- Red Post Collection: Quick Gameplay Thoughts: January 30, 9.2 Balance Hotfixes & More
- 1/30 PBE Update: Even More Tentative Balance Changes
- Patch 9.3 Server Maintenance
- PBE Preview: Papercraft Anivia & Papercraft Nunu & Willump
Table of Contents
- Ranked 2019 — /dev Update #2
- Quick Gameplay Thoughts: February 1
- Patch 9.3 Notes: Lore Edition
- Miscellaneous
- Reminders
Ranked 2019 — /dev Update #2
"Hey everyone, Ed “SapMagic” Altorfer here on behalf of the Ranked team. Earlier today, we dug into a few position rank-specific bugs that could affect the quality of matches in North America and Korea. In this post, I want to detail the issues we’re seeing and what we’re doing to fix them.
MMR Linking and LP Updates for Position Swapping
We found a couple bugs that affect how we handle position swapping and could lead to diminished match quality or incorrect LP updates in certain situations.
Issue: The links between MMRs are not tightening correctly for high-skill players.
We talked in /dev: Addressing Your Feedback on Ranked for 2019 about how, behind the scenes, we link your position MMRs together. We tighten those links the more you swap positions to prevent abuse or when you’re close to Master Tier where you have a single rank.
In our investigation, we found the link tightening wasn’t responsive enough for high-skill players. As a result, players who position swapped at Diamond+ found themselves in much easier matches than we intended.
Conclusion: We have fixed this bug for both North America and Korea as of today, January 31 at 2:00 PM GMT-8.
While some players were able to abuse this bug, we want to avoid solutions like resetting the Master+ ladder, which would punish the vast majority of players who earned their LP legitimately. Instead, now that we’ve fixed the issue, players who took advantage of it will get harder matches regardless of swapping, and their ranks will recalibrate as they lose games. We will monitor closely and make additional changes if needed.
We thought there was a possibility we could run into issues like this, so we preemptively locked Grandmaster and Challenger for the first few weeks of the season to ensure we could make fixes and respond to player feedback.
Issue: Some players may not receive properly diminished LP gains when they swap positions.
The issue above should’ve been mitigated by another detection system which determines when you swap positions and adjusts your LP accordingly. When you swap into a slightly easier game before link tightening has kicked in, you should get significantly less LP and have larger LP losses. While the actual position detection is working, it isn’t being taken into account in some cases when determining LP gains for the match.
Conclusion: Patching the previous bug should prevent the vast majority of cases where this logic is necessary, but we still believe adjusting LP like this is an important fallback for situations where—as we discovered today—MMR linking doesn’t work as intended.
We’re aiming to roll out a fix for this one before the weekend, within the next **24–48 hours**We fixed this bug on February 1 at 9:00 PM GMT-8.
Correctness of Match Data on Third Party Sites
With the introduction of position ranks, new information is necessary for third party sites to show all the right ranks. We’ve found some issues with the data we’re providing which can lead to inaccurate match details.
Issue: Some players are displayed as Master when they aren’t and vice versa.
Due to a problem with the underlying tech we’re using to calculate and track rank information we’re sending to the Riot API, we were giving third party sites incorrect data.
Conclusion: We fixed this bug yesterday, January 30 at 5:00 PM GMT-8, so players should be able to see who’s actually Master Tier and who’s not on third party sites.
Issue: Some matches may display the incorrect ranks for position played.
With position ranks, displaying the ranks for an individual game got a little trickier. Third party sites will eventually be able to display the ranks for each player’s position played, but we don’t currently provide the right information for them to do so.
Conclusion: We’re collaborating with the Riot API team and third party sites to provide them accurate data as soon as possible. Once a fix goes out, you’ll be able to rely on the ranks shown for matches again.
We’ll update this post with an ETA after we’ve resolved the other issues.
These bugs are painful to say the least, but on behalf of the team, thank you for helping us find and fix them as quickly as possible. If you’re seeing other issues that aren’t captured here or you just have feedback about the system, let us know in the comments. Thanks for all your help, and we’ll see you on the Rift."
Quick Gameplay Thoughts: February 1
"Hi folks,
Usual Disclaimers
These posts will often contain talk about future work we're doing, or planning to do, that isn't yet guaranteed to ship. The nature of the work could change or, depending on what we discover, projects mentioned may get delayed or even stopped. If you'd like to see a Tweet whenever a new one of these posts goes up: https://twitter.com/RiotMeddler
Rakan
We hit Rakan with a pretty substantial nerf to his W in 9.2, with the goal of reducing the reliability of his engage meaningfully. Looking at how that's gone it's probably hit him a bit too hard though. Current plan is to see how he performs in pro play, given the 9.2 changes were aimed at making him more balanceable between pro and regular play in part. Based off that and his solo queue performance we'll then either look at the best way to put some power back in general (outcome we think is likely) or at a way to give him power in solo queue without making him stronger in pro (less likely, but possible). That might include speeding his W up somewhat, since we erred on the large side with the change to make sure we got a meaningful difference but realize that comes at a meaningful feel hit for Rakan players. If that's not appropriate, or what we can do there's not enough by itself, other potential changes include an R buff (giving MS on button press towards enemies rather than on contact?) or Q buff (details TBD).
Jungle post 9.2
Wanted to share some initial observations and thoughts on how the 9.2 jungle changes have affected things so far. We're still digging into some of the data and the meta's still shifting a bit. These are preliminary current state of things notes as a result, rather than things we're certain will be the final stable state of the game
- The XP changes seem to have put junglers on average at the intended levels by game time. While they can look large on paper, effect on total jungler XP hasn't been that dramatic given changes to XP scaling (not base), don't kick in immediately and junglers do get some other sources of XP.
- We're seeing some junglers who rely on a strong power spike at 6 still doing fine, others struggling a bit. Need to look into those cases more.
- Power changes for different types of jungler haven't been that large in most contexts. Really high MMR (high diamond and above) does stand out though, with the same patterns seen in average MMR but to significantly larger degrees. As you'd expect junglers who focus on heavy farming and getting ahead of the XP curve, particularly those who use Challenging Smite well, have lost out on the most power. Kha'Zix and Graves are good examples of that, both of whom we felt could do with a bit of a power down.
On the flip side of things we're seeing tanks performing somewhat better, again especially in high MMR. That's pretty expected, given tanks tend to perform better whenever a position shifts towards lower economy (XP and/or gold) states, given their contributions are less scaling dependent overall. Tanks in average MMRs by contrast haven't changed too much, though do seem to be benefiting a bit. We're keeping an eye on their performance in both contexts. Good to see some stronger tank performance overall, bit cautious about discrepancies between skill levels and organized play versus solo queue potentially leaving them too dominant in some contexts while being fine elsewhere, given that's a trend we've seen in the past.
More thoughts plus probably some statistics to dig through to come, potentially next week. We don't have any systemic tuning planned for the jungle in 9.3. Will assess again during the 9.4 development cycle though, see if any follow up's needed to the 9.2 changes beyond individual champ adjustments. We have been looking at level 2 ganking, with a pretty heavy hit to Camille in 9.3. Xin Zhao we don't think's such an issue by contrast, given he's got more meaningful weaknesses than she does, so makes a much heavier trade off for that early strength.
Ranked 2019 so far
In case you missed it we posted a run through earlier in the week of how the ranked season start has been going. TLDR is so far fairly smooth, looking into a few things with position ranks though (NA and KR). Image here summarizes our current plans with more details available in the links below. Given there have been some meaningful changes to ranked versus last year and that we've got ongoing work on the system too it's a topic we'll keep talking about quite a bit over the next few months too, try and give you folks as much insight as we can into what we're thinking, what we'll be doing, what feedback would be especially helpful etc.
https://boards.na.leagueoflegends.com/en/c/developer-corner/8sVyEEQ5-ranked-2019-dev-update-1
Update since time of writing: We’ve been digging into some bugs/issues that have been affecting matchmaking for regions with position ranked. More on that at the link below, trying to get those addressed as soon as possible given how impactful this sort of stuff can be on your experience if you’re playing ranked.
https://boards.na.leagueoflegends.com/en/c/developer-corner/AnObL88Z-ranked-2019-dev-update-2"
On Renekton and Rek'Sai, Meddler commented:
"Renekton - We've backed off, given he's performing pretty well at present, especially since we've hit other laners who were performing stronger.
Rek'Sai - Nothing immediately planned, agree there's some work needed though (mixture of satisfaction and issues with power variation by MMR)."
When asked about VFX updates, Meddler replied:
"We're gradually working through VFX updates when time permits between larger projects. Likely, but not guaranteed, that there'll be some more in the future.
No current plans to update their icons. If we do another icons pass we'd be looking at whoever needs an update most as well, rather than who got VFX necessarily (champs most in need won't always be the same in both cases)."
On Master Yi, Meddler noted:
"Yi's a bit weaker, particularly in really high MMR. Shyvana seems to be doing well, though she got buffed at the same time, so harder to distinguish impact of jungle changes from her changes."
On Evelynn, Meddler replied:
"Overall looks like she's in an ok spot, though 9.2 likely left her a bit weaker. Looks like that's probably felt more in Diamond+, as with a range of other champs. I don't think she needs immediate changes, but do think she's one of around a dozen champs we should be monitoring pretty closely as jungle play stabilizes over another week or two."
On Sylas, Meddler commented:
"Want to give him a bit of time post hotfix buff and as players continue to learn him. Possible he's still on the weak side, don't want to over react and buff unnecessarily if we can avoid it though."
And on Aatrox, http://www.surrenderat20.net/2019/02/red-post-collection-ranked-2019-dev.html
No comments:
Post a Comment