Enemy Damage Types

[Galaxy Brain]

Hello everyone! In researching for a separate project, I took the time to scour as much as I could about the various NPC factions and what types of damage the bring to the table, and brought those to my own table!

 

 

Looking through a grand total of 117 enemy sub-groups, and a whopping 4979 instances of typed damage, I found the data you see above. The top portion compares the total number of instances of each damage type, so for example out of 4979 instances, Energy made up ~18%. The bottom shows the types compared to how they are split among the groups, so out of 117 factions you are likely to encounter energy ~21% of the time. 

 

This confirms some things we already know, like why it is important to get S/L defense, but I think it is truly eye opening to see the "Encounter Chance" per damage type like this. I had no idea how nearly non-existent Cold damage is for example! 

 

Click here for a link to the full sheet

[Caulderone]

Conceptually, I knew S/L/E were way up there, but I never would have imagined them being ~75% of all damage across mob types.

I never would have expected there to be so little Fire damage, either.  Cold is about what I expected.

 

edit:  Any way to know what percent of those N/F/C/T/P attacks actually have a S/L component?

 

edit2:  A quick glance through the chart, and it appears the Circle of Thorns and the Talons of Vengeance are among the most diverse groups.

Edited April 3, 2020 by Caulderone

[Galaxy Brain]

1 minute ago, Caulderone said:

Conceptually, I knew S/L/E were way up there, but I never would have imagined them being ~75% of all damage across mob types.

I never would have expected there to be so little Fire damage, either.  Cold is about what I expected.

 

edit:  Any way to know what percent of those N/F/C/T/P attacks actually have a S/L component?

Cold honestly shocked me with how infrequent it shows up.

 

As for finding data on how many are split damage, that is doable but will take a bit if time. Off the top of my head... it may be easier to count which ones *aren't* a mix

[Caulderone]

Snakes (50%), Arachnos Widows (44%), Hydra (50%), Snakes (24%), Vahzilok (27%) - Scary to actually see those Toxic numbers for certain defensive sets.

edit:  Missed Arachnos Tarantulas at 27%.  edit3:  Arachnoids (50%), how could I miss those.  I new they sucked with Toxic.

 

edit2 again:  I wonder if the same data would show how many attacks are positionless.  We always talk about positionless Psi attacks as weaknesses for certain sets.  I wonder how many there actually are.

Edited April 3, 2020 by Caulderone

[Galaxy Brain]

Going through again I could pick out the positions as well.

 

As for this data, knowing how defense works and that S/L is in over 1/2 of all powers, capping to that should cap you vs half the game regardless of the second type of their attacks. 

Edited April 3, 2020 by Galaxy Brain

[Caulderone]

Just now, Galaxy Brain said:

Going through again I could pick out the positions as well.

The things missing positions are the most interesting.

 

1 minute ago, Galaxy Brain said:

As for this data, knowing how defense works and that S/L is in over 1/2 of all powers, capping to that should cap you vs half the game regardless of the second type of their attacks.

Yeah, that was kinda why I asked about the other types and split type attacks.  We knew S/L defense was really good.  It may be even better than I thought.

 

S/L defense and S/L/E resists have frequently been lots of builds targets.  They may be even better than expected.

[Bopper]

Correct me if I'm wrong, but don't you have another analysis sheet like this, showing the different types of damage in the game? If so, what is the difference between the two analysis sheets (sorry, going off memory and haven't had a chance to dig for what Im talking about). 

[Bopper]

22 minutes ago, Galaxy Brain said:

Going through again I could pick out the positions as well.

 

As for this data, knowing how defense works and that S/L is in over 1/2 of all powers, capping to that should cap you vs half the game regardless of the second type of their attacks. 

If you do this, can you also pair everything? So columns with Smash-Melee, Smash-AoE, Smash-Range, Lethal-Melee, ...

[Caulderone]

7 minutes ago, Bopper said:

Correct me if I'm wrong, but don't you have another analysis sheet like this, showing the different types of damage in the game? If so, what is the difference between the two analysis sheets (sorry, going off memory and haven't had a chance to dig for what Im talking about). 

The other one has to do with mobs Resists, so how much damage they take.

 

This one is about their attacks, how much damage WE take and of what type.

[Bopper]

43 minutes ago, Caulderone said:

The other one has to do with mobs Resists, so how much damage they take.

 

This one is about their attacks, how much damage WE take and of what type.

Ah, that's right. Thank you. I knew there was a similarity but couldnt recall what it was.

[Call Me Awesome]

I seem to recall that years ago it was a "known fact" (for what that's worth) that 75-80% of all damage in the game had a S/L component to it, I wonder if that's accurate and if it's still the case.  I always took that as a given when planning a build and while I haven't actually crunched any numbers it does seem to work out close to that.

 

In any event S/L/E seem to account for 3/4 of all damage, and unless I'm mistaken the majority of Energy also comes with a Smashing component... with Ice frequently coming with Lethal and others likewise somehow including a S/L component it certainly makes sense.

 

Cold being the rarest damage in the game DOES surprise me however.  I knew it was rare but I'd thought it higher than Psi and Toxic.

[SuperPlyx]

On 4/3/2020 at 4:30 PM, Caulderone said:

The other one has to do with mobs Resists, so how much damage they take.

Got a link for that one?

[Caulderone]

3 hours ago, SuperPlyx said:

Got a link for that one?

 

[Galaxy Brain]

@Bopper, @Caulderone

 

Do you think you could assist me with weighing the enemy groups?

 

A flaw I noticed is that this is all assuming you were to just have an RNG cannon fire enemies at you from all groups, with no consideration towards how common certain groups are. For example:

• Tier 1 - Hydra are found really just in Perez Park, or even just a handful of specific missions. This makes the group very uncommon
• Tier 2 - Slag Golems, similarly to Hydra are found in 1 zone but it is a "main zone" (Sharkhead) and can be found in several story arcs
• Tier 3 - Praetorian Police are very common, but only to lvl 1-20 Goldsiders / some specific Trials. This tier is tricky as it is outside the "Found everywhere" rank, but still very common depending on your location and level range
• Tier 4 - Arachnos is found literally everywhere

Or something like that... if it makes sense? I'm hoping by weighing (and consolidating some mirrored groups) to get a more accurate representation of what we really see in-game.

[Bopper]

I'd be happy to help, but I'll likely have to rely on your guys' game knowledge as my skillset is mostly just number crunching. Are you wanting everything weighted, or is it best to weight things based on categories (red side, blue side, gold side | low level, mid level, high level, incarnates | etc)? 

[Redlynne]

1-10

11-20

21-30

31-40

41-50

51+ (Incarnate content, basically, but also some high level zones and task forces)

 

Blue

Red

Gold

[Galaxy Brain]

12 minutes ago, Bopper said:

I'd be happy to help, but I'll likely have to rely on your guys' game knowledge as my skillset is mostly just number crunching. Are you wanting everything weighted, or is it best to weight things based on categories (red side, blue side, gold side | low level, mid level, high level, incarnates | etc)? 

Those should all be included imo, and probably ranked as such (blueside > redside > goldside) or at least have marks showing what groups are exclusives. The lvl range I think would carry a bit more weight for ranking as say... hellions probably dont affect the meta much only capping at like lvl 15, while the new banished pantheon enemies being a lvl 50 group in a major arc will be tested more often in a way where you'd care about the damage types?

[Bopper]

I don't know exactly how you want to break things up, but would you want it broken out by our odorous arcs? Take everything from the level 50 section and make that a resistance breakdown, them step down to the next, all the way through. Spit balling here, just trying to visualize how you want the end product to look

[Redlynne]

Another way to think of it is that at early levels, yes there are different damage types, but the capacity of a build to defend/resist those damage types is quite limited, so there's a pretty strong leveling factor going on early on.  Compare and contrast that to how much of a differential there is in the 41+ game where builds are either finished or nearing being finished and the capacities of PCs to influence how much damage they're taking from hits is VASTLY greater than it was in the early game.

 

Point being that you'd want to "weight" the damage type distribution of the early game brackets much more lightly than the later stages as PCs grow and develop.

 

If you want to think of it in terms of "shares" of weighting, you could just add +1 the "share weight" of each bracket so as to yield a relative weighting like so:

 

1-10 = 1 share each

11-20 = 2 shares each

21-30 = 3 shares each

31-40 = 4 shares each

41-50 = 5 shares each

51+ = 6 shares each

 

You then determine the "spread" of damage types within each bracket and then multiply that prevalence by the "share weight" progression for that bracket so as to arrive at a loose sense of how common/concerning damage types are overall along with WHEN the balance shifts around in different level brackets.

[Galaxy Brain]

I like that idea @Redlynne! Perhaps even combining shares if a group spans multiple brackets (ie, Arachnos being 1-50+ would net them a massive weight). 

 

@Bopper, I think the lvl brackets may be the best bet, along side a way to show what content they are found in (such as a specific side, or even a specific zone within a side).

[Redlynne]

1 minute ago, Galaxy Brain said:

along side a way to show what content they are found in (such as a specific side, or even a specific zone within a side).

Again, you can resort to "share weights" for that.

Each shared world zone that enemy appears in for that level bracket adds +1 share to their weighting.

If the group appears in instanced mission content, give them an additional +1 share.

 

So in the blueside 1-10 bracket, Circle of Thorns appears in King's Row, The Hollows and Perez Park (+3 shares) and also appear in mission instances (+1 share).

In the blueside 11-20 bracket, the Igneous appear only in The Hollows (+1 share) and mission instances (+1 share) including the Cavern of Transcendence trial.

 

Just that right there shows that the Circle of Thorns is MUCH more common in the 1-10 bracket than the Igenous are in the 11-20 bracket, and the relative weighting reflects that (kinda).

[Bopper]

It really depends on how many groups there are and how frequent they show up. Calling it a share is fine, but the value of the share changes based on level. I rather treat it as conditional probability. Given a set of parameters (level, alignment), this group will show up X% of the time. Shares can kinda do that as long as you stay within the confines of the parameters.

 

But when you want to expand your data to reflect multiple parameters (e.g. levels 30-50, heros and villains) you'll want to make sure the units are equivalent. 

 

But we'll hash all that out as we go. Just need to start categorizing enemies and where they show up and how often.

[Caulderone]

I'm not sure how comprehensive this is, but it could be useful for reference:

 

https://paragonwiki.com/wiki/Category:Enemies

[Galaxy Brain]

11 hours ago, Caulderone said:

I'm not sure how comprehensive this is, but it could be useful for reference:

 

https://paragonwiki.com/wiki/Category:Enemies

That's a actually one of the same sources I used! 🙂

 

I think the shares is our best bet (and can be copy pasted into my enemy resistances guide too when I go back over it...)

 

Im hesistant on getting as granular as counting every zone something appears in as a share, so maybe just having "appears in one location" vs "found in multiple areas" split could suffice? This really only applies to Blueside too, as redside amd goldside have fewer zones but they generally have more content per zone.

[Galaxy Brain]

 

Something like this could be a jumping off point

 

 

Faction is weighted:

All > 2 > 1, possibly blue>red>gold as well...

 

Area:

All > multiple zones > 1 zone > mission arc

Edited April 26, 2020 by Galaxy Brain