General:Developer Deep Dive—Item Sets Part 2

Originally published October 15, 2024. The original article can be found here.

Delve into how the team balances item sets, both prior to and after launch, with this next developer deep dive blog.

Previously, ESO Combat Designer Nadav Pechthold shared the process behind the creation and production of the game's many (over 650!) item sets. In this second part of our conversation, we explore how the team goes about balancing item sets during their development and after they're live. Missed part one? Check it out here!

Power and Balance

When starting to think about item set balance, where do you start?

Nadav: Item sets follow a standard called Set Bonus Efficiency, or SBE for short. SBE means that the average power of an item set per slot is roughly the same. The two-piece bonus on a five-piece set might be weaker than the two-piece bonus on a two-piece set, but if taken as a whole, both sets will give you almost the same amount of power per item slot.

We use a set of mathematical formulas to determine the SBE with positives that increase and negatives that decrease the value. Giving stats and increasing damage or duration are some examples of things that can increase SBE, while complex proc conditions, cooldowns, and "curse" aspects are examples of things that lower SBE.

In short, we compare the good stuff the set gives you with how difficult the set is to use and divide that by the number of pieces you need to complete the set.

While the team is testing a set for balance, what kind of things are they looking for?

Nadav: The SBE gives us a great starting point by using objectivity to balance each item set, but often, the subjective nature of our game can skew said balance. An item set might overperform if combined with the right abilities or underperform if it's too difficult to use in actual combat. That's where testing comes in.

Each item set goes through numerous PvE and PvP playtests. Different members of our QA and combat teams try each set with various builds and playstyles to best simulate our player base in real-use cases.

We take notes throughout each playtest and run surveys and discussions with our testers to see how sets feel from the perspectives of the user, of people using the set in your group, and when going up against a set in PvP.

By thoroughly testing each set with real humans, we can reveal balance issues that our mathematical formulas might be unable to identify.

That doesn't necessarily mean a set is completely balanced at that point, right?

No. Balancing is difficult and sometimes we simply don't get it right. We have over 650 item sets, seven classes, and weapon and guild skill lines. We can never perfectly predict how these abilities will interact with each other, and our players always find unexpected and creative ways to use them.

That's why it's important to lean on as many sources for testing and monitoring what players are doing with our item sets, so that over time, we can tweak sets to be more balanced.

How does the team view power for different types of item sets?

Generally, an item set's SBE is compared specifically for its intended content. So, a set you'll find in a dungeon will be balanced around dungeon gameplay, and an item set you'll find doing PvP activities will be balanced around PvP.

It's still critical we test item sets in all possible environments, though, as a set intended for one type of gameplay might be stronger in another. For example, if we made an item set that deals a burst of damage to an enemy as part of a Trial set, it might be perfectly in-line with our SBE standards. However, when used in PvP, the burst might allow for alpha strike playstyles to get an unintended strong edge, which could lead to PvP-focused players being forced to participate in PvE content to get the best gear or even result in a disruption of the game's item meta.

So, while you might find a PvP item set that performs well in PvE and vice versa, they should always perform best in their intended content.

Does the complexity or the effort required to earn a set factor into its power? For example, when comparing a crafted set to one that drops from a dungeon boss?

Nadav: The short answer is yes, crafted and overland sets tend to be less complex, but the best way to think about it is to consider the players who are seeking a set out. For example, both Overland and Trial sets are designed for PvE, but an Overland set is likely to be acquired by players who aren't going to do Trials.

Maybe they're looking for an easier experience, so they'd want something more passive, whereas because Trial players are highly specialized, the sets they can find can be a little bit more active in the way they operate.

That said, when it comes to sets like those found in Trials, we do try to mix it up and give some simpler Trial sets on top of complex ones. We want players to have that option and be an efficient Trial player at any level of experience.

You're also balancing sets against the content and the game, right? How tricky is that?

Nadav: We try to use the SBE as our guideline, but balancing sets against content can be tricky, as both PvE and PvP are complex in their own ways. For example, if we buffed every item set in the game, we'd make every single dungeon boss much easier, and we can't do that.

Of course, we always need to consider PvP when balancing sets, too. We don't want to create item sets that have the potential to be abused in PvP and make the game less fun, and we don't want item sets that will encourage a limited number of viable playstyles.

Item sets need to be balanced in many ways

Public Tests and Launch

With the production and internal testing complete, what is next for a new item set?

Nadav: The team continues playtesting multiple times a week, but once all the teams have done their work and have completed their reviews, we add the item sets to the Public Test Server.

PTS is so important for us, and we monitor player feedback very closely. There's a LOT more of them than there are of us, and they're very creative, often thinking of things we haven't and finding things that we need to tweak.

If we feel a change to an item set will greatly alter how that set will function or perform, the team will discuss things together before locking the changes. Whatever tweaks we decide to make, big or small, we document everything.

How soon after a new item set goes live do you look at it?

Nadav: We look at it day one! Every time a new update hits live on PC/Mac and then again when it hits consoles.

We monitor our game's data using analytic tools, and our developers are also playing each new update to test out the content and the sets along with everybody else. We listen to our players, too, by reading forum posts, reviewing external websites, and watching content creators on Twitch and YouTube.

If at any point we see an item set with too low or too high of an adoption rate, we examine it right away. In some cases, when we examine a set, we might say to ourselves, "This item set needs to be changed, but it's not an emergency. We can wait until a future update to address it." If there's an emergency where, for example, a set is widely broken, we work to make the change as quickly as possible.

Do you have a good example of an item set that had to be balanced after it went live?

Nadav: When we first developed Sload's Semblance, we wanted a new set that dealt Oblivion Damage, which ignores mitigation, to combat really tanky players with high resist, health, and armor. We knew it had the potential to be devastating in PvP, so we were cautious about testing it in different PvP scenarios. When the set went to PTS, we closely monitored how players were utilizing it, and while it looked strong, we found no major issues at the time.

However, when the set finally went live, we saw many more players begin to use it in ways we didn't expect. They quickly figured out how powerful Sload's Semblance could be with the right build and set combinations, and it ran rampant.

I remember Sload's Semblance being a menace!

Nadav: People were hurling purple orbs of death all over the place! If we see an item set spike to the top of the charts while everything else is declining like that, that's when we take a harder look and ask, "Is this healthy for the game?"

As a result, we changed that set so its damage is spread out over a six-second duration and has a cap on how much it can do per tick, giving it more counterplay, and it's now in a much healthier place.

Oblivion Damage presents unique challenges

How challenging is it to balance an item set after it is live?

Balancing a live item set is hard. Just like players, we don't want our favorite gear to change, but sometimes, we have to for the sake of the health of the game. Sometimes that means a set gets buffed, sometimes it means it gets nerfed.

There is never a simple solution when it comes to changing our sets, and it is a balancing act. Buffing sets can seem to be easier than nerfing, but that can also lead to power creep and an arms race. For example, say there's one set that is extremely powerful. It makes more sense (and is much safer) to nerf that really strong outlier to preserve the game rather than buff everything else. If we can solve a problem with a targeted change, that's what we want to do.

When we buff or nerf a set, the intent is never to punish players, but to maintain a healthy game where no one item set becomes the "right" choice for all players. We want all item sets to be viable to different players.

Thank you so much for talking with us about item set creation. Before we go, does the team have any shoutouts to the #ESOFam they'd like to make?

Nadav: We've been fortunate enough to meet up with some of our fans at events and by random chance. The passion, excitement, and love they have for ESO continues to fuel that fire that pushes the team to make even cooler gear for them to play with. So, thank you to all of you who love our game as much as we do, and keep that feedback coming!

Another big thank you to Nadav for taking the time to dive deep into the process and philosophies behind the creation and balancing of item sets. ESO's item sets continue to grow and evolve with every release, and the team is always working hard to create new sets while keeping existing sets balanced. If you missed it, be sure to check out part one of this deep dive, too. What's your all-time go-to item set?* Let us know via X (formerly Twitter), Instagram, or Facebook.

*Editor's Note: Every stamina-focused character I have uses Hunding's Rage at one point or another!