SusEX and LakEX Average DPS Charts, Top 100 Players

Hey all, surprised nobody's made this yet with all of the discussion about job changes in SB. I've only just started HW MSQ, so I feel unqualified to make any assertive statements, but I thought a visualization of data might be useful for a lot of you.

All the data was taken from FFlogs, using the average DPS of the top 100 players of each boss.

The only thing I can say for certain is that I'm a bit miffed I used the boost on MCH. :p

http://i.imgur.com/gf9D13l.jpg

We did make one. It got removed. Tragic because data like this is crucial for us to discuss the balance and overall health of the game...

This thread isn't going to live for very long...

Agree that balance discussion without hard data is not particularly useful. Looking just at top logs, however, doesn't give you the full picture either. You're effectively selecting for good rng, short kill times, high balance and buff uptime and potentially all sorts of cheese. Just ask yourself, how many of those top 100 SAM logs will have had generous balance uptime vs the top 100 MCH logs?

Quote:

---

Originally Posted by Galvuu

We did make one. It got removed. Tragic because data like this is crucial for us to discuss the balance and overall health of the game...

---

Ah, well, if it happens it happens.

Quote:

---

Originally Posted by Fortune_Cookie

This thread isn't going to live for very long...

Just ask yourself, how many of those top 100 SAM logs will have had generous balance uptime vs the top 100 MCH logs?

---

I went with the top 100 figuring that they'd be at the top of their jobs and roles, thus giving a more accurate picture of how each job would perform when played at the highest levels. I'm still pretty new, so open to suggestions on data! Do you recommend expanding out to the top 500, or will these numbers just be inherently inaccurate?

Quote:

---

Originally Posted by D3VNT

Ah, well, if it happens it happens.



I went with the top 100 figuring that they'd be at the top of their jobs and roles, thus giving a more accurate picture of how each job would perform when played at the highest levels. I'm still pretty new, so open to suggestions on data! Do you recommend expanding out to the top 500, or will these numbers just be inherently inaccurate?

---

Even so, most of these top ones are dps padded/fluffed.

Nothing inherently wrong with looking at top x logs, so long as you're aware of the kind of bias this tends to introduce.

Job representation is an important aspect to bear in mind. There are 2.5 times as many SAM logs than MNK logs for Susano Ex, meaning there are more opportunities for SAMs to get lucky runs where crit & fight rng loves them. If you look at top logs you're going to pick up the effects of this. Competing jobs also create sampling bias - if a lot of good players switch from MNK to SAM (for example!!) the data may overestimate the actual gap between the 2 jobs. On top of this FFXIV has a few idiosyncratic effects: balance card, different comps with different synergies. If a particular "meta" develops, jobs not included will tend to suffer in the log rankings - they're almost certainly not as bad as the numbers will suggest at first glance.

You could look at the numbers at different percentiles (e.g. 50%, 75%, 90%) to see if there are obvious patterns.

Quote:

---

Originally Posted by Fortune_Cookie

You could look at the numbers at different percentiles (e.g. 50%, 75%, 90%) to see if there are obvious patterns.

---

This is generally more useful because we get a more "global" view of the jobs also (we should balance for the top players, but we can't disregard the more casual audience either, even if balance isn't inherently as important for them).
I can't wait until they filter out balance in the logs... that way I don't have to manually click 50 logs and check for single target balance usage -.-

Thank you! Very informative. Sometime this weekend I might make prettier visualizations going all the way through the top 50% in 10% increments.

Quote:

---

Originally Posted by Fortune_Cookie

Nothing inherently wrong with looking at top x logs, so long as you're aware of the kind of bias this tends to introduce.
There are 2.5 times as many SAM logs than MNK logs for Susano Ex, meaning there are more opportunities for SAMs to get lucky runs where crit & fight rng loves them.

---

And also more opportunities to get unlucky runs. This is nonsense. The larger the sample size the the less an outlier will effect the average, not the other way around. In other words, SAM currently has the most solid data.

*edit* nevermind I thought this was general dps data, not just the top 100. You are correct.

Was this based on the whole actual fight or the stone sky sea dummy?