Let’s call every detrimental effect that isn’t straight-up harm in a fight a “Condition.” Conditions allow for greater and more easily grasped tactical complexity in a fight sequence. In a tabletop RPG, you might be able to, in the fiction, knock someone down off their feet – but if that action has no consequences within the rules, there’s not a lot of incentive to attempt it, when you could just hit them again more solidly and end the fight faster. If the knockdown was an actual condition that cause a visible shift in the outcome of the battle, then there is incentive to perform that move. Conditions should optimally allow for modifications to the fight that give one side advantages that are useful but do not outright remove opponents from the combat.
Why the latter prerequisite? Conditions that outright stop action strike me as more trouble than they are worth. To me, the excitement and fun of a battle scene is in each participant’s ability to act. Enemies use their abilities to harm or hinder the players and the players navigate those abilities while also dealing out their own. In the hands of the players, abilities that outright stop actions can be easily applied on the most powerful enemies to render a situation too lopsided in their favor. Meanwhile in the hands of enemies you have an all-too-real moment where a player is incapacitated instantly by a condition, and has to pass a turn while doing nothing or effectively nothing. Being on the receiving end of this is hardly fun.
I’d like to share my thoughts on some common Conditions in tactically-complex Tabletop RPGs.
•Rooting/Snaring/Pinning/Paralyzing: This kind of condition has many names but all of them generally mean a cessation of movement. Most games separate the act of moving with other combat actions, so this type of condition is a pretty acceptable one in nearly all cases. Unless the game has few rules for supporting actions (such as giving allies bonuses from afar, for example observing enemy weaknesses) and no easily accessible ranged combat, being unable to move from your spot can force a tactical shift if properly applied, but without invalidating a character completely.
In games with grid-based movement where characters can move multiple units of range per turn (such as D&D 4e, with its 5-7 squares of movement) slowing is often a milder version. I don’t find slowing very exciting or relevant, and it often doesn’t force much of a change in a battle, but it has its place in granular movement systems.
•Numerical Penalties: -2 Attack/Defense/Armor/Damage etc. This is probably the least offensive condition you can apply in a game, though high penalties to attack or damage might start invalidating people’s actions. The size of your RNG will dictate what kind of penalties are relevant and what kind are much too high. For example, on a bare d20, a -1 penalty is basically nothing, -2 is probably the smallest that will ever be relevant and -4 is a fifth of your RNG and rather powerful. Meanwhile, on 2d6, a -1 is acceptable, -2 is relatively powerful, and -4 is likely to be utterly devastating.
Whether or not this penalty invalidates a character also depends on what kind of penalty it is. Penalties to passive stats like Defense always keep the character fairly relevant, but penalties to active stats like “hit rolls” come close to depriving a character of his or her role. However the difference is that the action can still be attempted at all and the system can have ways to mitigate low rolls, such as Miss effects or other fail foward mechanisms. So if your character has a -4 penalty in a 2d6 system with a fail forward mechanism, you still have the impetus and ability to act and contribute.
•Poison/Burning/Bleeding/Affliction: Damage Over Time effects are really hard to do “correctly.” Often they either do too little or do too much. Depending on the kind of health system in your game, 10 damage over time might be too much or it might be nothing, especially depending on the “time.” A Damage Over Time effect that lasts one turn has done practically nothing that a normal attack wouldn’t do – whereas one that lasts an entire battle might be too much. While Damage Over Time is usually incapable of rendering a character totally ineffective, it’s delicate in that overcompensating for its existence can render it meaningless (convenient potion chugging) or way too devastating (can’t be cured mid-battle, or requires way too much time to cure, such that it effectively takes the character out of a match for trying to restore itself).
•Impairing/Dazing/Stunning: This kind of condition is the tricky one. These conditions tend to remove a character’s ability to act at all, or to take actions relevant to the combat, often without a way to mitigate the effect. For most systems, completely depriving a character of the actions that contribute to victory is just a source for strife and frustration. An enemy that constantly stops your actions does not provoke you to change your tactics or force different input from you – it just makes you sit around doing nothing for that turn. It gets “cheap hits” on you. Often there’s not a lot of accessible ways to mitigate these conditions because they’re meant to be rare and powerful. Yet a lot of games fall into the trap that this “rare and powerful” condition should be the specific gimmick of a certain monster or character archetype, causing imbalance.
Unless you can “break out” of this impairment in time to take action, or it really is rare and powerful and won’t come into play often, it’s best to avoid having such a condition at all. Suitable alternatives can just be specific enemy abilities that can be worked around. For example, if a player is facing an enemy that stops only specific actions, or only stops actions when a certain specific trigger occurs, then you have a condition that forces a change in a good way, and challenges the player. This is different from getting struck and “Stunned” to inaction, helpless to overcome the effect.