WoW Theorycrafting 101 #1  When To Save Cooldowns?
“Should I Save Or Should I Go?” is one of the most commonly asked questions in WoW for any class/spec when it comes to cooldown usage. Okay, maybe not in those exact terms, but how do you actually decide if it is better to save a cooldown to line up with a damage amp, a big pack of mobs, or another cooldown or Trinket?
In this post I will go over some of the principles that guide these decisions, sort of a “mental checklist” that you can use for pretty much any situation in the game, as any class.
Part 1: Saving A CD for a Trinket/Other CD/Etc.
There are a few factors/variables that guide this decision.

The damage increase of whatever we’re thinking saving the cooldown for.

The amount of time we are thinking of delaying, and whether this costs us a cast or not.

Minor concern: The cost of casting the cooldown. This could be Resources, or a GCD.
Once you know, or at least mostly know, these variables, it becomes a pretty simple piece of math.
Let’s say we got:
 Cooldown A  3 minutes CD, increases damage by 20%.
 Cooldown B  4 minutes CD, increases Haste by 30%.
The question is: Is it worth delaying the 3minute CD for the 4minute CD every time to sync them up?
To find this out, we first look at the amount of time that we are delaying. That would be 1 minute, which is 33% of a 3minute cooldown.
Thus, we are delaying the CD by 33% of its cooldown. Over a long time, this effectively means that we would get 33% fewer casts of this cooldown. To justify this, we would need to be making that cooldown 33% stronger through delaying it. 30% Haste is not going to be a 33% DPS increase in most situations, although this can certainly depend on the class. A good rule of thumb however is that 1% Haste = 1% DPS, since you’re just doing 1% “more” of everything. Certain class mechanics can complicate this simple rule, though, so it’s not a hard and fast one.
Part 2: So What About Saving For AoE?
Most AoE abilities in WoW scale linearly with additional targets. An Explosive Shot on 2 targets does twice the damage of a singletarget Explosive Shot, so you can use the same rule as before. The damage increase of delaying for an extra target to be up is 100%, and thus you can delay an ability like Explosive Shot for a long time if it would ensure it does AoE damage.
But there’s more to it than just that. Look at the third variable  the cost of casting a cooldown/ability. Explosive Shot costs 20 Focus and 1 GCD to cast.
If it does double damage on 2 targets, which it does, then it is not just the case that a 2target Explosive Shot is just as powerful as 2x singletarget Explosive Shots… It is BETTER, because when you delay the ability for AoE, you are getting the same damage but for LESS FOCUS and 1 GCD instead of 2!
You can apply this same logic to most mechanics. All you need is the DPS increase of delaying which comes from knowing how much of a DPS increase an effect is, compared to the DPS loss of delaying which comes from losing casts of the spell, which is basically just a matter of dividing the total cooldown of the spell by the amount of time that you are delaying it for. In our case that was 180 (3 minute CD) divided by 60 (1 minute delay) = 3 (so, 100/3 to get your percentage).
A more common example perhaps is whether you should delay Wild Spirits for AoE. From a strictly DPS PoV, based on the above math puzzle, the answer is nearly always yes. It costs a global after all, and its damage scales linearly with additional targets up to its target cap.
However, in a practical Mythic+ scenario, it is not quite this simple. After all, if you delay your Wild Spirits for AoE right now, how can you be sure that doing so is not just costing you the opportunity to cast Wild Spirits on another AoE pack in the future? This uncertainty means that the safe choice is nearly always to cast Wild Spirits almost off CD in Mythic+, unless you know the rhythm and structure of the dungeon well enough to get rid of this uncertainty.
Another aspect that I will probably have to make another post about entirely, is the fact that raw overall DPS in Mythic+ is not really all that matters. What you’re actually optimizing for in competitive Mythic+ is time saved.
Keeping it short, imagine if a dungeon was 50% singletarget, and 50% 5target AoE fest in terms of time spent and it takes 10 minutes, and you do 10K DPS overall.
If you improve your DPS by 10% on singletarget, you save 30 seconds of time, and you gain 100 DPS.
This is because, assuming for argument’s sake that 5target AoE is 5x the damage of singletarget, your singletarget damage actually only contributed 10% of your overall DPS, even though it took up 50% of your time spent. 10% of 10% is 1%, so that’s 100 DPS.
Now, if you improve your DPS by 10% on 5target AoE, you STILL save just 30 seconds of time, but you gain 500 DPS.
So the time save is the same, but the DPS mostly comes from AoE. Because AoE damage is outright higher, increasing it nets higher Overall DPS numbers, even though the actual time saved is identical, or even lower in some cases.
Thanks for reading! I welcome feedback to these as it is my first stab at writing this. I might do more in the future if people found this useful. I tried to start with a basic, but universal problem. A lot of people honestly overestimate the complexity of WoW Theorycrafting and I want to help make it more accessible and simple to understand. At the end of the day, you can get almost all of it done with basic arithmetic combined with simple understanding of the game’s mechanics.