To begin, something that I really like about Last Epoch is the basic idea behind protections. Rather, I enjoy that they are flat values that can infinitely scale upwards to apply better and better resistance against certain elements, albeit with diminishing returns. However, the way in which they currently function constantly brings players to seek out other methods of mitigating damage, such as through glancing blows and sheer health. This is not due to the fact that they are not traditional percentages as in most ARPGs, but because they are tied to one’s maximum health stat as a scaling value.
For every point of protections you have, your effective health against that damage type raises by 1. Part of where the dilemma of protections come from is that, as you invest more in health, it becomes even less appealing to invest in resistances as you will need more and more to begin seeing effective resistance percentages. In addition, while protections come with much higher rolls than health, they are often not effective because they are only protecting against one damage type, while health mitigates all damage. Another issue with protections is related to its counter stat, penetration. Due to protections scaling additively with health, once enemies end up with massive health pools, penetration will do close to nothing against them. If every point of penetration lowers an enemy’s effective health against an element by one, and enemies lategame can have tens of thousands of health, penetration is rendered ineffective, even with the most extreme investment available.
The point about penetration is brought up frequently, and many have lamented about the over-presence of health and glancing blows for a while. However, with glancing blows on the agenda for being changed or removed, I feel that it is time to touch up protections so that they are on par with other methods of damage mitigation. I would like to propose a way to do this that not only makes protections more valuable, but also makes penetrations more effective.
My suggestion is as follows: Make protections scale against a level-determined value [comparable to base health] rather than maximum health.
In other words, say you need 100 protections to take 1/2 damage from an element at level 1, and then this caps at 500 at level 100, with 500 protection being 1/2 damage, 1000 being 1/3, 1500 being 1/4, and so on. Here are some topics that I think this change would address.
Building health would no longer make protections less appealing. Rather, protections would be just as effective at base and high health values, causing the player to actually be incentivized to balance both.
Enemies could be altered to have moderate, independent protection scalers that also go up with level, with high penetration investments actually causing noticeable damage increases as a result, no matter what their health is. As an example, consider that an enemy has a base scaler of 200, 500, or 1000 depending on tier. These would operate, in the eyes of the current system, as if they had 200, 500, or 1000 health.
Classes or items that grant large protection amounts would become noticeable. A void knight with this change would be able to feel their higher void resistance as they end up taking far less void damage than any other class, a change that is not as apparent when players stack 2k health with minimal protections.
Protections, with this change, would have a smoother curve of build impact from early to late game. Currently, protections are actually quite nice early, particularly void protection, but fall off as soon as the player gets health and glancing blows gear.
The exact protection scaling values could be debated further, but the main concept is to separate protections from health, while still requiring a higher investment as players achieve higher levels.
Anyway, please leave some feedback regarding this idea! It may not be the perfect solution, but I really think protections and penetration as player stats are in a bad state, and this may provide a better framework for balance against other mitigation factors.
EDIT: In reply to a Discord critique, I have slightly altered my suggestion to involve protection scaling values that adjust with level rather than having protection be tied to base health. It follows the same concept still.