Larian Studios Forums Baldur's Gate III Baldur's Gate III - Suggestions & Feedback A discussion on how the game's current combat mechanics design heavily restricts tactical options

Hey, I couldn't come up with a better title considering the wide range of topics I've been thinking on as I've been reading through these forums for the past few days, and I thought it might be a good idea to have some centralized thoughts on how the various combat mechanics in the game interact with each other, and how certain changes either through scaling back already existing mechanics or adding in new ones can open up more creative options. I have an entire Friday to think about this since I've clearly nothing more productive to do.

That said, some of this might sound like a wishlist too, but at the same time, I also recognize that acting on a lot of what is to be discussed here might not be possible, perhaps due to possible limitations within the engine used for BG3 and that the developers may have deemed it too much work to overhaul the engine to include some missing features from tabletop, as long as the majority of the public and the gaming press are still happy with the end result. That is neither here nor there, and this discussion is as much aimed at the developers as it is towards people on these forums who may end up becoming future developers themselves. I don't mean to make this yet another comparison between BG3, Tabletop (and Solasta by default), and D:OS2, but those are the most direct frames of reference available.

---

Combat balance tends to be a multi-faceted topic. Few people really know how to describe it, and while I will not pretend to be one of said few, I would at least like to think that I'm making an effort. At the moment, I personally feel that the game's tactics lean far too much towards pre-battle planning/setup, and once combat actually begins, there's very few options to turn a situation around if the original plan fails. The lack of certain options makes a battle highly reactive rather than encouraging anything proactive, outside of one constant mechanic as detailed below. I think a lot of the grievances about BG3's combat balance ultimately revolves around the action economy, more importantly in regards to bonus actions.

But before we get into that, we should talk about the high ground advantage/low ground disadvantage system, and how that particular design choice might be the main culprit behind many of the other seemingly problematic mechanics existing. In most other games, such a concept was largely used for controlling the field via forcing chokepoints while protecting ranged characters on high ground, and maybe extending their attack range on top of that. D:OS2 went the extra step of adding a damage modifier for characters on high ground. But BG3 is the first game I've seen that actually added an accuracy bonus to it by essentially doubling hit rate via advantage. This is highly problematic for the below reasons.

1) It is a free method of gaining advantage, in a game where other methods generally require using spell slots, a finite mid-battle resource. Even worse if said spells require concentration on top of that, such as Faerie Fire. Sure, Faerie Fire gives you advantage regardless of environmental factors, but it has a failure rate via Dexterity saves, it again requires concentration, and that means the opportunity cost of actually using it is far worse.

2) It heavily cheapens the value of buffs that increase your modifiers by a flat rate, such as Bless. Oh yeah, Bless happens to be a concentration spell too. One only needs to look at Solasta to see how the inclusion of high ground advantage affects the usage of Bless in both game's balance. In BG3, Bless is generally considered worthless, and hearing the developers say that people aren't using it because it's boring or not flashy enough isn't really something you want to be hearing from the creators of one of the most celebrated turn-based combat systems. In Solasta, it's considered one of the most powerful spells you can use, with flat modifiers being much more valuable in an environment where it is much harder to gain advantage rolls.

3) Ranged spells that target saves are inherently worse for damage purposes than spells that target AC, unless you are on even or low ground (and even then, the spells that target AC probably still have a higher success rate than saves on even ground). That said, you have plenty of reasons to be rushing to high ground ASAP, so this isn't something you'd consider unless things have already gone sideways. That said, spells that target saves usually do inflict half damage if an enemy succeeds on the saving throw (while spells that target AC would completely miss), but there aren't too many of those yet.

4) Since ranged characters on high ground essentially have permanent advantage against enemies on lower ground, all the tactical variety in the world suddenly becomes focused towards controlling that high ground as soon as possible. The value of AC suddenly has an exponential curve in such a situation, rather than a flat increasing bonus. Even worse, the encounter design knows this, from the sheer amount of encounters with easily accessible high ground points available to both players and enemies, with several enemies frequently starting on high ground themselves. At that point, your number 1 priority is to either knock them off or sneak to a higher point before a fight begins, sometimes missing out on dialogue if you really want a surprise round on top of that.

(That last part was or less what Tactician Mode in D:OS2 turned into if you wanted to be efficient due to the way encounters were designed in that game. For example, whichever party member actually spoke to Alexander at the end of Act 1 had a high probability of immediately dying afterwards once the fight broke out, as opposed to the entire party just ambushing the enemy group and skipping the dialogue entirely. Or the infamous scarecrow Alice, who usually had absurd initiative, would attack you no matter what even if you passed the persuasion checks, had crazy insane damage modifiers due to a personal buff, and usually opened up with a fireball that would immediately party wipe most players if they were to talk to her normally with all four party members present - which resulted in ridiculous tactics like having one party member talk to her to distract her while the other three party members went up the nearby ledges in the ruins to set up the field in their favor, or teleporting her to the nearby vendor who happened to be an overpowered former player character from the first DOS game. The latter is generally widely considered by the community to be the most direct way of dealing with her.)

One of the most common suggestions to mitigate the effect that this system has on this game's tactical focus would be to have high ground give the character effective +2 AC against ranged attacks from enemies on lower ground, which would act as a replacement tabletop cover system. But it remains to be seen if this is something that will happen.

Now, the devs may be looking at this and wondering how to allow melee characters to keep up. It would not be too much of a stretch to think that the backstab advantage system likely wouldn't exist if it weren't for high ground advantage. Even so, one would think that this combined with melee having the ability to threaten enemies with possible opportunity attacks would be enough to balance the relationship between the melee and ranged, right? But in terms of actual practicality, that's not the case, because of something else.

---

Jump/Disengage being coupled together and being considered bonus actions. Jump is perhaps a necessary tool to handle the verticality in this game, but in my opinion, it should only be used for crossing large gaps, jumping into pits with feather fall, or jumping up/down higher ledges mid-combat. It is an utter waste of the player's time to be using it outside of combat just to go up or down a ledge with no other path existing to just manually climb without it (such as the ledge that leads to the hidden perception check hiding a ruby), but that's another topic entirely.

(I should also mention how Jump by itself cheapens the value of field effects, but at the same time, this is less of a concern now that field effects now only trigger once per turn rather than being based on distance traveled within said field, like it used to do during the initial EA launch and how it works in D:OS2.)

The bigger problem is the Disengage being tied to it, and being considered a bonus action. This means melee characters running up to a ranged enemy often doesn't mean anything like it does in most other tactical RPGs, because the ranged enemy is just going to disengage and shoot at Gale in the back without any penalty anyways. You SHOULD be punished for leaving your ranged exposed, the same way you should actually be rewarded for threatening the enemy ranged too.

Making Disengage its own ability and changing it into a standard action would suddenly change a lot of this game's tactics. Players would be a lot more careful about the positioning of their ranged, while melee characters would suddenly prioritize going after the ranged even more and actually getting rewarded for pressuring them. Using that Dash action to catch up to the enemy ranged instead of hitting the nearest enemy suddenly seems like an attractive option, doesn't it? Or pull back to chase a melee enemy going after your ranged. Actually getting threatened would give you a reason to use spells like Shocking Grasp to cancel the enemy's potential opportunity attack and safely move away (especially if there are multiple ranged party members being threatened by that one enemy, meaning that wizard only has to use one action to allow everyone else to get away for free). Or maybe you wanted to bait the enemy to run up to your ranged so that they'd be conveniently next to a ledge to be shoved off of by a different character afterwards, or for said enemy to be immediately flanked and shanked with a sneak attack by your rogue. A bonus action Disengage means such unconventional tactics wouldn't cross the minds of most people.

At the same time, that would also remove the need for the ridiculous double 'Threatened' and 'Enemy Too Close' disadvantage malus, because both basically mean the same thing so why should they even be separate disadvantage penalties? Get rid of one of them, preferably 'Enemy Too Close' as Threatened is a far more specific term.

(On that note, the 'Threatened' range should really just be tied to the enemy's opportunity attack range. There is a complete lack of information in regards to exactly how far Threatened radius extends, which is just obfuscation of information in a tactical game such as this. If you're not at risk of taking an opportunity attack by moving away from a nearby enemy, you shouldn't be considered threatened. At the same time, tying threatened radius to weapon range would be a great reason to consider using reach weapons like spears too.)

---

The emphasis on wanting to control the high ground means that shoving by default is quite powerful. Bonus Action Shove however is quite overkill. It is normally a risky attempt to control the fight and cause a lot of free environmental damage, if not an instant kill if you throw someone into an out of bounds area. But making it a bonus action means it's less 'risk' and more 'free potential source of damage'.

That said, shoving an enemy out of bounds also results in a loss of rewards. Maybe if shoving was a high risk factor with becoming a standard action, this wouldn't be necessary, although it's more likely to be a programming limitation thing more than anything else at the moment.

Larian wants us to be creative, but immediately punishes players that want to utilize an immediate spammable solution in this manner. If the devs want to discourage people from using that kind of tactic in every fight by losing rewards, they should seriously look at how a fight would be tuned without the existence of bonus action shove or pre-battle field manipulation such as the use of barrelmancy. Some enemies' health appear to be tuned so high that there's little apparent recourse in certain cases. One can see this play out in their most recent stream in regards to the hag fight, with a lead developer going for the shove strategy immediately and doubling down on it when it repeatedly failed instead of trying to change tactics. It is proof that there is no actual flexibility in the system, and no other clear tactical options with the way they decided to approach the fight. Sometimes when you and the enemy are given so many options in regards to the action economy, and encounters are obviously balanced around this very idea, you generally default to the same strongest tactics and ignore 90% of the other tools given to you.

(At the same time, it's also possible that the hag fight is really tuned for level 5+ characters and that it is considered one of, if not the hardest fight in the game, but seeing someone struggle that much just because of a 1 level difference doesn't bode well for the tactical future of the game.)

---

There is one common thing to note about all of the above. They, along with the concept of barrelmancy and consumable items such as magical arrows and bombs being so readily available, are all things that every character, players and enemies, have easy access to. Character builds have little difference in how you use the above mechanics, aside from shove chance being tied to strength value (and even then it's 100% if you're in stealth). It is the major reason why there is criticism that classes may feel homogenized in this game. When you give everyone such powerful options, it actually limits the viability of tactical strategies you can employ, especially in regards to the bonus actions.

DnD is inherently a cost/benefit analysis system, but the 'cost' part of the balancing equation barely exists in BG3 as is. I have seen people often argue that you don't have to use such cheese options if you want to, but the problem is, the encounter balance will not operate under the same principles. You are still ultimately going to get pigenholed into securing the high ground ASAP even if you choose to ignore using disengage and shove.

---

It is known that Larian lowered enemy AC, out of a belief that players missing often was a frustrating design choice. However, they increased their HP of enemy mobs to compensate, which led to the balance between spells that targeted enemy AC (and thus subject to advantage/disadvantage) and spells that targeted saves being thrown off. Sure, spells that target saving throws do half damage if an enemy successfully saves against them, but that only applies to spells that use spell slots, not cantrips like Sacred Flame - the only damaging cantrip that Clerics currently have in BG3.

DragonSnooz had a more elegant solution to this dilemma, along with further reasoning why the lower AC/higher HP design choice worked to the detriment of players in the long term.


Basically, instead of lowering AC and increasing HP, altering proficiency values would have had the same effect as what Larian tried to do *without* disproportionately affecting the efficiency of certain spells, especially in regards to how effectively players can use them VS how enemies use them. It is really not a good thing when status effect spells like Sleep are more effective against player characters than enemies. That kind of stuff is deep within the realm of jRPG design. And we should know better by now.

For the uninitiated, all player characters start with +2 to proficiency. It goes up by +1 at level 5, 9, 13, and 17. Granted, we likely aren't reaching level 13+ in BG3, so let's assume we'd end up with +4 proficiency towards endgame. Proficiency affects attack rolls, damage, and DC of spells that target saving throws (DC = Difficulty Check, an enemy has to roll higher than the spell DC to succeed their saving throw against them). If Larian had instead adjusted proficiency bonuses and the proficiency curve to, say, starting at +3 and increasing by +1 at level 3, 5, 7, and 9 (effectively giving us +7 proficiency by an assumed endgame), it would have basically had the same effect as lowering enemy AC along with increasing the success rate of spells that target saving throws at the same time. And enemy HP would not have been needed to be adjusted to compensate for anything, meaning spells like Sleep become more relevant again. And if that wasn't enough, players now have an incentive to consider using buff spells like Bless again to overcome higher enemy AC.

---

Now, let's talk about the most requested missing features from tabletop in this game, and how the addition of those would actually add to the game's tactical variety without really detracting from anything else.

I suspect one of the reasons why Paladin isn't implemented yet is because Larian has not yet been able to program choice-based reactions yet, if they are even able to. Paladins at their core revolve around their ability to smite, after all. Spells like shield (which increases your AC until the start of your next turn, but you would only have the option to use it if an enemy attack overcomes your base AC to begin with, so there is no case where it'd be completely wasted) are vital to a Wizard's survivability, but the lack of it means Wizards are suddenly pigeonholed into ranged spellcasters rather than the battlemages that are possible in tabletop.

If Larian is able to get ready actions and the dodge action working, it'd be a huge boon to the tactical variety in this game. Solasta has a very basic implementation of ready actions at the moment (restricting it to attacking the first enemy that moves within attack range), but that and the dodge action are proactive options that do wonders in making players feel as if their turns are not being wasted, because both are very deliberate actions with a potential future payoff that can occur during an enemy's turn. Right now, there are plenty of situations in BG3, especially in regards to melee characters, where you can do little but move and then just end your turn doing nothing else if there is nothing in range to interact with. Maybe BG3 could go a step further and have a character ready action a shove on the first enemy to climb that ladder in front of them, instead of otherwise climbing down that ladder in order to get swarmed by the enemies below or stay up there doing nothing?

---

Quite frankly, I don't really want to bring up the argument that BG3 is considered by a faction of the community to be a D:OS2 mod, but a lot of the design decisions outlined above make it difficult to think of it as anything but that. The ultimate proof to see if Larian can shake off this image is if they ever get to de-couple Jump/Disengage, along with implementing ready actions and proper reactions into the game. I mention Jump/Disengage because when one really thinks about it from a programming standpoint, they appear to share the same traits as the various mobility/escape skills had in D:OS2 (Tactical Retreat, Phoenix Dive, Cloak and Dagger), which let you move large distances without provoking an opportunity attack if a melee enemy was right next to you. The closest thing D:OS2 had to a ready action and reaction was the Huntsman skill Reactive Shot, which let you execute up to 3 regular shots against enemies that entered an area of effect that the player marked out, but otherwise were forced to attack the first 3 enemies moving within with no choice input.

That's not to say that the developers aren't listening, as they did get rid of cantrips generating field effects on their own. But from a realistic standpoint, getting rid of field effects from cantrips was probably a very easy thing to do within their engine, compared to everything else being suggested here and elsewhere on these forums.

That said, I also understand that Larian is placing emphasis on their homebrew mechanics because of how their previous two games turned out. The D:OS games did not have a strict build/class system, they were free-form with an action economy that had everything from movement to actions taking from the same pool of resources. Trying to force that into a system such a DnD which splits up resources between movement points, standard actions, and free/bonus actions is another thing entirely. D:OS2 encourages you to go wild with your plans (so long as the game's encounter design allows you to, at least), while DnD deliberately restricts what you can do in order for you to weigh risk/reward and come up with creative solutions in spite of those limitations.

I should make it clear that there are some things that I find problematic, but from a logic standpoint, they are extremely low on my personal list of priorities compared to everything else mentioned above. They are food healing, barrelmancy, and to a lesser extent field effects. While cheap and immersion-breaking in most cases, they are ultimately still extra tactical options at the end of the day, and getting rid of those only means removing tactical options without really changing the balance in any meaningful way in their place. It's why I have no particular strong feelings towards loaded dice since it's really just a difficulty option setting. My grievances are mostly about the game's encounter design being largely balanced around everything else mentioned above, which do far more to restrict options instead of encouraging new tactics by their very inclusion. Address the above first, and then everything else should fall into place rather neatly.

While this might seem overly pessimistic, I imagine there are people on this forum who don't see things this way, and would have reasons of their own to refute the above points and try to convince me otherwise. I would encourage them to step up and give their thoughts here.