The second week of Mythic+ this season has performed quite well, and Nightstalker314 is taking a look at the weekly data provided by Raider.IO. Chinese data is once again fully accessible. The impact of these new metrics is significant. For example, we can look at the 18.3% drop-off in total runs for the regular regions (US, EU, KR, TW). Once we look at the global numbers, this shrinks to 14.64% because the drop on Chinese servers was only 9.56%.
Therefore, we will be highlighting the difference between global numbers with and without China and taking a look at the Chinese data in particular. For the comparison between seasons, we will stick to the four regular regions for now. The standard formats of dungeon split and key level split will be based on the global metrics from next week on. Let’s dive into the details.
Weekly Total Run Performance Compared To Other Seasons
As mentioned above, the drop-off from week 1 to week 2 is around 18.3%. This is quite a steep decline, but also quite similar to the value in Season 3 of Dragonflight. Within the graph, both lines almost overlap. Both seasons had a rather easy start. In both cases, much of the rating progress may have been made in week 1. So week 2 was already seeing more focus on just filling the vault with fewer extra runs.
In total numbers, TWW S3 is still ahead of S2 and slightly behind S1 (regarding the four standard regions). Global numbers, including China, seem quite massive and highlight China’s dominance compared to other regions. Week 1 of S3 had a total of 4,123,520 completed runs. Week 2 sits at 3,519,789. Last week, our estimate for the missing data in China put its total runs at around 1.68 million. It turns out that was quite off the mark. Numbers for week 1 were 1,727,773, and dropped to 1,562,512 in week 2.
China’s share of the global total was 41.90% in week 1 and has increased to 44.39% in week 2. Chinese numbers might be 50% or more of the global total during the later weeks of the season. Especially during dungeon week events or the start of the Turbo Boost in October.
Total Runs Across All Dungeons
The split across all dungeons seems to have evened out compared to week 1. The overrepresentation of Operation Floodgate keys last week might have been due to an error in the algorithm that generates your (new) key. Eco-Dome has become very popular now. Its reputation as the easiest dungeon may have spread and led to this. Halls of Atonement might benefit from the familiarity of being active in the first three seasons of SL. But its tuning is still rough. The remaining dungeons are rather close together, at least for the four standard regions.
Dungeon tuning is inconsistent, especially with the emphasis on Fortified in the lower gilded crests key levels (7-9). The average in-time percentage of 83.91% is a slight improvement over last week. Eco-Dome is still number one with 90.92%. But the push for higher item levels and being a 3-boss dungeon during Fortified week drops the in-time percentage quite a lot. Streets of Wonder (88.48%) got a massive benefit from the major nerfs during week 1. Further, a 5-boss dungeon is seeing the least impact from Fortified in general. Gambit also improved significantly (87.27%), and it seems that the boss tuning in week 1, combined with Tyrannical, was quite rough.
Existing Tuning Issues
Priory (85.74%) and Dawnbreaker (84.17%) dropped quite a bit, probably due to the impact of Fortified. Operation Floodgate (83.87%) barely changed. And Halls of Atonement (76.90%) as well as Ara-Kara (72.98%) are far behind. These two need some major adjustments. Trash in Ara-Kara seems to be turning into an issue during Fortified weeks.
The Global Average
Looking at the global numbers (including China), we see a similar split with Ara-Kara falling behind. Not just in terms of share but also success rate. Overall, the in-time percentage drops by 3.19 percentage points compared to the four standard regions. And most dungeons reflect that change with slight variation. But Ara-Kara, already being last place in the first graph, is now dropping off even harder (-5.22 points).
The Impact of Chinese Metrics
And once we look at the Chinese data, we can clearly see why. Just above 60% of all keys in Ara-Kara were timed. And the overall success rate is a full 4 percentage points lower than the global average. The mindset on asian servers in general seems to lean into running and finishing keys, no matter the outcome. But let’s look at that in more detail.
Split Across All Key Levels
The key level split for the four standard regions seems quite usual. A major focus on key level 6 for Hero track loot and the highest amount of runed crests. Followed by a bump within the gilded crest range towards the BIS vault key level. 29.18% of all keys are already completed at level 10. And 47% are done at 10 or higher. Compared to Season 2, key level 10 is seeing around 50% more activity during week 2, both in relative share and absolute numbers. For Season 1, the difference is even bigger, with 3 times the metrics.
The impact of Fortified at key level 7 is a drop-off of 9.6 percentage points when it was 12.6 in the previous week. From 9 to 10, the change is almost unnoticeable. A drop of 1.1 points compared to 11.45 in the previous week. All in all, it’s quite the improvement. Especially for keys at level 6 and above. Key level 12 represents the only major dip in the curve. Given the multiple incentives – a push towards 3k rating, more crests, and resilient keys – this key level is challenging players’ ambitions throughout the season.
If we look at the global metrics, the overall split doesn’t change that much. We can see a relative decline in lower keys, even for the early gilded crest levels. The focus shifts more towards high keys, with 31.93% being done at level 10. 10 or above already represent more than half of all keys with 52.59%. The success rates are slightly lower across the board. And a major dip is showing for key levels 10 to 12.
The Key Level Split in China
And here is where we see the massive impact of these metrics. 35.38% of all keys are already done at level 10. A value that we never saw at any time during S2 for the other four regions. Almost 60% of all keys are finished at 10 or higher. The share of runs done for gilded crests increased from 75% (other four regions) to 80% (global) to almost 87% in this region. And the success rates are a lot lower, reflect the effort to farm as many vault entries, crests, and loot as possible. So for the future weekly round-ups, keep in mind that on a global scale, the metrics on Chinese servers play a major role within the data.
Conclusion and Week 3 Forecast
The usual metrics look quite promising for a great final season of the expansion. The two remaining outlier dungeons, Halls and Ara-Kara, still need some decent tuning. For future analysis, we can’t really ignore the Chinese data, given that the region represents 40-50% of the global M+ activity. As long as the data remains accessible, we’ll compare weekly numbers on this basis. From time to time, we will highlight the major differences in the details. So keep in mind: The total numbers will look quite high, while the success rates will look worse.
As of this writing, hourly numbers for the US and EU were 11.1% and 11.5% lower than the previous week. The drop-off seems much smaller than in week 2. Chinese data for hourly entries could not be tracked during week 2. But it is safe to assume that their numbers dropped even less. But we will check on that in the next round-up.
