For points $\mathbf{a}, \mathbf{b}, \mathbf{x}, \mathbf{y} \in \mathbb{R}^d$ for some natural $d \ge 1$. I want to prove that if (Euclidean distance) $||\mathbf{x} - \mathbf{a}|| \le ||\mathbf{y} - \mathbf{a}||$ and $||\mathbf{x} - \mathbf{b}|| \le ||\mathbf{y} - \mathbf{b}||$, then for any point on line between $\mathbf{a}$ and $\mathbf{b}$, call it $\mathbf{c}$, $||\mathbf{x} - \mathbf{c}|| \le ||\mathbf{y} - \mathbf{c}||$
I know how to prove this equation using an argument derived from calculating the hyper-plane of equidistance points between $\mathbf{x}$ and $\mathbf{y}$, and then, using some more implications to argue that $\mathbf{a}$ and $\mathbf{b}$ must on the same side as the side of the hyperplane $\mathbf{x}$ because they are closer and then arguing that then $\mathbf{c}$ must be on the same side as the side of $\mathbf{x}$; however, I'm looking for a solution that directly derives the property that $||\mathbf{x}-\mathbf{c}|| \le ||\mathbf{y}-\mathbf{x}||$ using the properties of Euclidean distance.
I understand that we can parameterize $\mathbf{c}$ in terms of $\mathbf{a}$ and $\mathbf{b}$ as $\mathbf{c} = \alpha \mathbf{a} + (1 - \alpha)\mathbf{b}$ where $0 \le \alpha \le 1$, and that was my first step. Then, I tried to start from $||\mathbf{x}-\mathbf{c}||$ and reach that it is less than or equal to $||\mathbf{y} - \mathbf{c}||$ using a series of reformulations from my starting point and my parameterized definition of $\mathbf{c}$. However, I was not able to get anywhere. I'm quite stuck as to how to prove this in this manner, and any hints/advice would be greatly appreciated.
Thank you.