Ranged DPS, choose a weakness that would justify a buff

[Cetonis]

Casters in general SHOULD be higher damage than physical ranged

Yes, I said that. Don't straw man. Never did I say ranged should do just as much as other roles, in fact I literally said they shouldn't. It was the very first sentence, even.

But, that doesn't mean that casters have to deal as much as 10% more. If every non-ranged dps deals 10% more than the best ranged, that's when you're in the zone of taking a ranged at all becoming bad. 4-6% or so is closer to where you want to be. Basically, we're merely lucky right now that there are only those two or three outlier jobs.

As for SMN vs. BLM, I'd agree that SMN should do a bit less than BLM. But that's unrelated to the gap between ranged and non-ranged being larger than is probably ideal.

[Kabooa]

I'm still thinking we should have a fourth role to shove our selfish DPS into.

With nothing else accounted for, that lets you have everyone on the same starting line with a fatter full house attribute bonus.

Then you could balance the machinist, black mage, and samurai against each other, which is less of a hassle than doing it with everyone else, and then balance the others against the rest in their category.

A more grand reorganizing might also do some good, but for a starting baseline, it solves a few things.

[Shurrikhan]

I'm still thinking we should have a fourth role to shove our selfish DPS into.

Why remove opportunities for unique takes on a role just to shunt anything that isn't overly homogeneous into a new set of likewise homogeneous jobs?

I don't see what's wrong with one Ranged being far more dependent on indirect contribution than the rest. Unlike Dragoon, it isn't a unique gear class and just as much as there might be slight annoyances in being dependent on your party to fully preform there is equal advantage in being less dependent on your own direct performance, so as long as the gear class (Aiming) is never going to be shafted as a whole, there's nothing to fear in having a particular unique Ranged among the others.

In the end, we aren't attempting to balance aDPS, only rDPS, and only against the increased opportunities the toolkit provides (in terms of mistakes forgiven, more granularly reduced skill-ceiling via more flexibly timed skills, or up-time tricks permitted, etc.) for the player and their party. I'm not sure exactly what good a grand reorganizing would do, though admittedly I'm coming into the line of thought with far less forethought than you likely have.

[Kabooa]

Why remove opportunities for unique takes on a role just to shunt anything that isn't overly homogeneous into a new set of likewise homogeneous jobs?

However little we make the claim, SE has shown that they respond to player feedback. They may not respond in the way we want them to, but in terms of end-result their aim isn't bad. So long as our most common feedback boils down to Nambers, their going to respond to us in terms of Nambers, and the nambers, while questionable, are rarely the result of unintended math. (As in, you can compare SSS expectations to what jobs actually put out and they usually line up)

The current tuning is a little off, but not awful. You still want one of each, you just don't really consider a second ranged in a min max week 1 / speedclear setting. There's nothing inherently wrong with it, but the perception created is you only take ranged for the bonus, but it isn't Ranged who lost their spot this tier, that's largely been melee despite being the 'stronger' role.

A fourth role gives us another tuning lever to work with, and more tuning levers means we can make smaller changes and get the same effect with a little more precision.

Using our current paradigm to show some examples of things we simply can't do.

[Dancer and Bard cannot threaten Machinist in damage contribution, else Machinist is obsolete.]
--Repeat for the other two roles with Samurai and Black Mage.

"Obsolete" might seem like a gross exaggeration, but we've already seen this happen to Black mage and samurai. Perhaps machinist, but Stormblood I honestly thought it was fine, but now they're in the same boat. This is despite being capable compared to the content. They simply brought nothing of value to a team at their various points, and little advantages mean quite a bit.

[The value of non-damage utility is not constant.]
How much is mobility worth? How much -should- it be worth? What about a raise? Job specific boons? This problem will still persist but it is grossly apparent when jobs exist in a role who have nothing unique to offer.

Cross role examination creates some glaring examples where some jobs have the boon tax and others don't. There's no reason Summoner, Bard, and Dancer shouldn't be equivalent in damage output. Their supplementary kits are all comparable. There's no reason Red Mage and Summoner shouldn't be similar for the same reason.

Machinist and Red mage should have comparable damage - the free mobility vs the value of verraise.

[Reduces our Arms Race.]
Circular potency buffing is likely to still happen but their scale should be smaller. The machinist getting buffed doesn't mean the bard and dancer must be buffed which would mean the red mage must be buffed, which would mean the melee must be buffed, which would mean- etc etc.

If Machinist is out of tune, then it needs only be adjusted based on Samurai and Black Mage.

If you want more numeric ramblings, here's some.

I see a fairly diverse amount of feedback from various players when it comes to Shadowbringers job direction, so my suggestion included a few other considerations Namely it tries to hit a few notes.

1. The RDPS>PDPS supporters
2. Not alienating the PDPS=RDPS jobs.
3. Re-emphasizing a burst window (Though not necessarily a 120s mystic alignment)
4. Minimizing the impact / tuning needed in the world at large (Since this would literally be for the benefit of Savage and Ultimate only)
5. Encourage party diversity (AKA not hard lock a comp)

In my opinion, the blurb from the previous page aims to accomplish all that.

This is going to be a lengthy one, and we'll be using some arbitrary number values to illustrate the concept.

Quick Primer: RDPS vs PDPS

RDPS or "Raid DPS": This is the baseline damage that the job brings to a group. This includes personal DPS and all DPS given by buffs and minus all DPS taken by buffs.
PDPS or "Personal DPS": This is the end amount of damage sourced from your character, including all buffs.

Defining "Out of the Box": I've been using this term, but basically this is raw RDPS the job does assuming a raid size of 1. Therefore, an 'out of the box' RDPS value of 12,000 with Ninja whose trick attack brings on average 2% bonus RDPS has a PDPS value of about 11,760. If Trick attack was an average RDPS increase of 10% (which is like, the same as having 5 dancers), then its PDPS is around 10,800.

To illustrate just how ridiculous that would be in practice, any job that increases the raid's total DPS by 10% would, at the suggested numbers to follow, have to have a PDPS of around 3,000.

A ton of gobblygook.

Let's use the first example first, with a balanced team of two tanks, two healers, and one of each type of dps (Melee, ranged, caster, cannon). We'll use tank = 7500, healer = 5500, and DPS = 12,000 'out of the box'. The total raw RDPS = 74,000.

We'll say each of the Melee, ranged, and casters are the same weight - on average, the support is worth about 2% more RDPS and it always lines up cuz magic. 74,000 * 1.02^3 = approximately 78,529. With a Cannon, you double that RDPS increase for them, netting an additional 749 DPS. 79,278.

Let's say instead of the cannon we add another melee. The same calculation is 74,000 * 1.02^4 = 80,099.

Not bad, it's pretty close.

A small differential, but with the additional role we have yet another tuning lever in the attribute bonus. Team 1 has all roles represented and Team 2 has not.

Lets apply a 1.08 (representing 8% attribute bonus) to team 1 and 1.05 to team 2.

Team 1: 85,620
Team 2: 84,103

What about a double cannon team?

Team 3: (74,000 * 1.02^2 + (24000 * 1.02^2 - 24,000) ) * 1.04 = 81,077

All of this assumes 100% performance by everyone involved. Supposedly SE tunes in unupgraded Tome Gear and slashes that maximum performance, but I can't say for certain. Titan is currently around 75,000 or so, so lets apply a "Casul tax" and slash our totals (We'll assume the presented numbers in this system are in i450 unmelded) by about 15%.

Team 1: 72,777
Team 2: 71,487
Team 3: 68,915

No week 1s for anyone. A fully pentamelded set (we'll use DH because it's easy) over a 0 meld set is an extra 2180 stat points. With DH, that's 36.33% direct hit chance or about 9.075% more damage. Pentamelded team looking like

Team1: 79.872
Team2: 78,456
Team3: 75,634

Week 1s for many people, though multiple cannon jobs are taking a nasty hit.

What if we applied the increased cannon buff to attribute bonus as well from the party system?

In Team 1 we see an increase from 14,578 to 15,658. Total gain 1,080.

In Team 3, we see an increase from 13,498 each to 14017. Total gain 519 each. Still rough for them.

What if the "Cannon buff" itself was a flat 3% rather than being a full house bonus?

In team 3, we see an increase from 13.498 each to 13,757. Even worse.

...Both?

13,498 to 14,536 each. Total increase of 2077.

Team 3 100% total: 83154
Team 3 casul tax total: 70,680
Team 3 pentameld: 77,572.

Summary

System Change
New Role: Cannon. Jobs in this role gain double the effect of party buffs and enemy vulnerabilities, and role attribute bonus.

"Normalized" RDPS and PDPS levels.

Team composition
Team 1 (2 tanks, 2 healers, 1 of each dps role)
Team 2(2 tanks, 2 healers, no cannons, 3 other roles covered)
Team 3(2 tanks, 2 healers, 2 cannons, two other roles covered)

Theoretical Maximum
Team 1: 86,700
Team 2: 84,103
Team 3: 83,154

'Minimum ilvl tune'
Team 1: 73,695
Team 2: 71,487
Team 3: 70,680

'Pentamelded'
Team 1: 80,880
Team 2: 78,456
Team 3: 77,572.

It's late, but you get the idea. These three team set ups cover a fair range of party compositions (though obviously not all), but as a parting note, it seems if we do the doubled attribute buff and cannons adding a flat +attribute bonus rather than being a full set bonus, the stat bonus could be reduced.

[Shurrikhan]

[The value of non-damage utility is not constant.]
How much is mobility worth? How much -should- it be worth? What about a raise? Job specific boons? This problem will still persist but it is grossly apparent when jobs exist in a role who have nothing unique to offer.

Cross role examination creates some glaring examples where some jobs have the boon tax and others don't. There's no reason Summoner, Bard, and Dancer shouldn't be equivalent in damage output. Their supplementary kits are all comparable. There's no reason Red Mage and Summoner shouldn't be similar for the same reason.

Machinist and Red mage should have comparable damage - the free mobility vs the value of verraise.

This is honestly the only issue I feel is inevitable, no matter what changes elsewhere, between jobs which include and do not include indirect contribution (especially of a damage vs. non-damage variety). The rest mostly balances itself out with only the sad requirement on a highly indirect contributor must have an entire party of skilled players to contribute its full rDPS while a wholly direct contributor needs only itself to provide its full rDPS (and can offer indirect contributors more in turn).

In all other cases, I'd honestly be happier with removing Role-based stat bonuses altogether and having a tighter balance for individual jobs relative to all other jobs than adding more sub-types of Roles.

That said, I now feel like I need to remember what a friend told me about another MMO he was playing. If I recall correctly, mobs and bosses there had defense and most classes all had about the same damage calculations based on their stats; the difference was that the direct damage jobs had higher raw stats to work with. And because Defense worked against raw stat to a degree (I don't know how), they'd be less affected by Defense, making up for the disadvantage they'd otherwise face in larger parties for having no party-based damage contribution without overpowering them in your typical smaller-party content (where party-based contribution wasn't as hefty anyways).

As weird as all that sounds, I'd honestly rather go that route, looking at how damage itself works in order to better allow for balance between the varying levels of (in)directness in contribution as to be balanced across various party sizes, rather than just thrusting A, B, C, and D upon any and all parties (all too likely in place of actual balance).

[Cetonis]

Looking at current fflogs statistics is fairly dubious, nobody is fully geared and any groups that are serious about pushing for speed early would have fed their melee/casters and starved their dancers. And few groups are doing those hard speed pushes, not that there ever are very many to begin with, but you don't even have the much more numerous soft "let's optimize with what we've got" groups really getting that going just yet, not until there's a firmer idea of kill times at max gear to line up buffs and such.