You missed that 146 clear by Barb. CompletedTime timestamp is 1578023227000 or GMT : Friday, January 3, 2020 3:47:07 AM. I don’t understand why you would use any data source other than Blizzard’s API.
This means Barb is probably getting nerf next patch.
This was a blue post commenting on Crusader doing solo GR 146 on EU non-seasonal.
This barb from CH did 146 solo with 9477 Paragons.
Clearly, barb is in gray. Without hard cast, it may still under perform a little bit. Only few players can properly use hard cast right now.
Bliz and players still need some time to see the true power of zodiac rend. The data of barb era 12 is pretty close to WD era 11. Frequently changes the numbers of certain build is not a reasonable choice. This is also why gray zone was introduced by Matt.
Currently, the only issue is crusader in NS. Fix that issue, buff DH Nec and fix bazooka, buff other wiz builds(channeling builds, etc). In my opanion, that’s is what they should do next season.
I differ with certain elements of your interpretation.
With additional time, the top GR clears will only increase. It will be interesting to see what Blizzard’s blog post says. At the top end of the leaderboard, the top barb clear is +5 GRs over the target of 141. Within 1-2 GRs is considered good balance. The gray is 3-4 GRs. 5 GRs or more is most likely “correction” zone. Matt has made a couple posts already about this.
See:
The data is much closer to wizards of era 11 than witch doctors. Double check the median of the top 10 worldwide GR clears and GR efficiency (overall and at specific GR levels for example GR 125 and GR 120). You can also look at GR tiers as a function of leaderboard ranks. For example, rank #100 era 11 wizard cleared GR 131 while currently barbarians are at GR 130 and this era is a couple of months old.
I do think Matt means this single data. They will balance it according to the entire database, distribution. Not as simple as your data.
Top 1000 mean or similar, are not reliable. E.g. Top 1000 barb players may have higher avg paragon than WD. That’s why Prokhan’s analysis is more reliable than yours. Try his method.
Prokahn’s method was used in section 4 that discusses “Greater Rift Efficiency”. This method takes into account paragon levels. You may want to look carefully at the graphs and the table. The conclusions between the different metrics are largely consistent.
Let me walk you through section 4. There are 3 graphs.
The first is GR efficiency for era 11. You will notice that the black dots are outliers. These dots represent wizard era 11 GR efficiency. It is clear that this class is an outlier and overpowered in comparison to all other classes.
The second graph is era 12. The data is more noisy (presumably because the era is younger). However, even with all the data points you can notice that the yellow triangles (representing) crusaders are grouped toward the upper edge of the distribution.
The third graph is a simple representation of only four classes where one can directly compare era 11 wizards and witch doctors to era 12 crusaders and barbarians. In general, era 11 witch doctors represented by red triangles remain grouped separately (for the most part) while the 3 other classes are more overlapping (where era 12 crusaders are the most overpowered in terms of GR efficiency).
Thanks for the explanation. Here is my suggestion.
In my opinion, people use a curve (e^ - x like functions) to approximate the data. Before we say two curves are close enough, first we have to focus on an interval on the horizontal axis. In your data, we should focus on the right half of the interval (Cuz we all care more about 130+ clears not 110 clears), then calculate the area between the two curves.
My personal conclusion is:
barb is close to wiz on mid-level grifts: e.g. 120-130.
is more close to WD on top-level grifts e.g. 135+. (This is what we care most)
I do not know that it is fair to claim that “we all care more about 130+ clears”.
If you want to only consider 130+ GR clears, than it makes sense to look at the worldwide leaderboard to increase the number of data points that are available for analysis. At the very top end, you can look at the top 10 worldwide leaderboard as done in section 2.
For ease, I took the worldwide data from the zero empathy leaderboard.
The area under the curve for GR clears 134 and above for era 11 witch doctors is 186.8 while for era 12 barbarians is 267.0. As you mentioned their difference in GR efficiency becomes even more pronounced at lower GR levels where barbarians are more efficient.
I could not go down to GR 130 because the ZE leaderboard only lists the top 200 worldwide (and the rank #200 barb cleared GR 133).
Thanks for posting another wall of data, without really understanding what it is you’re posting, other than taking a not so subtle swipe at Barbs, which seems to be your only agenda.
Here’s some more data for you, that actually has context.
While taking a break from seasons I’ve been playing around with a few different classes and builds on non-season, rattling through GR90’s while leveling gems, trying out new sets, etc.
At Paragon approx. 1,500, my Rend Barb can do GR 90, on average in 3-5 minutes, depending on rift.
While messing around with other classes and builds, I am so far up to 5 builds that can do GR 90 faster, on average, than Rend Barb.
Lets see if we can agree on some things in the wall of data.
In era 11, the data shows that wizards were “overperperforming”.
Do you agree with this statement? Yes or No
Do you think that it was reasonable that in patch 2.6.7 that wizards were nerfed? Yes or No
In era 12, the data shows that crusaders are the most “overperperforming” (even more so than wizards in era 11)?
Do you agree with this statement? Yes or no.
Do you think that it would be reasonable that in patch 2.6.8 that crusaders will be nerfed? Yes or No
In era 12, does the data demonstrate that necromancer are the weakest class? Yes or no
Do you think the necromancers should be buffed in patch 2.6.8? Yes or no
To your knowledge, the data that is presented is accurate? Yes or no
To your knowleddge, the data analysis was done as described? Yes or No
To your knowledge, has Matthew Cederquist made several post about game balance (two of which were linked earlier in this thread)? Yes or no
The data that I presented does not inform what classes can speed run GR 90 @ 1,500 paragon. It also does not inform which classes can serve a RGK, trash killers, or zDPS in speed runs.
I am not going to tell you how powerful Rend barbs are. There is data for you in the OP and other posts of mine in this thread to assess and draw your own conclusion. The era 12 analysis is clear. Barbs are the second strongest class using distinct metrics. This thread has looked at top GR clears, GR tiers relative to leaderboard rank, and GR efficiency