Data science investigations into the mechanics of the world's greatest role playing game

[Morrus]

This is... interesting!

https://github.com/cyberscribe/d20datascience#d20datascience

I'll leave it to better mathematicians than I to unpack, but after plugging in AC, hit mods, HP, damage, and a ton of other stuff, cyberscribe on github seems to conclude that AC and hit mod, and inititiative, don't matter much (except in extremes), but damage and HP are of high importance.

"In short, an entity's ability to deliver damage (dam_max) rapidly and take damage over time without dying (hp) are the most significant controlling factors in combat success, and should be considered as the most influential when determining creature challenge rating."

 

[Zardnaar]

And this is a surprise? I figured this out in 2014 with the -5/+10 feats. The best debuff is death and if you can out last or out heal the opponents you win.

Basically reducing damage or stacking temp HP is good. The old fiend pact fighter1/Warlock XYZ, Abjurer and Light clerics, lore bards all contribute to outlasting the opponents along with things like the life domain and healer feat along with Life cleric 1/Druid or Bard XYZ. With bounded accuracy AC doesn't matter that much and AC 15 is more or less the average and even at 18 its not that drastic a change. Even thief with/healer kit and bonus action 1 hp healing is annoying.

 

[abirdcall]

Not sure about the math either but supports my intuition so it must be true.

 

[Staccat0]

Did they factor in rerouting a river into the mouth of the dungeon?

 

[jgsugden]

When you factor in only the surface level of the game, sure. However, there are too many variables unincorprated in a model like this to get an accurate total picture and many of those are campaign specific.

 

[Oofta]

I've tried my own simulations with code, and as far as I can tell it doesn't make a huge difference. Things like high ability scores, what type of monster you're facing, style of play are all just as important if not more so.

So yeah, I take white-room math "proofs" with a grain of salt. Like much of theoretical physics it might be right or it could be totally BS.

 

[abirdcall]

I've tried my own simulations with code, and as far as I can tell it doesn't make a huge difference. Things like high ability scores, what type of monster you're facing, style of play are all just as important if not more so.

So yeah, I take white-room math "proofs" with a grain of salt. Like much of theoretical physics it might be right or it could be totally BS.

I think they are more useful for what they discredit.

For example, some say that initiative bonuses are very powerful. I think that's silly.

 

[5ekyu]

I think they are more useful for what they discredit.

For example, some say that initiative bonuses are very powerful. I think that's silly.

Whether init matters much or not depends on what you do with it.

In the post i think they referenced how init can affect "real contests" due to severly impactful openings.

Did you open your high init start with a fear or hypnotic pattern? Did it divide the enemy costing them attacks, position, control?

The basic problem with analysis like this and how "conclusively" it presents what is "better" and what "matters" is how much it leaves out.

AC matters more when you get to disadvantage the enemy attacks.

Init matters more when it delivers effects that control the ability of enemies to even get to make attacks. How does your high damage rogue do with blindness after the save fails against the high AC opponent whose AC matters little?

Its very easy to calculate highs and lows if you reduce combat to closets and grunts, but they say almost nothing about actual rpg performance in any type of significant encounter i can remember from actual play.

 

[Satyrn]

This is... interesting!

https://github.com/cyberscribe/d20datascience#d20datascience

I'll leave it to better mathematicians than I to unpack, but after plugging in AC, hit mods, HP, damage, and a ton of other stuff, cyberscribe on github seems to conclude that AC and hit mod, and inititiative, don't matter much (except in extremes), but damage and HP are of high importance.

"In short, an entity's ability to deliver damage (dam_max) rapidly and take damage over time without dying (hp) are the most significant controlling factors in combat success, and should be considered as the most influential when determining creature challenge rating."

Interestingly, the DMG already calls those out as the two key factors in determining CR.

 

[Guest 6801328]

And this is a surprise? I figured this out in 2014 with the -5/+10 feats.

You know what I most admire about Zard? His ability to pat himself on the back with sufficient vigor to fracture his arms, and then to still be able to type forum posts. Perhaps he's really Thri-Kreen?