Step 1: Compute $P^2$ directly by squaring the matrix: $P^2 = \begin{pmatrix} 1 & 0 \\ 0.5 & 0.5 \end{pmatrix}\begin{pmatrix} 1 & 0 \\ 0.5 & 0.5 \end{pmatrix}$.
Step 2: Row 0 of the product: $(1\cdot1+0\cdot0.5,\ 1\cdot0+0\cdot0.5) = (1,0)$. Row 1 of the product: $(0.5\cdot1+0.5\cdot0.5,\ 0.5\cdot0+0.5\cdot0.5) = (0.75,\ 0.25)$. So $P^2 = \begin{pmatrix} 1 & 0 \\ 0.75 & 0.25 \end{pmatrix}$, and its $(1,0)$ entry is $0.75$.
Step 3: Cross check by enumerating sample paths from state 1 to state 0 in exactly two steps: path $1\to0\to0$ has probability $0.5\times1=0.5$, and path $1\to1\to0$ has probability $0.5\times0.5=0.25$. These are the only two ways to land on state 0, and being mutually exclusive their probabilities add: $0.5+0.25=0.75$, matching the matrix computation.
\[\boxed{0.75}\]