$\lim_{x \to a} f(x)=\lim_{h \to 0} f(h+a)$
The above seemingly obvious result is what I want to prove.
My idea was to "set" $x$ to be equal to $h+a$ and then prove using the epsilon delta definition that $\lim_{h \to 0} h+a=a$ or that $\lim_{h \to 0} x=a$. This would (i think) mean that $\lim_{x \to a} f(x)=a$ since $x=h+a$ and ${h \to 0}$ implies that ${x \to a}$.
I didn't like it because we somehow had to "set" $x$ equal to $h+a$ in the beginning. I wonder if there is a better way to prove it.
Let $\lim_{x\rightarrow a}f (x)=c $ (assuming limit exists). Let $\epsilon >0$. Then there exists a $\delta >0$ so that $|x-a|<\delta \implies |f (x)-c|<\epsilon$. Then $|h-0|=|(h+a)-a|<\delta \implies |f (h+a)-c|<\epsilon $.
So $\lim_{h\rightarrow 0}f (h+a)=c $.
Most do same thing for when limit doesn't exist.
====
I see by your comment you didn't understand this
Definition: $\lim_{w\rightarrow a} f(w) = c$ means the for any value $\epsilon 0$ there exist a $\delta >0$ so that the following is true: If $|w - a| < \delta$ or in other words if $w \in (a - \delta, a + \delta)$ then $|f(w) - c| < \epsilon$ or in other words $f(w) \in (c-\epsilon, c + \epsilon)$.
Now $w \in (a-\delta, a+ \delta) \iff w-a \in (0 -\delta, 0+ \delta)$.
So $\lim_{x\rightarrow a}f(x) = c$ means for any $\epsilon > 0$ there exists a $\delta$ so that $a-x \in (0-\delta, 0+\delta) \implies f(x)=f((x-a) + a) \in (c - \epsilon, c + \epsilon)$.
And that is the definition of $\lim_{x - a \rightarrow 0}f((x-a) +a) = c$. Now we can just relabel $x - a$ as $h$ where $h = x - a$. So $\lim_{h \rightarrow 0}f(h + a) = c = \lim_{x \rightarrow a} f(x)$.