Adps vs Rdps. Which one to go by?

Ive become more active on the forum boards and one thing that I see a lot is people talking about using Rdps vs Adps when it comes to job balancing. I personally believe, that while Adps is important, balancing should weigh more heavily into Rdps. Especially when you look at the DPS jobs. Take Blm and Mch, for example, in the Asphodelos raids. Outside of Phoinix, Blm is number 1 or 2 in both Rdps and Adps, while mch is dead last in rdps and and 7th or 8th in Adps in all the Asphodelos raids. (Based off fflogs of all percentile over the last 2 weeks) I say, when looking at job and role balancing Adps is the first initial step, its your constant, but you have to take it a step farther. In the game, job balancing only really effects game play in group settings such as dungeons, raids and trials. So why wouldnt we base balancing more on the functionality of each job in a group/party setting (rdps)?

Both.

To try and simplify an argument that's been ongoing in several threads - RDPS shows DPS minus single target and aoe buffs, + the damage others gain from the buffs you give, and ADPS shows DPS minus single target padding, but including AoE buffs you receive from the party. https://www.fflogs.com/help/rdps

What this essentially means is, a job with sufficiently high ADPS is bursting within 60/120s windows which increases overall group DPS in a way that can't be measured purely by the RDPS number, just like how the ADPS number can't account for damage the job gives with their buffs.

There is no simple and clean and easy "which one" answer.

The answer is both.

Samurai is the 5th highest RDPS at max at the moment, and yet if you go look at the top speed kills for 3/5 of the fights, it is used in the comps. This is because of their high ADPS, very strong personal dps and burst that's dumped into buff windows, which is complimented by a comp that gives buffs for them to dump their burst into.

https://www.fflogs.com/zone/statistics/44/#dataset=100
https://www.fflogs.com/zone/rankings/44/

You cannot look exclusively at either one or you are not grasping the entire picture.

Aside from that, Machinist and ranged physicals/casters not named BLM are on the weaker side in general, but that's a more complex issue, imo, and is something I've never seen people come to a consensus on - how close should Machinist be to BLM and Samurai ADPS? Who should be the lowest RDPS/ADPS? Etc? Look in a hundred threads and see a thousand answers, with dozens of factors taken into account, from difficulty(which people argue for and against,) APM, melee vs. ranged, cast times vs. instant cast, res vs. no res...yadda yadda.

Quote:

---

Originally Posted by Blutiful

You cannot look exclusively at either one or you are not grasping the entire picture.

---

This. SO MUCH this.

When you're looking at FFLogs in order to make determinations you have to understand that you're not asking a question that has an answer that can be given by one piece of data. You have to look at all the data, and see what the trends are, and decide whether the metric you're looking at is valid to the question you're asking.

So, let's say you're a purple parser (rarely orange) and you'll looking to decide if the job you're taking is holding your group back. Is looking at top speed runs going to help you? No. It's irrelevant to your needs. Is looking at the percentiles people using that job get going to give you information? No, that doesn't tell you anything comparing different groups. (Percentiles of performance are within a job only, not compared between jobs, obviously). Would looking at 99th percentile be helpful? Maybe, if you're actually being realistic when you think you can achieve that performance. But you know what? You probably won't, or don't have to. That's okay. That's not your goal here.

But if you start looking at the rDPS/aDPS spreads (that's all percentiles) favoring the top half of the whisker/box plot? Well now you can see 'Okay, this is where I am, and that's where I'd have to be in those jobs to get improvement. Now you're comparing where you are to where you might be on other jobs--cause a 80percenter on a monk miiiiiight not be an 80percenter on a reaper, after all!

And sometimes you just have to accept the question you're asking doesn't have an answer in fflogs. How do you rank job balance? How does SE rank job balance? Are 1% differences even significant in FFlogs? (tentative answer: probably not)

But the truth is--choosing one data point and saying 'Ah yes, I looked at the top parse and it verifies my thesis' doesn't lead to valid data. You might be right, but you might be wrong. Or you might be asking a question that doesn't count.

FFLogs does a great job, but data interpretation is hard, and the temptation to decide 'this, this one number, right here, that's all the info I need.' It rarely ever is. There's too many confounding factors, there's too much variance, hell, there's too much error margin inherent in FFLogs to come to many of the conclusions people do.

I made this post because in a lot of the forums Ive been on, its a constant argument on whether you should go by adps or rdps. My point wasnt to say that you should only use 1 over the other or that rdps/adps should be the only factors and I'll admit that I probably misinterpreted them. What I was meaning, is that Adps seems to be a better measurement of personal dps while Rdps seems to be a better measurement of your dps in a group or party setting. In my opinion, Adps would a good metric to use as a baseline and then look at rdps and other factors to see how a job is doing when in a group. If there are massive changes in job rankings like in fflogs or dps numbers, it could indicate that a job is over or under performing in a group setting. I will also admit that to me rdps would seem like a more valid measurement for balancing jobs because when talking about balance its usually in a group or party setting.

I could be completely off the mark. But since we dont know exactly how SE balances jobs, places like fflogs are going to be the best resources we have to make comparisons and base our opinions off of. It shouldn't be used as a finite definite answer, because like its been pointed out there are a multitude of variables to consider. However, its been shown in the past that if there is a problem with a job's performance it will usually show on places like fflogs. I made an inference that because Mch seems to be the only job that has major jumps between where they rank on fflogs and given community feedback Ive seen that there is an issue with the job.

I guess my point was that just because a job is doing well by Adps, doesnt mean its performing well in a party and to look at rdps to get a better understanding. Like you've said you cant go by just one or the other for a definitive answer. But by looking at these we can base personal conclusions on a jobs effectiveness and then be able to have a discussion on them.

The real answer, put simply; Just don't make a Bard, its low on all charts.

Focus more on RDPS as its signifies buffs from everyone, in a raid setting. This can be applied to dungeons, as if that even matters to be honest. And work on your rotations.

Both is the answer.

After 6.21 buffs, PLD had a high rDPS because it would simply hit hard.
But aDPS wise it was the lowest.

Because while PLD was hitting quite hard, it wasn't putting the most of its potency within the raidbuffs, meaning it would generate less value from buffs.

Both. In theory, a job that does great rDPS does less aDPS to compensate. If one does both low rDPS and aDPS, something needs looking at. If one's high in both, it might be busted.

Funny that people can openly discuss these topics and fflogs on the official forums even though parsers are supposedly against the TOS.

Quote:

---

Originally Posted by Gruntler

Would looking at 99th percentile be helpful? Maybe, if you're actually being realistic when you think you can achieve that performance. But you know what? You probably won't, or don't have to. That's okay. That's not your goal here.

---

Not too important but I would argue that looking at 99th percentile is never helpful nowadays, even if you have the skill to compete with the best players. An orange parse (95th percentile) is achievable in a normal clear of whichever boss, but those last 4 percentile are often entirely up to RNG, requiring dozens of optimized kills to get that "perfect" run.

Doesn't matter how well you play, if your big potency attacks don't crit you're pretty much guaranteed to not get that 99.

99th percentile is unhelpful for a number of reasons. A lot of it (especially this expansion) is crit RNG, due to how damage is so compacted into 2 minute bursts with high potency hits. There's also the problem with feeding- usually when someone's that high, it's because they want to be. So the run was catered to them. Uptime may have been bent around that person. Strategies may have been tweaked to make sure that one player had as much padding as possible. If it's a buff-dispenser job, you can ensure the rest of the party is high personal damage so their percentage buffs have magnified output, increasing that player's rDPS.

Now, looking at the specific actions that person did during their run is definitely helpful if you're looking for optimization tips. But if you go in expecting to get above purple or low orange on that alone, you're not gonna have a good time.

Saying "another job has higher aDPS means it will burst higher!" is a total fallacy that fails to account way too many variables when it comes to burst, one of the bigger ones being that a higher aDPS job tends not to provide any raid buff in counterpart, and is generally tuned on its raw base damage potency higher than the others (namely SAM, BLM to name the obvious suspects). It is true that one has higher aDPS than the other since it has higher burst potential, but saying that it does bigger burst because it has higher aDPS is a total non sequitur. Only one implies the other, else you'd get DRK having higher aDPS than anybody and yet, that's not true, but it's still the biggest burst job in the game with NIN. They do not, however, have higher aDPS for the simple reason that their base potential power is lower than SAM for instance, because they do provide raid buffs (and DRK is a tank anyway).

aDPS is a metric that only accounts for how well a team plays into their own raid buffs, and is also unique to a single team composition (a different composition will play with different raid buffs, and the metric will change a little at equal performance level). On a minor note there, rDPS also has a flaw, which is that it also tends to skew in favor of raid buffers (like DNC) that are playing in selfish heavy compositions, which naturally inflates their rDPS charts. This is due to the fact that both metrics are a complement to each other in that they try to funnel a rating on an individual when in reality the performance is at a team level. To make it simple, if you account for aDPS, you do not account for how well your team mates played into your buffs. If you account for rDPS, you do not account for how well you play under your team buffs (because that value is redistributed to them).

In short, even though as I said above you can definitely find small flaws:

- rDPS is great to compare what each job provide to a team on a GENERIC fashion.
- aDPS (the successor to raw personal dps, aka the same with the solo buffs like cards removed to prevent padding) is an old and flawed metric that has way too many variables not accounted for. It CANNOT be used to compare burst potentials, even though it is definitely affected by them. It's just good to know how you played into your team raid buffs for a specific run only.
- nDPS is the metric you're looking for if you want to know the base raw, unmodified, unbuffed power a job has on its own. It is the best tool at your disposal if you want to know how well you're faring compared to other people playing the SAME job on a rotational basis only (still affected by the native rng nature of each hit but that's another story).

If you want to account for a job pure burst potential, you'd need to take a nDPS chart and extract the 20ish seconds that cover each burst every 2 minutes.

Burst is important enough to take into account because it is often reflected in charts and if you check them on a regular basis (every couple of weeks) you'll see some jobs moving up or down slightly on rDPS for example, because people are adjusting to the fights themselves, and the way jobs can play around downtime or especially kill times (that will allow or not another burst before the kill) means that nothing is completely fixed into stone. As people gear up and progress, some jobs suddenly crop up a little because they fare better with certain kill times, a little more than others.


Edit: I do believe fflogs is focused on the wrong part of the play, which is personal performance, which also comes from wowlogs I believe (?), and more generally how as a society we value things. Or maybe that's because fflogs favors this that everybody follows, hard to say what came before between the egg and the chicken. Either way, both feed into each other and we end up with a maximized individuality, with individually focused metrics instead of having a bigger focus on party total dps and kill times, which always seemed the more natural way to go about things, to me at least. Don't get me wrong, there is still value in individual metrics, but it's pretty telling that percentiles on individual pages are centered exclusively on individual metrics rather than party wide.

Quote:

---

Originally Posted by Semirhage

Both. In theory, a job that does great rDPS does less aDPS to compensate. If one does both low rDPS and aDPS, something needs looking at. If one's high in both, it might be busted.

---

rDPS is not just the amount of raid buffs a job provides. It also includes their own personal raw contributions (nDPS).

If this was the case BRD or DNC would be at the top of the rDPS charts, yet they are not.

Quote:

---

Originally Posted by Valence

rDPS is not just the amount of raid buffs a job provides. It also includes their own personal raw contributions (nDPS).

If this was the case BRD or DNC would be at the top of the rDPS charts, yet they are not.

---

Yes I'm aware that rDPS isn't buffs alone. I meant that if a job both dishes out and makes strong use of incoming buffs, that may be a balancing red flag. If it doesn't play well in the dishing out or receiving game, that may also be a red flag.

(We also split them into relative categories by subrole to sidestep the arbitrary meleewalker > BLM >> other casters > ranged hierarchy.)

wait, thought Rdps is your personal DPS + the DPS your buffs bring to the party (and Adps is just your pure dps, without any buffs)

Quote:

---

Originally Posted by Shiroe

wait, thought Rdps is your personal DPS + the DPS your buffs bring to the party (and Adps is just your pure dps, without any buffs)

---

rDPS is your personal DPS + the DPS you provide to the party - the DPS you gain from others

aDPS shares the same base: your personal DPS + the DPS you gain from others (except solo buffs like cards to prevent padding, with the exception of dancer partner buffs that are accounted for). aDPS is essentially identical from what you'd get from a piece of software that makes yoshi sad, minus the solo buffs like cards (thus why it was called adjusted dps... I know, the names are really bad).

Your DPS shaved off any buffs (doesn't account what you receive, doesn't account for what you provide), is nDPS.