Maths Olympiad Prep

Track / Stage 6 / 44 of 400 #1044 of 1964

Problem 1044

National olympiad, first round
Algebra Difficulty 6.0 Prove it

## Task 2 - 211242

Two persons A and B play the following game:

In the system of equations

a1x+b1y+c1z=1a2x+b2y+c2z=1a3x+b3y+c3z=1 \begin{aligned} & a_{1} x+b_{1} y+c_{1} z=1 \\ & a_{2} x+b_{2} y+c_{2} z=1 \\ & a_{3} x+b_{3} y+c_{3} z=1 \end{aligned}

A first assigns one of the coefficients ai,bi,ci(i=1,2,3)a_{i}, b_{i}, c_{i}(i=1,2,3) with a natural number of his choice.

Then B assigns one of the remaining coefficients with a natural number of his choice, then A again, then B, and so on, until finally A assigns the last (ninth) coefficient with a natural number.

A wins if, after these assignments, the system of equations (1) has exactly one real solution (x,y,z)(x, y, z).

B wins if, after the assignments, the system of equations (1) has no or infinitely many real solutions (x,y,z)(x, y, z).

Investigate whether B can enforce a win through appropriate assignments in every case.

This one wants a proof. Work it on paper, then read the official solution and mark yourself. Be honest about it: the record is only any use to you if it is.

Official solution

Let (x,y,z)(x, y, z) be a solution of the system of equations (1). Then it holds that

(a2c1a1c2)x+(b2c1b1c2)y=c1c2(a3c2a2c3)x+(b3c2b2c3)y=c2c3(a1c3a3c1)x+(b1c3b3c1)y=c3c1 \begin{aligned} & \left(a_{2} c_{1}-a_{1} c_{2}\right) x+\left(b_{2} c_{1}-b_{1} c_{2}\right) y=c_{1}-c_{2} \\ & \left(a_{3} c_{2}-a_{2} c_{3}\right) x+\left(b_{3} c_{2}-b_{2} c_{3}\right) y=c_{2}-c_{3} \\ & \left(a_{1} c_{3}-a_{3} c_{1}\right) x+\left(b_{1} c_{3}-b_{3} c_{1}\right) y=c_{3}-c_{1} \end{aligned}

From this, by multiplying with b3,b2b_{3}, b_{2}, and b1b_{1} respectively and adding, we obtain

Dx=(c1c2)b3+(c2c3)b1+(c3c1)b2 D \cdot x=\left(c_{1}-c_{2}\right) b_{3}+\left(c_{2}-c_{3}\right) b_{1}+\left(c_{3}-c_{1}\right) b_{2}

where

D=a1b2c3a1b3c2+a2b3c1a2b1c3+a3b1c2a3b2c1 D=a_{1} b_{2} c_{3}-a_{1} b_{3} c_{2}+a_{2} b_{3} c_{1}-a_{2} b_{1} c_{3}+a_{3} b_{1} c_{2}-a_{3} b_{2} c_{1}

Furthermore, we obtain

Dy=(a1a2)c3+(a2a3)c1+(a3a1)c2 and  D \cdot y=\left(a_{1}-a_{2}\right) c_{3}+\left(a_{2}-a_{3}\right) c_{1}+\left(a_{3}-a_{1}\right) c_{2} \quad \text { and }

Dz=(b1b2)a3+(b2b3)a1+(b3b1)a2 D \cdot z=\left(b_{1}-b_{2}\right) a_{3}+\left(b_{2}-b_{3}\right) a_{1}+\left(b_{3}-b_{1}\right) a_{2}

Therefore, the system of equations (1) has, due to (2), (4), and (85), exactly one real solution if D0D \neq 0, so that A wins; in the case D=0D=0, no or infinitely many real solutions, so that B wins.

Hence, B can enforce the win with the following strategy, i.e., achieve that D=0D=0:

1. Without loss of generality, assume that A first assigns the coefficient a1a_{1} (all other possibilities can be reduced to this by swapping equations or swapping unknowns). Then B assigns the coefficient c2c_{2} with 0 and achieves

D=a1b2c3+a2b3c1a2b1c3a3b2c1 D=a_{1} b_{2} c_{3}+a_{2} b_{3} c_{1}-a_{2} b_{1} c_{3}-a_{3} b_{2} c_{1}

where the coefficient a1a_{1} is now fixed.

2. If A now assigns the coefficient c3c_{3}, B assigns the coefficient b2b_{2} with 0 and achieves

D=a2b3c1a2b1c3 D=a_{2} b_{3} c_{1}-a_{2} b_{1} c_{3}

If, on the other hand, A assigns another coefficient, B assigns the coefficient c3c_{3} with 0 and achieves

D=a2b3c1a3b2c1 D=a_{2} b_{3} c_{1}-a_{3} b_{2} c_{1}

In both cases, in the obtained representation of DD, at most one coefficient is fixed (in the first case, this is exactly the coefficient c3c_{3}).

3. If A now assigns another coefficient, in at most one of the two products (from the obtained representation of DD) two fixed coefficients will occur. B can then assign another coefficient with 0 in a product with the maximum number of fixed coefficients and achieve

D=a2b3c1 or D=a3b2c1 D=a_{2} b_{3} c_{1} \quad \text { or } \quad D=-a_{3} b_{2} c_{1}

where in each case at most one coefficient is fixed.

4. Regardless of which coefficient A now assigns, B can achieve that D=0D=0 by assigning the still free coefficient with 0.

Regardless of any further assignments made until the end of the game, B has thus enforced the win, as in the case D=0D=0, the system of equations (1) has no or infinitely many solutions.

## Adapted from [3]

Source: NuminaMath-1.5, licensed Apache-2.0. Statement and solution reproduced as published; topic, difficulty and ordering added by this site.