Buffs are cool in concept (and I like Green Mages) but I feel like they are rarely implemented well.

One of my favorite examples of buffs done badly is Valkyria Chronicles. That game has a “command point” system that allows you to take a certain number of actions per round. A point can be spent to have a unit move and attack, but you can also spend points to execute “orders.” These include special options like evacuating fallen allies and buffs like increasing a soldier’s damage against certain enemy types. The trouble is, despite just being a numerical calculation it’s not at all obvious when it’s better to use an order to increase your damage output and then attack versus saving the command points to just attack multiple times.

This is because of two problems. First, the UI doesn’t explain how much difference the buffs actually make. If the numbers were clear you could do the math and figure out when the buffs were worth using. Cosmic Star Heroine leans heavily in this direction and is a great example of how to do this well. It has an unusual combat system in which most abilities can only be used once before they need to be recharged - the recharge action then recharges all of your depleted abilities. But it costs a turn, so it’s usually worth using most or all of your abilities first. Many abilities have buff-like effects or other interactions and all the numbers are clear, so it’s a complex but feasible analysis problem to figure out the optimal abilities to use in what order for the current situation. Buffs are thus an interesting tactical option that are often but not always correct to use.

The other problem with the Valkyria Chronicles approach is that the buffs are expensive and short-lived, meaning it’s often a close call whether they’re worth using even if you did have the numbers. World of Warcraft, at least back when I played it, was an example of the opposite approach here - most buffs were cheap enough compared to their duration that they were always obviously correct to use even if you didn’t look at the numbers. This meant they were more of a rote chore than a tactical decision but the result was a system that encouraged class diversity and balanced teams. If you had a Mage in your group everyone would have buffed Intellect, if you had a Priest everyone would have buffed health, and so on. These buffs might as well have been automatic passive bonuses, but they achieved their goal without creating any confusing decisions or making you stop playing to solve a math problem.

Games like Valkyria Chronicles where you can choose to trigger a short-lived no-numbers buff instead of taking an additional action give the player a math problem that is both uninteresting and obscured. There always is an optimal choice, it’s just hard to suss out. So you can either interrupt the action to look up and crunch the numbers or play on knowing you are often not doing the best thing - which is frustrating in games like Valkyria Chronicles that grade and reward you based on your efficiency.

In practice, I usually avoided using the buff orders. Having a unit move and attack was more fun than picking a buff from a menu anyway.