 1
Next, we consider U = 6. Then either {N , H} = {9, 7} or {N , H} = {8, 7}. Say H = 7. 23
= cot ∑ (tan−1 (n + 1) − tan−1 (n))
If N = 9, then again I = 0. The largest possible value of M is 4, so we have
n=1

= cot(tan−1 24 − tan−1 1)
9 0 4 O 1
=
+ 7 4 4 T tan(tan−1 24 − tan−1 1)
1 + tan tan−1 24 tan tan−1 1
1 6 4 8 C =
tan tan−1 24 − tan tan−1 1
1 + 24 25
The remaining digits are 2, 3, and 5, so C = 5. Then PU MaC = 16 485. = =
24 − 1 23
If N = 8, then I = 9 as there is a carry-over. The only possible value for M is 5. Then we
√ √
have 3−1 3+ 3
SE-D2 Problem In △ABC, cos A cos B cos C = and sin A sin B sin C = . Find
8 9 5 O 8 8
+ 7 5 5 T tan A tan B + tan A tan C + tan B tan C
.
1 6 5 0 C tan A + tan B + tan C
The remaining digits are 2, 3, and 4, but no two of the three add up to the third. √
9−2 3
Since for U = 6 there is only one solution, we conclude that the largest possible value for Answer
3
PU MaC is 16 485. Solution A non-right triangle
√ satisfies
√ tan A + √
tan B + tan C = tan A tan B tan C. Thus,
3+ 3 3−1 3+ 3 √
tan A + tan B + tan C = ÷ =√ = 3 + 2 3.
Difficult 8 8 3−1
23 n Consider
SE-D1 Problem Evaluate cot ∑ cot−1 (1 + ∑ 2k).
n=1 k=1

25 cos(A + B + C) = cos((A + B) + C)
Answer
23

37 38
 = cos(A + B) cos C − sin(A + B) sin C Answer 107
= cos A cos B cos C − sin A sin B cos C − sin A cos B sin C − cos A sin B sin C Solution Let Q(x) be the expected value of P(x). By linearity of expectation, the expected
= cos A cos B cos C(1 − tan A tan B − tan A tan C − tan B tan C) value of P(4) is Q(4). Now Q(x) is also a polynomial of degree at most 3. We have
√ 0 + 50
3−1 Q(O) = E(y0 ) = = 25
But cos(A + B + C) = cos(180○ ) = −1 and cos A cos B cos C = . Thus tan A tan B + 2
8
1 √ 40 + 60 + 80
tan A tan C + tan B tan C = √ + 1 = 5 + 4 3. Q(1) = E(y1 ) = = 60
( 3 − 1)/8 3
√ √ 10 + 40 + 70 + 80
Q(2) = E(y2 ) = = 50
5+4 3 9−2 3 4
Therefore, the desired expression is √ = .
3+2 3 3 10 + 30 + 40 + 70 + 90
Q(3) = E(y3 ) = = 48
54 5
SE-D3 Problem How many prime factors does ∑ (−1)⌊(n+2)÷3⌋ n2 have?
n=1 Since Q(x) is a polynomial of degree at most 3, R(x) = Q(x) − Q(x − 1) is of degree at
Answer 3 most 2, S(x) = R(x) − R(x − 1) is of degree at most 1, and T(x) = S(x) − S(x − 1) is
constant. Then
Solution We group every six terms:

54 9 2 T(4) = S(4) − S(3) = (R(4) − R(3)) − (R(3) − R(2))

∑ (−1) n = ∑ ∑((6n − i)2 − (6n − 5 + i)2 )
⌊(n+2)÷3⌋ 2
n=1 n=1 i=0 = R(4) − 2R(3) + R(2)
9 2
= (Q(4) − Q(3)) − 2(Q(3) − Q(2)) + (Q(2) − Q(1))
= ∑ ∑(12n − 5)(5 − 2i)
n=1 i=0 = Q(4) − 3Q(3) + 3Q(2) − Q(1) = Q(4) − 144 + 150 − 60
9 2
= ( ∑ (12n − 5)) (∑(5 − 2i)) = Q(4) − 54.
n=1 i=0

= (6(9)(10) − 5(9))(9) = 55(81) = 5 ⋅ 11 ⋅ 34 Similarly, T(3) = Q(3) − 3Q(2) + 3Q(1) − Q(0) = 48 − 150 + 180 − 25 = 53. Since
T(4) = T(3), then Q(4) − 54 = 53 ⇒ Q(4) = 107.
Therefore the number has 3 prime factors.
1
SE-D5 Problem A graph is defined in polar coordinates by r(θ) = cos θ + . Find the smallest
SE-D4 Problem Let y0 be chosen uniformly randomly from {0, 50}, y1 from {40, 60, 80}, y2 from 2
x-coordinate of a point on the graph.
{10, 40, 70, 80}, and y3 from {10, 30, 40, 70, 90}. Let P(x) be a polynomial with degree at
1
most 3 such that P(i) = y i for i ∈ {0, 1, 2, 3}. Find the expected value of P(4). Answer −
16

39 40
 Solution For polar coordinates, we have Solution This is a direct application of the Chicken McNugget theorem: (13 − 1)(12 − 1) −
1 = 131.
1 cos θ
x = r cos θ = (cos θ + ) cos θ = cos2 θ + 5
(2i − 1)π
2 2 SSF-4 Problem Simplify ∏ tan .
cos θ 1 1 i=1 20
= cos2 θ + + −
2 16 16 Answer 1
1 2 1
= (cos θ + ) − Solution
4 16

1 1 π 3π 5π 7π 9π π π π 3π π 3π π
Hence, the smallest x-coordinate is − , when θ = cos−1 . tan tan tan tan tan = tan tan ( − ) tan tan ( − ) tan
16 4 20 20 20 20 20 20 2 20 20 2 20 4
π π 3π 3π
= tan cot tan cot =1
Senior Division Semifinals 20 20 20 20

SSF-1 Problem The sum of squares of deviations of 10 observations from the mean 50 is 250. SSF-5 Problem Find the monic polynomial, with rational coefficients and of least degree, such
What is the coefficient of variation? Express as a percentage. that √ √ √ √
Answer 10% 1+ 2+ 1+ 2+⋯
√
250
Solution The standard deviation of the data is σ = = 5. Thus, the coefficient of is one of its zeros.
10
5
variation is = 10%. Answer x 4 − 2x 2 − x − 1
50
SSF-2 Problem Let ⌊k⌋ denote the largest integer not exceeding k ∈ R. If x and y are real numbers Solution Let x be the above expression. Then
such that y = 2⌊x⌋ + 3 = 3⌊x − 2⌋ + 5, find ⌊x + y⌋. √ √
x= 1+ 2+x
Answer 15 √
x2 = 1 + 2 + x
Solution Let ⌊x⌋ + {x} = x. Then y = 2⌊x⌋ + 3 = 3⌊x⌋ − 1. This implies ⌊x⌋ = 4, and √
x2 − 1 = 2 + x
y = 2 × 4 + 3 = 11. Then ⌊x + y⌋ = ⌊x⌋ + y = 4 + 11 = 15.
x 4 − 2x 2 + 1 = 2 + x
SSF-3 Problem In a market eggs are available by sets of 12’s or 13’s. What is the largest number x 4 − 2x 2 − x − 1 = 0
of eggs that cannot be bought as a combination of 12’s or 13’s?

Answer 131 Since x 4 − 2x 2 − x − 1 is not factorable, the degree cannot be reduced.

41 42
SSF-6 Problem If 1, ω, ω2 are the three cube roots of unity, find Consider an arbitrary point (a, b, c). The distance of this point to (t, t, 1) is

2018 √ √
∏ (1 − ω + ω ).
j 2j (a − t)2 + (b − t)2 + (c − 1)2 = 2t 2 − 2(a + b)t + (a 2 + b2 + (c − 1)2 )
¿
j=1
Á a + b 2 (a + b)2
Á
À
= 2 (t − ) − + (a 2 + b2 + (c − 1)2 )
2 2
Answer 21346
a+b
Solution When j is divisible by 3, ω j = ω2 j = 1, which implies 1 − ω j + ω2 j = 1 − 1 + 1 = 1. This is minimized when t = . Then, the distance of (a, b, c) to the line x − y = 0, z = 1
2
When j is not divisible by 3, ω j ≠ 1. Since ω3 j = 1, we have a+b a+b
is the distance of (a, b, c) to ( , , 1). Similarly, the distance of (a, b, c) to the
2 2
a−b a−b
ω3 j − 1 = 0 ⇒ (ω j − 1)(ω2 j + ω j + 1) = 0 ⇒ ω2 j + ω j + 1 = 0 ⇒ 1 − ω j + ω2 j = −2ω j . line x + y = 0, z = −1 is the distance of (a, b, c) to ( , , −1). Therefore, the locus
2 2
of points can be solved by the equation
Therefore,
√ √
x+y 2 x+y 2 x−y 2 x−y 2
2018 ⎛ 671 3 ⎞ (x − ) + (y − ) + (z − 1)2 = (x − ) + (y − ) + (z + 1)2 .
∏ (1 − ω + ω ) = ∏ ∏(1 − ω + ω2 j+3i ) (1 − ω2017 + ω4034 )(1 − ω2018 + ω4036 )
j 2j j+3i 2 2 2 2
j=1 ⎝ i=0 j=1 ⎠
This simplifies to x y + 2z = 0.
⎛ 672 3 ⎞
= ∏ ∏(1 − ω j + ω2 j ) (−2ω2017 )(−2ω2018 )
⎝ i=1 j=1 ⎠ SSF-8 Problem How many paths of shortest length are there from (0, 0, 0) to (3, 3, 6) if one can
672 only go up, forward, or right by integral units, and the path can only pass through points
= (∏(−2ω)(−2ω2 )(1)) 4ω4035 (x, y, z) that satisfy x + y ≤ z?
i=1

= 4672 ⋅ 4 = 4673 = 21346 Answer 2640

SSF-7 Problem Find the locus (equation describing the set) of points (x, y, z) that are equidistant
to the lines x − y = 0, z = 1, and x + y = 0, z = −1.
Answer x y + 2z = 0
Solution The two lines can be parametrized as ℓ1 ∶ (t, t, 1) and ℓ2 ∶ (t, −t, −1), for all t ∈ R.
Solution We find the number of ways using a diagram. For any point P we place a number
#(P) denoting the number of ways one can go there from (0, 0, 0) that satisfy the above
condition. Thus #(3, 3, 6) is the answer. Now, if at least one of x, y, or z is negative, then
#(x, y, z) = 0; #(0, 0, 0) = 1, and for all other points #(x, y, z) = #(x − 1, y, z) + #(x, y −
1, z) + #(x, y, z − 1). Now we compute the values up to (3, 3, 6).

43 44
 Solution Consider (x+1)k , and let c i be the coefficient of x i in its expansion, i ∈ {0, 1, . . . , k}.
Then if we substitute x = ±1 in (x + 1)k = c0 + c1 x + ⋯ + c k x k , we have

⎧
⎪
z=6 ⎪
⎪2 = c0 + c1 + ⋯ + c k
k

z=5 ⎨ ⇒ c0 + c2 + c4 + ⋯ = 2k−1
⎪
⎪
⎪0 = c0 − c1 + ⋯ + (−1)k c k
z=4 ⎩
z=3
On the other hand, substituting x = i gives
z=2
z=1
kπ kπ
z=0 2k/2 (cos + i sin ) = (1 + i)k = c0 + ic1 − c2 − ic3 + c4 + ⋯ + i k c k
4 4
5
kπ
5 15 We then take the real parts of both sides to get 2k/2 cos = c0 − c2 + c4 + ⋯. This implies
2 4
1 k−1 kπ
1 2 4 3 10 15 a k = c0 + c4 + c8 + ⋯ = (2 + 2k/2 cos ).
2 4
1 1 1 1 2 2 1 3 5 5 Thus, a2019 − 2a2018 = (22017 − 21008 ) − 2 ⋅ 22016 = −21008 .
z=0 z=1 z=2 z=3
SSF-10 Problem Determine all functions f ∶ R ∖ {0, 1} → R satisfying the relation f (x) +
14 56 28 168 420 48 360 1320 2640 1 2(1 − 2x)
f( )= for all values of x ∈ R ∖ {0, 1}.
1−x x(1 − x)
9 42 84 14 84 252 420 20 144 540 1320
x +1
Answer f (x) =
x −1
4 18 42 56 5 28 84 168 6 40 144 360
1 1 2(1 − 2x) 2 2 2
Solution Let y = ⇒ x = 1− . Then f (x)+ f (y) = = − = −2y.
1 4 9 14 1 5 14 28 1 6 20 48 1−x y x(1 − x) x 1 − x x
z=4 z=5 z=6 1 2 2 1 2
Also, f (y) + f ( )= − . Let z = . Then f (y) + f (z) = − 2z.
1− y xy 1 − y 1− y y
Therefore, there are 2640 paths.
1 1 1
k Note, however, that y = and z = together imply x = . Then as above,
SSF-9 Problem Let a k the sum of the coefficients of x 4n , where n is an integer from 0 to , 1−x 1− y 1−x
4 2
inclusive, in the expansion of (x + 1)k . Find a2019 − 2a2018 . f (z) + f (x) = − 2x.
z
Answer −21008

45 46
 Therefore, we have Thus,

1 1 1
f (x) + f (y) + f (z) = + + −x−y−z 29032019 − 8032019 − 4642019 + 2612019 + 2019 ≡ 2019 (mod 1897)
x y z
1 1 1 ≡ 122 (mod 1897)
f (x) = − + − x − y + z
x y z
1 x 1 1
=  − (1
−
x) + −
x− + (1
− ) SSF-12 Problem In a class of 10 students, the probability that exactly i (i from 0 to 10) students
x x −1 1−x x
x +1 passed an exam is directly proportional to i 2 . If a student is selected at random, find the
f (x) =
x −1 probability that s/he passed the exam.
11
Answer
SSF-11 Problem Find the remainder when 2903 2019
− 803
2019
− 464
2019
+ 261
2019
+ 2019 is divided 14
by 1897. Solution Denote by P(N i ) = λi 2 the probability that exactly i students passed, where λ is
the constant of proportionality. Then
Answer 122
5 7
Solution Note that 1897 = 7 ⋅ 271. Now, 10 10 >⋅ 11 ⋅ 
10
 21
>
 1
1 = ∑ P(N i ) = ∑ λi = λ 2
= 385λ ⇒ λ =

.
i=0 i=0 6
385
2903 − 803 = 2100 ∣ 29032019 − 8032019
From law of total probabilities, the probability that student A passed is
464 − 261 = 203 ∣ 4642019 − 2612019
7 ∣ 29032019 − 8032019 − 4642019 + 2612019 10 10
i i2 1 10 3
P(A) = ∑ P (A ∣ N i ) P(N i ) = ∑ ⋅ = ∑i
i=0 i=0 10 385 3850 i=0
1 10 ⋅ 11 2 11
Similarly, = ( ) = .
3850 2 14
2903 − 464 = 2439 = 271(9) ∣ 29032019 − 4642019
SSF-13 Problem Find the largest prime factor of the sum of the products of the nonzero digits of
803 − 261 = 542 ∣ 8032019 − 2612019
the positive integers less than 1000.
271 ∣ 29032019 − 8032019 − 4642019 + 2612019
Answer 103

Solution Denote by p(n) the product of the nonzero digits of n. We first append two
zeroes to the front of one-digit numbers and one zero to the front of two-digit numbers.

47 48
 Now, the sum of the products of the digits of the transformed positive integers less than SSF-15 Problem What is the value of
1000, zero or otherwise, is equal to ∞ ∞
m2 n
∑∑ ?
m=1 n=1 3m (n3m + m3n )
(0 ⋅ 0 ⋅ 0 + 0 ⋅ 0 ⋅ 1 + ⋯ + 9 ⋅ 9 ⋅ 9) − 0 ⋅ 0 ⋅ 0 = (0 + ⋯ + 9)(0 + ⋯ + 9)(0 + ⋯ + 9) − 0 ⋅ 0 ⋅ 0
= (0 + ⋯ + 9)3 − 03 9
Answer
32
The sum of the products of the nonzero digits can be attained by changing all the zeroes 3n
Solution Let S be the given sum, and a n = . Then
above into ones. Therefore, the sum is 46 −1 = (46−1)(46 +46+1) = 45⋅2163 = 3 ⋅5⋅7⋅103,
3 2 3
n
so the largest prime factor is 103. ∞ ∞ ∞ ∞
1 1
S= ∑∑ = ∑∑
SSF-14 Problem Define the sum of any two points (x1 , y1 ), (x2 , y2 ) on the Cartesian plane to be m=1 n=1 a m (a m + a n ) m=1 n=1 a n (a m + a n )
the point (x1 + x2 , y1 + y2 ). Furthermore, for two sets A and B, define their Minkowski
Adding both summations results in
sum to be the set A + B = {a + b ∣ a ∈ A, b ∈ B}. Let A = {(x, y) ∈ R2 ∣ x 2 + y2 = 1} and
B = {(x, y) ∈ R2 ∣ ∣x∣ ≤ 2, ∣y∣ ≤ 2}. Find the area of the region enclosed by A + B. ∞ ∞
1 1 ∞ ∞
1
2S = ∑ ∑ ( + )= ∑ ∑
a m (a m + a n ) a n (a m + a n ) m=1 n=1 m a n
a
Answer (12 + π) square units m=1 n=1
∞ 2 ∞ ∞ ∞ 2 2
1 n 1
Solution We translate A around the Cartesian plane such that the center of the circle is in = (∑ ) = (∑ n ) = ( ∑ ∑ n )
a
n=1 n n=1 3 m=1 n=m 3
B. Then the region enclosed by A + B is as shown in the following figure: 1 2 1 2
⎛∞ ⎞ ∞ 2 ⎛ ⎞
=⎜ 3m ⎟ = ( 3 1 3
∑ m) = ⎜
3 ⎟
⎜∑ 1 ⎟ 2 3 ⎜2 ⋅ 1⎟
⎝m=1 1 − ⎠ m=1 ⎝ 1− ⎠
3 3
B 3 2 9 9
=( ) = ⇒S=
4 16 32
A
Senior Division Finals
Thus, the area of A+B is equal to the sum of the area of square B, the four rectangles around
SF-1 Problem It is known that the only rational solution to the equation 10x +11x +12x = 13x +14x
B, and four quarter-circles on the corners: 2(2) + 4(1)(2) + π(12 ) = 12 + π.
is x = 2. Find all irrational solutions.

Answer There are no irrational solutions.

49 50
 Solution Dividing both sides of the equation by 13x ≠ 0 gives us express it as the sum of Fn and y = x − Fn . Note that since Fn < x < Fn+1 , 0 < y < Fn+1 − Fn =
Fn−1 . Since y < Fn , from the induction hypothesis, we can express it as a sum of one or
10 x 11 x 12 x 14 x
( ) +( ) +( ) =1+( ) . more non-consecutive positive Fibonacci numbers. On the other hand, since y < Fn−1 , the
13 13 13 13
representation of y will not contain Fn−1 . Thus adding Fn to the representation of y (to yield
The left-hand side is a decreasing function of x, and the right-hand side is an increasing x) will work. Therefore the proof is complete.
function of x; therefore, the equation must have at most one solution, which is x = 2. Thus,
√
there are no irrational solutions. SF-4 Problem Solve the equation x 3 − 3x = x + 2, where x is a real number.
2 1 2 √
SF-2 Problem The three altitudes of a triangle have lengths , , and . Find the inradius of −1 − 5 4π
9 5 17 Answer 2, , 2 cos
the triangle. 2 7
Solution It is clear that x ≥ −2. When x > 2, we have x 3 − 4x = x(x 2 − 4) > 0 ⇒ x 2 − 3x > x,
Answer
1 √ √
18 and x 2 − x − 2 = (x − 2)(x + 1) > 0 ⇒ x > x + 2. Then x 3 − 3x > x > x + 2. Hence,
Solution Let a, b, and c be the lengths of the sides corresponding to the altitudes with x ∈ [−2, 2].
2 1 2 a
respective lengths , , and . Then, if A is the area of the triangle, we have A = = Then let x = 2 cos α, there α ∈ [0, π]. We have
9 5 17 9
b c 1
= , so a = 9A, b = 10A, and c = 17A. Then the semiperimeter is s = (9A + 10A = √
10 17 2 8 cos3 α − 6 cos α = 2(cos α + 1)
17A) = 18A. Since the area of a triangle is equal to the product of its inradius r and its √
1 α
semiperimeter, then s = 18A = 18sr ⇒ r = . 2 cos 3α = 4 cos2
18 2
α
SF-3 Problem Prove that any positive integer can be expressed as a sum of one or more positive cos 3α = cos
2
Fibonacci numbers, no two of which are consecutive. (1 and 2 are considered consecutive
a a
Fibonacci numbers.) Then either 3a − or 3a + is a multiple of 2π. When 0 ≤ a ≤ π, we have x = 2 cos 0 = 2,
2 √ 2
Solution 4π −1 − 5 4π
x = 2 cos = , and x = 2 cos .
5 2 7
Proof. We show this is true by induction. For n = 1, 2, and 3 this is true as all three are SF-5 Problem A tetromino is a shape composed of four congruent squares such that each square
themselves Fibonacci numbers. Also we see that it is true for 4 = 1 + 3. It is obvious that this shares a side with at least one of the other three. How many ways can one tile a 2 × 18 block
is true for all Fibonacci numbers, so we consider numbers that are not. Now we show that with 9 tetrominoes, such that the tetrominoes fully fill the block and do not overlap?
if this is true for all integers less than a Fibonacci number Fn , then it is true for all integers Answer 3025
less than the next one, Fn+1 . Say we have a number x between Fn and Fn+1 . Then we can
Solution First, we establish a lemma.

51 52
Lemma 1. The squares of the Fibonacci numbers, F12 = F22 = 1, Fn2 = (Fn−1 + Fn−2 )2 (n > 2), shaded figures show the remaining area after placing the tetrominoes specified, and the
satisfy the recurrence relation number inside denotes the number of ways possible to tile the shaded area.
2n−2 2n−4
Fn2 = 2Fn−1
2
+ 2Fn−2
2
− Fn−3
2
. II
a n−1 OO a n−2
II
Proof. From the definition of the Fibonacci numbers, we have Fn = Fn−1 + Fn−2 and Fn−1 =
2n−2 2n−4
Fn−2 + Fn−3 ⇒ Fn−3 = Fn−1 − Fn−2 . Then,
2n−3 2n−1

Fn2 = Fn−1
2
+ 2Fn−1 Fn−2 + Fn−2
2 JJ
b n−2 b n−2
= Fn−1
2
+ Fn−2
2
+ (Fn−1
2
− Fn−1
2
) + (Fn−2
2
− Fn−2
2
) + 2Fn−1 Fn−2 LL
2n−1 2n−3
= 2Fn−1
2
+ 2Fn−2
2
− (Fn−1 − Fn−2 )2
= 2Fn−1
2
+ 2Fn−2
2
− Fn−3
2
Therefore, a n = a n−1 + a n−2 + 2b n−2 . Shifting the base by one unit results in a n−1 = a n−2 +
a n−3 + 2b n−1 . Subtracting the two and combining like terms, we get a n = 2a n−1 − a n−3 +
Denote the following letters to the seven tetrominoes. 2(b n−1 − b n−2 ).

Similarly, given a block corresponding to b n , there are two ways to fill the two upper-rightmost
II cells.
L J
OO SS ZZ 2n−2 2n−3
T
LL II
a n−1 b n−1
First, note that the tiling may not have T, S, or Z tetrominoes, because if one of these tiles
are placed anywhere in the 2 × 18 block, the number of squares on both the left sides of the 2n−2 2n−1

tetromino will be odd and thus cannot be filled up by other tetrominoes. The tiling can only
Thus, b n = a n−1 + b n−1 , or b n − b n−1 = a n−1 . Shifting the base by one unit, we have b n−1 −
consist of I, O, L or J tetrominoes.
b n−2 = a n−2 . This can now be substituted to the equation for a n , giving a n = 2a n−1 + 2a n−2 −
Let a n denote the number of ways to tile a 2 × 2n block with tetrominoes, and b n denote the a n−3 .
number of ways to tile a block that results from removing the two lower-leftmost squares
Note that this recurrence means that a n are squares of Fibonacci numbers, since a0 = a1 = 1
from a 2 × (2n + 1) block. Thus, we are finding a9 .
and a2 = 4. Therefore, from Lemma 1, a n = Fn+1
2
. When n = 9, a9 = F10
2
= 552 = 3025.
Given a block corresponding to a n , there are four ways to fill the rightmost column. The

53 54
Mathematical Results Cosine Law

AM-GM Inequality For △ABC, c 2 = a2 + b2 − 2ab cos C. (p 31)

For any positive real numbers a1 , a2 , . . . , a n ,

Law of Total Probabilities
a1 + a2 + ⋯ + a n √
≥ n a1 ⋅ a2 ⋯a n . If B1 , B2 , . . . , B n are non-empty sets such that any two are disjoint and B1 ∪ B2 ∪⋯∪ B n = B,
n
then
n
Equality holds iff a1 = a2 = ⋯ = a n . (p 3) P(A) = ∑ P (A ∣ B n ) P(B n ).
i=1
Apollonius’s Theorem
(p 48)
On △ABC, if D is the midpoint of BC, then AB + AC = 2(AD + BD ). (p 27)
2 2 2 2

Pythagorean Theorem
Bretschneider’s Formula
△ABC is a right triangle with right angle at B iff AB2 + BC 2 = AC 2 . (p 5)
The area of any quadrilateral with side lengths a, b, c, and d (a and c, and b and d opposite)
is Sine Law
sin A sin B sin C
1√ 2 2 For △ABC, = = . (p 7, 32)
A= 4p q − (b2 + d 2 − a 2 − c 2 )2 a b c
4
√
= (s − a)(s − b)(s − c)(s − d) − abcdx,

α+β
where p and q are the lengths of the diagonals, x = cos2
, α and β are a pair of opposite
2
a+b+c+d
angles, and s = is the semiperimeter. (p 3)
2

Chicken McNugget Theorem

For two coprime numbers p and q, the greatest integer that cannot be written in the form
ap + bq where a and b are nonnegative integers is (p − 1)(q − 1) − 1. (p 42)
Disclaimer: Not all of the problems here are original. Some are lifted from, or based on, other
material. All information provided here is for educational purposes only.

55 56