FFXIV, Job Balance, and How Communities Discuss Performance

FFXIV is having something of a banner raid tier at the moment, for reasons both bad and good.

On the good side – Abyssos Savage is being very well received as an interesting raid tier to play. It has interesting and engaging boss mechanics from first to last fight, stretches the performance requirements of players in a positive way mostly, and is a fun tier to progress through by virtue of strong music. Coming off the success of Dragonsong Reprise, high-end raiders have been eating well this expansion in FFXIV.

On the bad side, Abyssos has exposed some tuning issues with the current state of the game’s balance, but it has also been an interesting look into how MMO communities discuss the balance of their games.

It speaks well to Square Enix’s track record with high-end raid tuning that the patch 6.21 boss nerf to Abyssos: The Eighth Circle (Savage) was the first raid nerf in the game in six years. The team is normally so on-top of their balance targets that they haven’t had to retune a live boss since Heavensward. That is a hell of an accomplishment, given that sometimes Blizzard can’t even go six hours without retuning a boss fight! Their response to player struggles with the tier reveal an interesting window into how the game is balanced and also, perhaps, a potential flaw.

The Design and Balance Paradigm Of FFXIV Is Fundamentally Broken

This title sounds doom and gloom, so let me first say that “fundamentally broken” doesn’t mean unplayable or all that bad. FFXIV’s job balancing rules are built around the idea that boss fights in 8-player content (the balancing target for job performance) are not full uptime encounters for melee players. Melee DPS do the most damage by design because they are built to have downtime in fights, where a mechanical interaction or boss phase forces melee to be unable to attack the boss. Casters do a smidge less for the luxury of increased uptime at range, and physical ranged DPS do the least damage because they have no cast times and full mobility, so being forced out of melee means they can still hit the boss with their full rotation. Within each role type, there is a DPS tiering based on the utility brought to the raid. Caster jobs with a raise spell to bring back KO’d party members? That’ll cost you a bit of DPS performance. Have some sort of damage buff for the raid? It’ll cost you raw personal DPS.

In the past, this model made a lot of sense, because it kept things relatively balanced under ideal play. Melee could hit harder per ability and per hit because they wouldn’t always hit, casters get big nukes at the cost of needing to manage cast times around mechanics, and physical ranged DPS do less per hit because they can always hit the boss unless that target is completely un-attackable. In the end, performance bands tightly together and things look great. So what has changed?

Encounter Design Influences Job Design and Balance

A big part of the current dilemma facing the game is that boss design has become far more accommodating of melee play, with most common group strategies allowing for full (or virtually full) uptime. Melee suddenly look a fair bit better than most other DPS because their balance is built assuming they’re going to have to take a break from hitting the boss every now and then, and they rarely actually have to take that break. The game has become increasingly less punishing for most of the caster jobs, too – Summoner in Endwalker has exactly two rotational spells with cast times at all, and Red Mage is built around the Dualcast mechanic, spending short cast time spells to be able to then use longer ones instantly. Even Black Mage, famed for its immobility, isn’t that bad – in Endwalker with two charges of Triplecast and the power of Aetherial Manipulation, you can pool Triplecast for movement mechanics or simply stay put and then Aetherial Manipulate yourself to a party member who is actually respecting the mechanic (what a sucker that person must be). Physical Ranged DPS still get their softer damage full uptime, but when everyone else is getting full uptime, suddenly a physical ranged is looking a little weak.

How did this happen? Well, boss hitboxes got larger, max melee range got further and further from the actual boss model, and the number of mechanics that force large amounts of movement decreased. Both mid-tier Abyssos bosses are perfect examples of this, with Hegemone in P6 having a hitbox that reaches 3 times the width of the model in all directions, and Agdistis in P7 has a hitbox so large that you can hit it from the back of the room without consequences (not to mention that Agdistis also has no positional requirements for melee!). Mechanics like P6’s Chorus Ixou or P7’s Bough of Attis have little challenge for melee uptime because the positioning is very forgiving. For melee in particular, the increased number of auto-centering bosses also makes their lives easier, because you can camp a particular spot in the room and be relatively assured that your (now limited) number of positional-bonus attacks will properly be delivered at the flank or rear.

In the opposite direction for casters, high-movement boss fights can be very punishing to a caster who is caught casting at an inopportune moment. Black Mages struggle with a lot of fights in Endwalker because of rapid positional changes challenging their uptime even with their movement tools – Endsinger EX quad-planets did this, EWEX4 (still keeping that under spoilers just in case) has a lot of rapid movements that mean most PF Black Mages I’ve tried the fight with are just awful, and Red Mages have to very carefully consider movement. Summoners getting stuck in Ifrit phase at an inopportune moment can feel really bad, and while the temptation is to use Swiftcast to get a Slipstream out during Garuda, if you are called upon for raise support, that really feels bad too. Physical Ranged DPS in Endwalker might as well be a buffing/support role, as the reason most groups tend to bring them is because they offer DPS buffs to the melee, and since Machinist is the “selfish” DPS of the 3 jobs in that role, it suffers representation for it.

Even outside of the DPS role, the tightness of DPS checks has kept a strong damage meta for tanks and healers too. Astrologian and Scholar are the de facto strongest healer combo for a clear party because both bring raid DPS buffs, and a lot of clear groups in early weeks of the tier have had a tank composition of Dark Knight and Gunbreaker for the high damage output compared to Warrior and Paladin, who were both buffed in the recent 6.21 patch alongside the P8S nerfs to help this issue. White Mage and Sage both have decent personal damage, but bring no buffs for the party to help the main DPS players, so the end result is that your serious early-week prog groups don’t run them as often.

Job Balance Is (Not) A Problem, Maybe

So the prevailing sentiment coming out of the first couple of weeks of Abyssos Savage is that DPS balance in FFXIV feels pretty bad, maybe even a low point. Yet, that isn’t quite true at the same time.

In week 1 clear parties, sure, there were some warning signs like 0 Reaper, Machinist, or Red Mage clears of P8S, and a perilously small number of White Mages, Warriors, and Paladins. It was also possible to end up like a friend of mine, whose week 1 group was “comp-locked” out of the kill of P8S on the first week. That feels terrible and like something that is not typical of the FFXIV raiding experience at the high-end. That is a bad outcome that the team should be working to correct, and the rapid response with the nerf to P8S is probably the right initial response given the overshot tuning target.

However, there’s a mixed discussion to be had about the state of things going forward and job balancing needed. The balance paradigm of FFXIV’s jobs needs a serious analysis from the development team before it becomes a larger problem, and a part of that also means rethinking how boss fights are built. If boss fights remain on their current trajectory, then melee probably needs to be rebalanced as a role around the idea of near-full uptime. If that is the case, then physical ranged DPS also need to be rebalanced, but upwards in order to do more damage. The reason this is an issue is because uptime strats are no longer a difficult puzzle to untangle on new bosses, especially for melee – the hitbox is a mile wide, so what’s the struggle? Melee is missing that sauce in current play, because optimizing for uptime is so much easier to do and even playing safe with some mechanics doesn’t carry a high DPS penalty or require multiple GCDs of recovery movement. There’s still a bit of optimization that melee role players can do, but it’s not a lot and mostly boils down to the same basic idea on every fight – once you know the full timeline and requirements, you can press your rotation into the contours of the fight.

An alternative solution is to get more creative on bosses – smaller hitboxes, bigger movement mechanics, melee forced downtime, and the like. This would allow the current balance to reach a relative stability and not require as much retuning. This is something that I think is a better approach overall, because the balance paradigm of FFXIV stays strong with encounters that have that puzzle aspect to solve. If I’m melee, I should need to figure out how to maintain contact with the boss constantly while doing mechanics. If I’m a caster, I should need to find the right ways to position and cast in order to maximize my performance. And if I want to turn my brain down a bit and just worry about a rotation and meeting basic mechanical checks, physical ranged DPS is there with full uptime and mobility. When the encounter design plays to it, this job balance paradigm really works – there is a reason that FFXIV has been seen as a relatively balanced game for most of its life. But we need to explore one more aspect of this topic…

The Community Is Overreacting In Some Ways, Too

The current balance situation is also not that bad, really.

Like, here’s the thing – it sucks for week 1 raiders, absolutely, and there are groups that had to wait until week 2 to get those first P8S kills in. That isn’t a good scenario to have playing out and that’s not a viable approach to a balanced MMO raiding scene. It sucks that if your prog team had a Red Mage and Machinist in week 1 with a Paladin tank, your chance of clearing P8S was very much reduced. Likewise, it’s perhaps not ideal that even getting a week 1 clear required a gear funnel to a melee – that’s common in week 1 groups anyways, but the week of tomes before Savage meant that you really wanted all your raid gear on a single melee player and that player holding 500 Causality by Savage week in order to immediately get a tomestone weapon after P6S and then upgrade that weapon after P7S.

At the same time, though, the thing is that as of week 2, every job has cleared, even pre-nerf P8S, even prior to the Warrior/Paladin DPS buffs. If you’re meta-slaving now in week 3, you’re behind the times, especially as prior week raid gear, tomestone gear, and general fight readiness has improved. It was a genuine problem that balance was messed up enough that not every job got a clear in week 1 despite so many of the normal players getting to P8S, but at present, it’s just simply not the case that you need a meta group composition to just simply play. In fact, that is part of why FFXIV’s model shines – for week 1 groups at the cutting edge, the balance really matters because a 1% DPS difference can make a huge gap or comfy kill, depending on what direction the difference goes in, especially since you’re only going to have crafted gear, week 1 normal raid drops, and any tomestone gear the raid hustled up to get there, plus maybe some EX trial weapons, but for most players, that difference fades away into nothing as gear pushes you further and further ahead. Groups going in week 1 might have had one DPS player near 620 item level by the time of P8S, but a more laidback group is going to have most DPS near that mark by the time they get there, not to mention that tanks and healers will also have more gear and be more powerful. There’s no need to push to play meta jobs normally – where that failed this tier is that this used to be the ideal behind the whole tier, including week 1 groups, but it wasn’t this time.

Now, does that mean I think that underperforming and underplayed jobs don’t need a look? Nope, quite the opposite in fact. Because I am not, in fact, a game designer (not yet, at least!) I can only speculate, and to avoid putting myself into a Dunning-Kruger hole, I’ll avoid making any specific prescriptive statements, but balance is definitely worth a look. In addition to fight design as I discussed above, individual jobs do need a looking at. FFXIV is entering that bloat-zone that long-lived MMOs so often do, where bolted-on abilities and changed spellbooks have reached a point of potential incoherence, depending on the job. Not everything needs the Summoner or Monk treatment, certainly (a lot of players would hate that, in fact), but there should be a fair effort made to consolidate where possible and to try and envision what makes each job unique and interesting, and then to build the job and balance it around that. I think that for the future state of the game, a lot of work is due to revamp the jobs, even if the actual revamp is relatively narrow in scope (and it certainly should be in some cases). A big part of that, however, boils down quite simply to needing to ensure that no job feels unneeded or unviable in raid content. FFXIV normally does such a good job of this that it makes issues like the current tier stand out sharply.

In conclusion, that’s kind of the whole story of this tier – good raid boss design and mechanics, decent levels of balance, but the outliers on both boss tuning and job tuning overlapped in a really unfortunate way that made both look like huge, looming problems. And in truth, at least on the job balance front, some attention is definitely still needed. FFXIV is still far off from falling into the WoW groove of wide performance gaps – jobs still have a relatively good banding of performance and post-nerf P8S still seems to present a great challenge to many raid groups while being an obtainable goal – but for the first time in a while, that gold medal of balance wasn’t really quite reached, and it feels like there is still work to do at Creative Business Unit III to get the game back to being a gold standard in endgame content balance.

3 thoughts on “FFXIV, Job Balance, and How Communities Discuss Performance

  1. All of this doesn’t make sense to me still. If the average pug doesn’t have a DPS meter, and isn’t supposed to use out of game tools like wowanalyzer to improve their performance, either 1) all pug groups fail, 2) the game is rotationally extremely easy, or maybe 3) some players can just carry

    All this meta talk doesn’t make sense to me because if people are expected not to have a benchmark for their own and their teammates’ performance, they literally will just push whatever buttons, so the game can’t be hard.


    1. In practice, the higher-end the content in FFXIV is, the higher the odds at least someone in the group is running ACT and logging the fight to FFlogs (built by the same developer as WarcraftLogs, in fact). FFXIV also has an analyzer site built by the community that allows people to plug in an FFLogs URL and spit out recommendations to improve their job performance.

      Having said that, I do think it is an extreme stretch to suggest that not having a DPS meter means no one could ever find ways to improve or that things cannot be hard. DPS meters do make it easier to find that kind of stuff and log review after lets you drill down into specifics with greater ease, true, but the way mechanics unfold in FFXIV fights make it pretty easy to tell as a group and individual when someone is making mistakes – mechanical failure is punished with Damage Down or Vulnerability Up debuffs that persist and are visible to the whole party, and there is definitely a point where a group might reach an enrage timer and know that players dying to mechanics or getting slapped with these debuffs hurt even without knowing the specific number tied to it. The only time a damage meter would be more necessary is if the group plays the mechanics perfectly and still hits the enrage – it’s possible for individual players to read their job’s theorycrafting and correct without directly knowing it increased their damage by x% to do so, but it would be easier with a meter and log.

      Rotationally, a lot of FFXIV’s jobs have pretty complicated and rigid openers for their rotations, with a lot of emphasis placed on lining things up around raid buff windows on the even minutes of a fight. There is a range of complexities to them, and most are recoverable if you make a mistake – at the cost of some amount of DPS performance. You could maybe carry a single learning player through a fight, but they’d either need to be doing a baseline bad rotation and mechanically perfect, or doing a great rotation and failing to recoverable mechanical checks (and not everything is recoverable in Savage and up content). It’s not that different from most Heroic guilds in WoW that I’ve been in.

      But all of this assumes that most parties have 0 people logging data, which is definitely not the case in Savage or up, or assumes that not having a damage meter means that things cannot be challenging, which is a bizarre claim to make absent any evidence, to be honest. Logging does make it objectively easier to spot specific issues with rotational gameplay, but those are far from the only issues that level of raid gameplay has even in WoW.


      1. So, take for example my journey in trying to learn how to dps in WoW as a tank main. I gave frost mage my best shot, exhaustively reviewed what little there is to the rotation, consistently performed badly, logged and checked buff uptime, even posted a log on the mage discord. The reason I knew I was a burden to the team was because I could clearly see my DPS was dogshit. I was determined not to push higher than 10s until I could improve, and eventually gave up on frost because it was disheartening to take perhaps the easiest DPS class rotationally and just totally suck at it. That’s even outside of any in-game criticism (I was never in a group where it disbanded, or completed, and someone said “mage sucked”, as an example). And my DPS wasn’t poor because of mechanics per se, because I knew what I was supposed to do and didn’t die, it was piss-poor execution of 1) positioning and 2) ABC, probably because I was unaccustomed to having cast times and constantly was repositioning rather than pre-positioning and planting.

        Simultaneously, I was playing ret paladin and struggling with totally different things. Situational awareness was challenging because I’m staring at my weakauras. Lining up the CDs (this was s2) and not holding them, or holding them, at the wrong times. Dying a lot. In this case I could see how failure tagging would help, but it’s blindly obvious when I die on 1/3 of the pulls and suck on the pulls I live through.

        But when the talent selection switched for ret due to 1 pc, that’s when I started to feel like it was worth it. Tank threat buffs also helped ret a lot. Being able to see my DPS hit 80 or 90k mid-pull, miles higher than anyone else’s, and keeping up with dps even outside CD windows – that’s what made me think, yeah, finally I actually CAN do this.

        All that aside, experiencing runs from a tank perspective, watching my team’s CDs to see when to pull big, and especially failed runs where bosses took too long to die and the healer couldn’t keep up with attrition, and being able to recognize – very importantly – what was my fault for over pulling vs very low DPS making dungeons impossible to complete or time. All of that is hugely facilitated by having a DPS meter.

        This outside of judging my guildmates’ capabilities for raid or m+ and helping to review logs to assist my teammates in improving their gameplay or examining improvement possibilities in my own, which is what your comment focused on.

        It’s true I don’t play FFXIV, but I think it’s equally bizarre to expect and set as a standard NOT using third party tools because of the occasional toxicity that comes along with it.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.