If $f(t)=g(t)u(t-t_0)$, then the Laplace Transform, $F(s)$, of $f(t)$ is
$$\begin{align}
F(s)&=\int_0^\infty f(t)e^{-st}\,dt\\\\
&=\int_0^\infty g(t)u(t-t_0)e^{-st}\,dt \\\\
&\int_{t_0}^\infty g(t)e^{-st}\,dt \tag 1
\end{align}$$
Now, we enforce the substitution $t-t_0=t'$ into $(1)$ to obtain
$$F(s)=\int_0^\infty g(t'+t_0)e^{-s(t'+t_0)}\,dt' \tag 2$$
Using $(2)$ with $g(t)=t^2$ and $t_0=5$ yields
$$\begin{align}
F(s)&=\int_0^\infty (t+5)^2e^{-s(t+5)}\,dt\\\\
&=e^{-5s}\int_0^\infty (t^2+10t+25) e^{-st}\,dt\\\\
&=e^{-5s}\left(\frac{2}{s^3}+\frac{10}{s^2}+\frac{25}{s}\right)
\end{align}$$
Note the in general, if $F(s)$ is the Laplace transform of $f(t)$, then $e^{-st_0}F(s)$ is the Laplace Transform of $f(t-t_0)$. For the function $f(t)=t^2$, the Laplace transform is $\frac{2}{s^3}$.
Therefore, the Laplace transform of $(t-5)^2u(t-5)$ is $e^{-5s}\left(\frac{2}{s^3}\right)$. But then note that $(t-5)^2u(t-5)\ne t^2u(t-5)$. And this was, I believe, the issue with the OP.