wow Dragonflight Mythic+ Season 1 DPS, Tank and Healer Log Rangkings: Week 1.5

[Starym 3,800]

 

We're taking a mid-week look at the Mythic+ logs, as some major tuning arrived this week and we wanted to give it a chance to take effect.

Note: The logs are based on POINTS, and not actual damage or healing, meaning they log the timed completion for the specs, with higher keys getting more points, obviously. The time in which the dungeon is completed is also a factor, but a much, much smaller one, as it grants very few points if you do it significantly faster than just any in-time completion. We're also using the Normalized Aggregate Scores numbers, for clarity, meaning the top spec is marked as 100 and then the rest are ranked in relation to that peak point.

All Keys

95th percentile

We'll just take a quick look at the rankings this time around, as we'll be back next time with a full week's worth of tuning data. Outlaw pushes Arcane down, but the rest remains very much the same as in our first look last week, with Feral sneaking in past Shadow to the final top 10 spot. The bottom is also the same, with Warlocks having a pity party, as Fire gets joined by Beast Mastery near the end.

Mythic+ All Keys 95th Percentile Data by Warcraft Logs.

All Percentiles

Subtletly is the mover here, as it pushes Marksmanship down, with the Hunter spec also getting bullied out of the top 3 by Arcane. Windwalker and Demonology avoid all the chaos around them and stay put in 5th and 6th, as Havoc pushes through Balance into 7th. Enhancement also moves past Feral to claim the last top 10 spot. Meanwhile the bottom is looking just like in the high percentiles, except it a different Hunter there with Fire and the Warlocks.

Mythic+ All Keys All Percentile Data by Warcraft Logs.

Tanks

Protection Warrior made its way to the top, as both it and Blood DK pushed Vengeance down out of its pre-patch dominance, as Paladins and Druids commiserate at the bottom.

Tank All Percentile Data by Warcraft Logs.

Healers

Preservation makes a name for itself here, as it almost manages to topple Balance out of first place, and the Priests bemoan their pre-patch performance.

Healer All Percentile Data by Warcraft Logs.

 

For even more in-depth data for each individual key head on over to Warcraft Logs. And if you're interested in more info on the specs themselves you can always check out our class guides, as well as our Mythic+ guides.

[MiNiGiANT 26]

Destro lock is last after the buff... omg...

i was thinking to roll back to prot pala but that is also last...

[TheRagingwolf 258]

These stats are just not reliable. Click into the Warcraft Logs site and look at how insanely lopsided all of the numbers are. In some charts, there might be 30 runs for a DK and only 5 for a Druid. Then consider what you are seeing above includes ALL item levels so, if those 30 DK runs have an average item level higher than the 5 Druid runs, well, obviously the DK runs have an advantage.

This is the only consistent thing you see in the real numbers over there, how uneven the information is. Unless the classes and races are evenly represented then all these charts do is show you which is getting the most play, not which is truly better.

Edited December 23, 2022 by Madyrn

[Synapss 0]

45 minutes ago, MiNiGiANT said:

Destro lock is last after the buff... omg...

i was thinking to roll back to prot pala but that is also last...

Play what you like dude^^ I'm destro and in any dungeon or raid I did I never felt like I'm the worst x) I'm closer to the best than the worse. Probably coz I have decent gear ;  but I still don't have 2/4 bonus sets or 2 best trinkets. Indeed If you play a "low" spec, you better not skip some stages and take your time to get gear coz your spec doesn't compensate your lack of stuff ; but that's it. Except If you're Mythic raider or expect to do MM+20 keys ; playing best specs isn't necessary and you should play what you like in my opinion. And that's even more the case with tanks and healers, the difference between them is almost non-existent here^^ 

[ArthasStormrage87 52]

1 hour ago, Madyrn said:

These stats are just not reliable. Click into the Warcraft Logs site and look at how insanely lopsided all of the numbers are. In some charts, there might be 30 runs for a DK and only 5 for a Druid. Then consider what you are seeing above includes ALL item levels so, if those 30 DK runs have an average item level higher than the 5 Druid runs, well, obviously the DK runs have an advantage.

This is the only consistent thing you see in the real numbers over there, how uneven the information is. Unless the classes and races are evenly represented then all these charts do is show you which is getting the most play, not which is truly better.

Facts!!!

[Timmhey 0]

"Healers

 

Preservation makes a name for itself here, as it almost manages to topple Balance out of first place, and the Priests bemoan their pre-patch performance."

 

Did i miss something here? Resto druids are not a thing anymore?

[MiNiGiANT 26]

On 12/23/2022 at 9:44 PM, Synapss said:

Play what you like dude^^ I'm destro and in any dungeon or raid I did I never felt like I'm the worst x) I'm closer to the best than the worse. Probably coz I have decent gear ;  but I still don't have 2/4 bonus sets or 2 best trinkets. Indeed If you play a "low" spec, you better not skip some stages and take your time to get gear coz your spec doesn't compensate your lack of stuff ; but that's it. Except If you're Mythic raider or expect to do MM+20 keys ; playing best specs isn't necessary and you should play what you like in my opinion. And that's even more the case with tanks and healers, the difference between them is almost non-existent here^^ 

Long story short, the game is 10 times better if your class is S tier.

[TheRagingwolf 258]

47 minutes ago, MiNiGiANT said:

Long story short, the game is 10 times better if your class is S tier.

When those tiers are based on bad information, you should play what you like.

[MiNiGiANT 26]

On 12/26/2022 at 6:58 AM, Madyrn said:

When those tiers are based on bad information, you should play what you like.

I think the information for these rankings are very accurate.

The other problem is ppl whatching it and make groups based on this, so it doesnt matter if the information is good or bad, if your class is the last here you not gona be invited to good groups.

Edited December 27, 2022 by MiNiGiANT

[TheRagingwolf 258]

7 hours ago, MiNiGiANT said:

I think the information for these rankings are very accurate.

The other problem is ppl whatching it and make groups based on this, so it doesnt matter if the information is good or bad, if your class is the last here you not gona be invited to good groups.

Except it is not accurate, or even close to being accurate. Anyone that understands distribution, averages, and sampling knows that these rankings are antithesis of accuracy.

What they are, however, are a prime representation of popularity. The Warcraft Logs website allows you to view the information regarding the number of runs per class, item level, etc. Not a single chart gives equal representation when it comes to these factors. That alone makes the "Show All" chart which we are presented with in these articles woefully inaccurate in terms of class ranking.

Relying on these charts for even a casual idea of which class stands out in terms of performance is a massive mistake, but especially for group comps.

As an example, look at the table at the bottom of this page https://www.warcraftlogs.com/zone/statistics/31#boss=2587&bracket=1 . Direct your eyes to the far right column labeled "Parses". Even on a set boss, Eranog, with a set difficulty of Mythic and a tighter item level range of 376-378, the parses are all over the place. Now, this is a much better representation of performance than the "Show All" which we are continuously presented on here, but it is still completely inaccurate since the parses are not equal.

For any of these charts to be accurate, you have to have equal parses, on the same fight, with those parses in the tightest item level range possible. Otherwise, it is merely a representation of which class is getting played the most. That, of course, leads to many playing those same higher played classes, which leads to a more heavily skewed and inaccurate log.

It's simple statistics and while Warcraft Logs is a solid utility for the game, in this instance, it is being used wrong.

So, yes, it does matter that this information is wrong, ESPECIALLY because it will be used to build groups.

Edited December 27, 2022 by Madyrn

[MiNiGiANT 26]

13 hours ago, Madyrn said:

Except it is not accurate, or even close to being accurate. Anyone that understands distribution, averages, and sampling knows that these rankings are antithesis of accuracy.

What they are, however, are a prime representation of popularity. The Warcraft Logs website allows you to view the information regarding the number of runs per class, item level, etc. Not a single chart gives equal representation when it comes to these factors. That alone makes the "Show All" chart which we are presented with in these articles woefully inaccurate in terms of class ranking.

Relying on these charts for even a casual idea of which class stands out in terms of performance is a massive mistake, but especially for group comps.

As an example, look at the table at the bottom of this page https://www.warcraftlogs.com/zone/statistics/31#boss=2587&bracket=1 . Direct your eyes to the far right column labeled "Parses". Even on a set boss, Eranog, with a set difficulty of Mythic and a tighter item level range of 376-378, the parses are all over the place. Now, this is a much better representation of performance than the "Show All" which we are continuously presented on here, but it is still completely inaccurate since the parses are not equal.

For any of these charts to be accurate, you have to have equal parses, on the same fight, with those parses in the tightest item level range possible. Otherwise, it is merely a representation of which class is getting played the most. That, of course, leads to many playing those same higher played classes, which leads to a more heavily skewed and inaccurate log.

It's simple statistics and while Warcraft Logs is a solid utility for the game, in this instance, it is being used wrong.

So, yes, it does matter that this information is wrong, ESPECIALLY because it will be used to build groups.

Yes its shows what class played the most, but ppl play the best classes the most, and if you play ingame the dps between classes are bigger than represented here, as a 388 destro lock a 377 marksman hunter did the same dmg than me after the marksman nerfed and the destro buffed, in the same gear they do a lot more dmg in m+.

[TheRagingwolf 258]

17 hours ago, MiNiGiANT said:

Yes its shows what class played the most, but ppl play the best classes the most, and if you play ingame the dps between classes are bigger than represented here, as a 388 destro lock a 377 marksman hunter did the same dmg than me after the marksman nerfed and the destro buffed, in the same gear they do a lot more dmg in m+.

The intent of Warcraft logs is to demonstrate through the numbers which class is doing better. However, when those numbers are not properly distributed, the results are inaccurate. The fact that we are merely seeing a popularity contest when we are led to believe it is a representation of performance is the problem. It means these charts are 100% inaccurate in terms of performance. Your statement that

Quote

Yes its shows what class played the most, but ppl play the best classes the most

misses the whole point. Warcraft logs is meant to tell us which classes are truly best, by the numbers, but by virtue of how this data is being utilized, all we see are what is played the most and NOT what is best. Just because a lot of people play one class over another does not mean it is best, it just means it is what those people play. Specifically, the limited number of people bothering to take the steps needed to collect data and submit it to Warcraft logs.

So, my point stands, these charts are completely inaccurate as a representation of actual performance of classes and stand only to show us which classes are being played the most. Because of this, the data should not be presented as a "this class is best" but more of a "this is the class being played the most".

As for your example, that is only one experience which can be affected by many variables. Anecdotal evidence is not representative of the overall picture, and is less reliable than utilizing results from a system such as Warcraft logs if it is used correctly.

Edited December 29, 2022 by Madyrn

[MiNiGiANT 26]

10 hours ago, Madyrn said:

The intent of Warcraft logs is to demonstrate through the numbers which class is doing better. However, when those numbers are not properly distributed, the results are inaccurate. The fact that we are merely seeing a popularity contest when we are led to believe it is a representation of performance is the problem. It means these charts are 100% inaccurate in terms of performance. Your statement that

misses the whole point. Warcraft logs is meant to tell us which classes are truly best, by the numbers, but by virtue of how this data is being utilized, all we see are what is played the most and NOT what is best. Just because a lot of people play one class over another does not mean it is best, it just means it is what those people play. Specifically, the limited number of people bothering to take the steps needed to collect data and submit it to Warcraft logs.

So, my point stands, these charts are completely inaccurate as a representation of actual performance of classes and stand only to show us which classes are being played the most. Because of this, the data should not be presented as a "this class is best" but more of a "this is the class being played the most".

As for your example, that is only one experience which can be affected by many variables. Anecdotal evidence is not representative of the overall picture is less reliable than utilizing results from a system such as Warcraft logs if it is used correctly.

Well i dont care about the details, i just stopped playing m+ coz all of my charachters are last in these lists and i dont get any invite or if i do i invited to a no clue group...

[TheRagingwolf 258]

6 hours ago, MiNiGiANT said:

Well i dont care about the details, i just stopped playing m+ coz all of my charachters are last in these lists and i dont get any invite or if i do i invited to a no clue group...

The details are what matter. The reason your toons do not get invited is because these charts are being used incorrectly. That is a huge problem and is leading to those groups making decisions with bad information which likely leads to greater difficulty.