This was supposed to be a comment, but it became too long.
When you want your students to understand the material, I think it's good practice to avoid remembering procedures (I avoid the term 'trick' because I want to distinguish between two types on tricks: IMO, there is a good and a bad kind) as much as possible. I think they are harmful for understanding of the topic. A much more useful skill is to derive formulas on the spot. I have to admit that there are examples in which it is just not feasible to do this (for example, I would advise to just remember the quotient rule as it's too complex to derive on the spot every time, but I think it's also good to be able to derive it, so that forgetting a rule is no big deal).
So, to recap, I think it's better to teach techniques which enable students to come up with formulas. Remembering a technique (normally, I'd refer to a technique as a trick, rather than a formula or abbreviation) usually works way better than remembering a formula (because there is usually some intuition behind the technique, and our brains just are better at remembering ideas than they are at remembering raw formulas).
I can (hopefully) show what I mean, by an example.
Consider two students that have learned how to find roots of a quadratic equation.
Student A has learned to write $x^2 + bx + c = 0$ in the form $(x-a)^2 = d$, so that you can apply a squareroot on both sides, then solve for x to get: $ x = a \pm \sqrt{d} $. Ideally, he would understand that we translate the graph of the polynomial (by doing a substitution for x) to get rid of the $bx$ term, get an equation that is easy to solve, and then translate/substitute back again.
Student B has learned $$ x = \frac{-b \pm \sqrt{b^2 - 4ac}}{2a} $$
Now suppose they learn about cubic polynomials and their roots. Using some clever geometric tricks, it is possible to solve $x^3 + px + q = 0$.
Student A's knowledge of the technique now enables him to see a similar possibility here. IF you have $x^3 + ux^2 + vx + w = 0$, and you can find an $a$ such that the quadratic term will vanish (just like the first-order term vanished in the quadratic equation), you can write any cubic equation in the simpler form. So student A may realize that the solution to the simpler cubic equations is enough to solve for all cubic equations.