What makes a stat "interesting"

[Shurrikhan]

I really don't think it's possible to make passive stats interesting, they can be made more "desirable" though. For example, if a stat gives you a drastic dps increase or scales really well to the point that it REALLY outshines others then you're going to want as much of the stat as possible. I don't necessarily think that's interesting though. To give you an example, a few years back WoW had a stat called armor penetration, it was the highest dps boost for all classes that dealt physical damage and it's scaling was ridiculous to the point where it became a godly stat. They kept having to nerf it and it's scaling over and over again but it STILL managed to outdo all the other passives, eventually they gave up and ditched the stat entirely in to something new.

This is something I've talked with others ingame about a fair bit, and on here a few months back. The way I see it, gearing itself (choosing gear paths, or gunning for a particular piece) can be made interesting through any of three ways:

• [Gear affects gameplay] You establish a gameplay style you like as made possible by gear (in our case, the only actual affector is Speed, but same idea), and try to maintain it in what even-value gear choices you do have, or start experimenting and rounding out new gameplay styles from what you don't have a choice over. A particularly high-stats drop could be your pride or, temporarily, the ultimate annoyance as you have to readjust things to support it, but regardless you do feel like there's significant interaction with your gear and stats themselves.
  
  
• [Class-optimals] There is a clear best and/or worst stat for your job, and you try to maximize the prior and avoid the lower. Unlike a more balanced setup that allows for more gameplay styles, this puts a greater urgency on gearing, as two players differing only in secondary stats could have noticeably different outputs.
  
  
• [Scalars] In ARR, Crit was originally recommended for most classes to a certain level, but after said level the crit chance as a portion of one's base chance being increased per stat point diminished, making it lose weight as the amount increased. De-exponential. Inversely, the portion of the remaining GCD affected by Skill Speed increased with the more Skill Speed one acquired. Exponential. Determination was linear. In this system, the most worthwhile stat is dependent on the other stats in play, and their relative equilibriums. Even if the stats are individually balanced at a given point, there's constant significant renewal of their weights as gear progresses.

I'm personally most fond of the first system, but not by much. I like the idea of mostly having control of my gameplay, so I like for gear to last, and/or for the variance not to be too huge, but for there also to be a number of distinct styles supported within any given tier collection of gear pieces. I don't dislike charting out gear paths to maximize my best stat and avoid the worse, or theory crafting for the best set of stats at a given time. They all kind of work. It's just a question of which has the most FF feel, or just generally is most attractive to most players.

____________________________________

On WoW's Armor Penetration itself, and similar quandaries:

The main issue with Armor Penetration was certain oversights in how the mechanic actually worked, not the least of which was that you could remove more than 100% of armor in order to then deal bonus damage. Armor Pentration removed flat amount of armor per point. However, because after exceeding 100% armor reduction the bonus damage being dealt as a percentile while the stats was accrued as a flat sum meant that those with lower armor were more vulnerable to its effects. If it took only 500, rather than 1000 ArP, to ignore the whole of a mage's armor, then by 1500, the mage would be taking twice the bonus damage as the other guy. It doesn't make much sense for the term, destroyed clothies in PvP, and in PvE it functioned as a nigh exponential stat. Were it to have capped at 100% armor, it would have instead been an auxiliary stat much like Accuracy (or in their case Hit and Expertise) that at most causes physical attacks to never lose priority even against armored enemies. Instead it caused hybrid attackers to even deprioritize their DoTs and magic damage in favor of direct physical sources, wherever possible. That's oversight, as you could call virtually any exponential scalar, especially after whatever point it starts to wholly overwhelm all other options, just as ArP did. It's quite avoidable. However, that example also wouldn't be applicable to XIV, as we don't have mob Defense anyways.

____________________________________

Thinking on it more, it is my opinion that [Class-optimals] is what will end up happening no matter what they do. At their most basic level with combat, MMO's are just number generators with a fancy user interface and sooner or later someone is going to figure out which stat yields the most dps increase. A blizzard dev once commented that he was shocked what players would do even for a theoretical 1% dps gain and that is inevitably what happens. You either make the stats good and then someone figures out which one is best or they become so weak they don't make much difference at all, at which point someone with a spreadsheet will theorycraft and figure out which one of those paltry stats offers the best increase. That is the nature of MMO's nowadays. I don't think it's a bad thing personally, but that again brings me to why I don't think stats can be interesting, because no matter what the stats do, someone will eventually figure out what makes the numbers the biggest, spread the word, and everyone will go for it.

The thing is, though, if the devs so desire, the same theory-crafting can be done to be sure that all stats are balanced. It just requires ensuring that scalars are of equal value to the class's tool-set. And an equilibrium or 0-derivative between different stats can be calculated just as easily as a proven (not just hinted at) "best" stat. I'd wager that the main reason class-optimal stats seems to be the given is that for many MMOs differing ideal secondaries are a selling point of class identity. In some cases this was even used in place of class-specific gear. It's leather, usable by any and all leather-users. Who should get it? The one for whom it has ideal secondary stats.

That being said, it's not like people are going to toss out an upgrade just it's not the BiS (due to secondary stats). An alleged upgrade can be rather disappointing, weight-wise, if non-optimal, but especially when the primary stat weight gains are as strong per ilvl increase as they are in XIV, there are no truly wasted pieces upon acquisition unless you already have equal or higher tier gear compared to the place you're looting. The odd thing about this is that this vertical progression above all paradigm has caught a ton of flak from the community, sometimes for its effects, and sometimes on principle. But if most pieces are to remain real options, either of two things have to happen: the primary stats must diminish the secondary stat weight gaps between pieces over time/progression, or the secondary stats themselves need to be more closely balanced. Otherwise you end up with either a smaller actual gear pool, or secondary stats changed to "auto-fit" the class wearing them (at which point you have only vertical progression left).