Larian Studios Forums Baldur's Gate III Baldur's Gate III - Suggestions & Feedback Robust numerical analysis of BG3 vs RAW DnD combat (AC/HP/Advantages)

I've already touched this issue on other places here but I feel this topic deserves its own thread. Some of you did read the reddit post with quick analysis of the HP vs AC issue. I've taken this a little bit further and incorporate proper fight simulation with ciritcal hits and misses and with advantages and disadvantages. My goal was to compare RAW (Rules As Written) DnD with modified BG3.

The testing scenario contains 1st level immortal STR 16 Figher vs 1 million of Goblins. For simulation I did use standard Monte Carlo method with Mersenne Twister random generator. I did some tests with even more random cryptographic generators but results for rolling dice were roughly the same.

For modded BG3 I used Goblins with AC 8 and HP 14. Also 80% of attacks were made with advantage and 20% were made normally or with disadvantage to simulate backstab and high ground rule. For RAW DnD I used Goblin with AC 15 and HP 7. 80% of attacks were made normally while 20% have either advantage or disadvantage. This simulates no backstab nor high ground advantage.





Results are surprising. Total time for killing 1M of Goblin is roughly the same in BG3 as in RAW DnD. Larian manages to keep same average numbers while significantly reduce the long miss streaks. Those horrific max values of round-to-kill-a-goblin for RAW DnD are very rare but number of goblins which takes more then 4 round to kill in RAW DnD are significantly high. BG3 is more consistent (see the standard deviation) than RAW while possibly a little bit slower judging from minimal value and median. Kudos for them for changing martial combat to feel better. But this change has very deep consequences (see e.g. this thread), including very big indirect nerf of:

• damaging spells and abilities vs DC (no advantage for those spell, enemies have more HP but same DC)
• spells and abilities which grant advantage or impose disadvantage (because backstab or high ground is easier and cost no resource and everybody can use it, no team cooperation is needed)
• spells or abilities requiring concentration (concentration needs to be hold for a little bit longer due to higher median but concentration checks are much more frequent due to damage from surface effect)

If we are telling Larian to discard AP/HP changes and backstab/high ground rule I think we should present an alternative system which gets rid of those long miss streaks or at least the bad feeling associated with them. My proposals so far:

Do not interper all misses as misses

This is certainly not a silver bullet but it can help especially with bad feelings. Depending on how close was attack roll to AC, there could be a hit with no damage accompanied with an animation (sword hit a shield or helm, sparks, dust, chips) or sound (taunts, shouts). For example we could have some uniqe shout after 4 misses in a row.

Pseudo-random die roll distribution

This is implemented especially in competetives games like Dota 2 (see their wiki). In this implementation the event's chance increases every time it does not occur, but is lower in the first place as compensation. Results are more consistent but statistically the overall probability is same as true random generation.

Simultaneous enemy AI

This is another indirect method. The miss itself is not that bad if we can strike again as soon as possible. This may be a reason while some people still advocates for RTwP. In RTwP everything is happening simultaneously and you need to wait for next attack roll 6 seconds in max. But in turn-base mode, when every enemy takes their turn in series you may wait a whole minute! Since RTwP is off the table let at least grouped enemies play their turn simultaneously.

Larger party and smaller groups of enemies

This is controversial proposal and may go either way, but since we are brainstorming ideas here lets include it. The general idea is to reduce the ratio of player controlled characters vs number of enemies to allow player simply play more often. And also elevate the party concept of BG. Maybe your 3 warriors missed this turn and your main wizzard is out of spell slots, but your Arcane Trickster still has a Magic Missile to not risk another miss.

And that is all I have at the moment. What are you thought about those solutions and about the analisys itself? Did I make a mistake maybe?

All you need to do is cast Sleep in BG3 and in Solasta and you don't need calculations to know how different it feels. =)

In Solasta I dropped 3 goblins that were hiding behind cover with bows with a tactically placed Sleep. And I couldn't get that spell slot back after combat by just clicking on a button. It was a choice that mattered right there, and in the immediate future. In BG3 I dropped one goblin that woke up immediately. I never prepared Sleep again.

BG3 combat feels like a cheaty DPS king of the hill race in comparison. I don't exactly feel like a tactical mastermind when I position or bonus jump my guys behind an enemy for free backstab advantage before each attack.

The main thing I want out of BG3 combat after 5e rules and stats have been restored, is simply that the AI turns go faster. I won't mind the miss streaks if I don't have to wait that long for my turns. The AI is really taking forever and that's why missing feels extra bad.

I do want more animations that reflect the armor and shield deflections that are happening. BG3 is falsely making the characters look incompetent every time they miss. Armor doesn't make you miss, it soaks all the damage.

Simultaneous turns of enemies can cause a terrible mess.
First, it requires a much better AI, which is always a problem in games.
Secondly, it is much more difficult for the player to follow what the opponents are doing, which is important.

Various miss animations in the long run are unlikely to help either, because it's still a miss.
Maybe if they introduced a PoE system where to reduce the number
miss Obsidian introduced a graze mechanic which made you deal less damage instead of a miss.
Only this again can cause problems without increasing HP.

Larian doesn't think this is what will sell copies of the game; they know DOS2 sold millions of copies, so the closer to BG 3 is to that, the more they anticipate they will repeat those results.

(For the record I agree they should try and remain closer to the source material for rules; HP bloat makes caster's feel awful to play and demonstrates a fundamental mistrust and misunderstanding of 5e)

The game will certainly have different levels of difficulty. The current difficulty level is hard (or very close to it), on lower difficulty levels HP will certainly be lower, so it shouldn't be a problem.

The principle needs to be how to make the most out of the D&D system rather than start changing how it works. Changing it will lead to a chain reaction which we are already seeing. Lowering AC and increasing HP makes all spells that target DC or HP weaker. Then you need to address that. Then you need to change something else.

In the case of misses, they should just embrace the fact that armor soaks all or nothing in the AC system. Animate those soaks. It would also give better visual feedback WHY you are not doing damage. Which gives you a clue which tactic works best against that enemy.

And I want my tanks to feel like tanks instead of elusive butterflies. I want to hear metal clashing instead of a woosh.

Just that doesn't fix the reason why they changed it.

Is it possible to get numbers for HP bloat and no king of the hill or backstab advantage? Just RAW chance to hit.

Then maybe look into a change in damage with spell save vs goblins to show the nerf.

Afterwards there could be more changes from the 5e RAW simulations for:

The change in Hp bloat and AC and the change in cantrip for firebolt (extra fire over time),
how much longer a PC can survive with the abundance of eating food every BA
How much more damage a goblin does with infinite flaming arrows spells, etc.

Even form this first simulation and graph ,you can already extrapolate the slippery slope Larian went through:

Players complain about missing too much, decrease AC
Enemies died too quick, increase HP
Enemies were being hit but still lasted too long, grant non RAW advantage situations

Mix this in with PC food gobbling, infinite scrolls/spells/bombs/arrows for enemies, ground effects, barrels and can trips. Also realize this feedback only applies to lower level combat, where later on the HP bloat and AC reduction will make spells with saves even more useless than attacking directly against AC.

This sums up my feelings about it quite well. I wouldn't object to having smaller battles either (or a larger party for that matter ).

What about increasing the hit chance through surfaces i.e ?
If you stick to the rules for every spells/cantrip and HP/AC.

Exemple with fire bolt.
- 1D10 damage
- OR surface effect if you target the ground
=> 1D4 damage/turn if you stay on the surface + "burning"
=> Burning : -2 AC (because you're not totally focus on the combat, you're burning)

Exemple with frost ray
- 1D10 damage
- OR surface effect if you target the ground
=> +2 to attack rolls against you (trying to stay on your feet, it's easier to hit you)
=> Saving throw when you move not to prone or when you take damages (advantage on you if your face is on the ground)


This is only an exemple but I think surfaces (and more) could be treated as an addition to D&D instead of becoming a "mix".
This kind of thing could reconcile the best of both worlds and add many value to the game.

Your wizard would have to choose : surface effect to help my companions or damage to kill ? What should I do ?

Another exemple could be higher ground.
A reasonable +2 or +3 to attack roll seems fine to me if you're higher. Important, but less significant than an advantage.

Of course many other things has to change for it to work but combats could become way more tactical if Larian try to offer us possibilities instead of mechanics that totally guides things.

My complete suggestion about combats is for another day.

1)Did anyone tell you you should work in the game industry? ^^ I'm not able to comment on you making a mistake when it comes to the data interpretation, I used to work on way easier problems than the analysis you did here. But since I took part in most of the discussions here please see my comments below.

2) Remember any conclusion reached in this thread must be added to the general feedback or separately as one of the mega-threads ! Otherwise Larian will probably never see it.

3) My comments.


Do not interper all misses as misses

A)I think that's a very important point and I love that you came up with it. It doesn't solve the problem per see but it does both bringing the game closer to tabletop ( where a miss is interpreted by your GM) and gives more immersion. I love the idea of more outcomes added to miss than simply nothing happening.

B) One of the cheapest and easiest way to achieve it is to add voice lines on miss /critical miss. It serves the purpose you described here aswell as several others after me. Like bounding more to your character's personality.

Pseudo-random die roll distribution

This might decrease the general frustration of too many misses one after another but will effectively be a major buff for players who are already powerfull due to other factors like sleeping being what it is right now.

Simultaneous enemy AI

I've talked about it in a different thread about combat lenght. The thread itself was very generalistic but we came up precisely with this idea in the end. Despite the " issue with tracking enemy movements" it doesn't require a swap to RTwP, is an realistic upgrade to TB in general and I simply think it's perfect when it comes to reducing the wait time for players.

If 2 of my characters can act, then 3 enemies and then 2 of my characters again the 3 enemies can act all at the same time. In the end what you need to track is your hp not what the enemy does one after another.

Also if done well enought it's a huge upgrade to TB games in general I think.


Larger party and smaller groups of enemies

A Larger party will result in more actions made by the player( side note:good thing!) but due to more permutations of skills, buffs the player even more. While it counters the misses I think the power creep resulting in this requires more enemies or stronger enemies to begin with which would itself give the average amount of misses per combat at the same rate.

I see mainly issues with this solution as it doesn't solve the right problem. Not even speaking about the fact it increases the combat lenght which was an issue itself mentioned multiple times. I wouldn't bring this up as a solution to the reduction of misses amount even if it COULD be one admitting the enemy power is balanced vs 4 players.

In conclusion I would concentrate on :
Simultaneous AI + Misses interpretation as a solution to the problem.

This is amazing. Thank you for taking the time to run such a detailed simulation to dive deeper into a very interesting discussion. One thing I am curious about - not sure if it's an excessive amount of work to do - is if we adjust the input data to match the game-stats better, I wonder how that changes the results?

I ask this because the 8 AC, 14 HP ratio may be an exaggerated ratio.

Some thoughts on the conclusions:




Can we consider this a universal nerf, if the increased HP falls within range of other CR 1/4 creatures? For example, 14 HP is definitely high for goblins RAW, but it isn't an absurd amount for CR 1/4 foes. In 5e, we've many CR 1/4 monsters with similar or higher HP:

• Albino Dwarf Warriors (Tomb of Annihilation) - 30 hp, 13 ac
• Kenkus (Monster Manual) - 13 hp, 13 ac
• Giant Frog (Monster Manual) - 18 hp, 11 ac
• Ash Zombie (Lost Mines of Phandelver) - 22 hp, 8 ac
• Abyssal Wretch (Mordenkainen's Tome of Foes) - 18 hp, 11 ac
• Blink Dog (Monster Manual) - 22 hp, 13 ac

The point is that a HUGE range of HP/AC ratio fits within the same CR (1/4) as goblins under 5e RAW. If anything, it's a extremely low ACs (8, 9) present on certain goblin types that might be the problem.

If Larian had chosen one of the above to use RAW instead of adjusting goblins, we might end up with a similar result, and it'd still fall inside the intended balance of 5e D&D.




The main benefactors of easier to obtain advantage/disadvantage would be the martials IMO - since most casters gain tricks to impose advantage relatively early. I'm not 100% against this rebalancing since casters tend to be more powerful than martials in 5e anyway. We do have to factor in the additional Bonus Actions (Shove) and easy resting in BG3 to the overall caster vs. martial balance equation though. Ultimately, I do think getting advantage too easily is overkill with the lowered AC (since it messes up criticals and also bless, etc), and I'd like to see that adjusted.

In regards to the team cooperation aspect, would you mind elaborating what you mean by that? Or is that referenced in another thread?




Agree 100%, non-dodgeable surface effects fundamentally breaks 5E imo.

Pretty cool analysis here, just goes to show that we can't just port pure martial classes to BG3 without change. Missing streaks are ok in pnp, because you or the DM can narrate those as an intense fight. But you can't do it in BG3.

I don't think adding animations for miss streaks will resolve the issue - half of the time I am not really looking at it anyway.

Using pseudo-random dice rolls can be a solution, but it will add too much complexity to the game and confuse players. Is the miss counter attached to a PC, enemy, or a PC/enemy pair? How would switching targets effect your chance to hit, etc.

I think lowering AC/increasing HP is the most straightforward way to do this. Plus, there are plenty of creatures at CR 1/4 at similar HP/AC levels, as Topgoon mentioned above. So you can just think about goblins as re-skinned enemies that already exist in 5e.

Should some spells be adjusted for it - perhaps.

Вы должно быть шутите? Какой это уровень сложности? Прекратите заниматься подсчётами и просто осматривайтесь повнимательнее вокруг. В игре, (спасибо разработчикам!!!) Огромное количество возможностей уничтожить противника без магии, атак ближнего или дальнего боя. Закидайте врагов огненными или кислотными смесями, скидывайте их с обрывов, валите на низ разный хлам, наэлектризуйте воду и так далее и тому подобное. Если использовать все предоставленные инрой возможности, на данном уровне играть становится довольно просто. На мой взгляд даже слишком.

Thanks for more precise data. I will adjust the simulation. I am also doing some other changes but you know, it's hobby project. I am or may not finished it


By cooperation I mean you normally need at least two characters to get advantage. Warrior get advantage when wizard cast Hold Person or cleric cast Guiding Bolt. Or another warrior just graple the opponent. You are not supposed to get advantage being low level, alone, on every turn.


You are right. I also think about it would be nearly impossible to implement PRG in DnD/BG3. There are too many dynamic factors and relations.


Actually yes, haha. But to your comment. I agree that combination of Simultaneous AI + Misses interpretation looks as the most promising. Rhobar121 mention the difficulty for the player to follow what the opponents are doing which is valid argument. It could create a mess, especially in large-scale fights. And even in simultaneous execution some actions nees to happen in order, like casting bless or charm person. But all of this can be decided on background and the game will just present the result in one compact execution.

@Zahur Don't forget to reference this discussion in the feedback thread against "mega threads" to make sure larian takes a look at it. Would be a shame if they missed your work.