Is $$2\prod_{i=1}^{n}{p_i} - \prod_{i=1}^{n}{\left(p_i + 1\right)} - \prod_{i=1}^{n}{\left(p_i - 1\right)}$$ even and negative for $n > 1$, where $p_i > 1 \hspace{0.07in} \forall i \in \left[1,n\right]$?
Here are a few quick WolframAlpha computational verifications:
$$2pq - (p+1)(q+1) - (p-1)(q-1) = -2$$
$$2pqr - (p+1)(q+1)(r+1) - (p-1)(q-1)(r-1) = -2(p+q+r)$$
$$2pqrs - (p+1)(q+1)(r+1)(s+1) - (p-1)(q-1)(r-1)(s-1) = -2\left(p(q+r+s)+q(r+s)+rs+1\right)$$
How do we prove the general case?
Yes, that expression is even and negative for $n > 1$.
It is even because $2\prod_{i=1}^{n}{p_i}$ is even, and $p_i+1\equiv p_i-1 \bmod 2$ for all $i$ so that $\prod_{i=1}^{n}{\left(p_i + 1\right)} \equiv \prod_{i=1}^{n}{\left(p_i - 1\right)} \bmod 2$.
We show that it is strictly negative by induction. You prove the base case $n=2$ yourself. Now assume
$$2\prod_{i=1}^{n}{p_i} - \prod_{i=1}^{n}{\left(p_i + 1\right)} - \prod_{i=1}^{n}{\left(p_i - 1\right)} < 0$$
Write this as $$2R-S-T < 0$$ Then the equivalent expression with $n+1$ terms is
$$2pR - (p+1)S-(p-1)T$$ (where we have written $p$ for $p_{n+1}$). But this is just
$$p(2R-S-T)-S+T$$
which is less than $p(2R-S-T)$ (and therefore $< 0$) because $S=\prod_{i=1}^{n}{\left(p_i + 1\right)}$ is clearly greater than $T=\prod_{i=1}^{n}{\left(p_i - 1\right)}$.