Table of Contents
- [UPDATED] Looking at upcoming changes for 5.18
- CertainlyT on Mordekaiser Rework and Solo/Duo lane.
- Upcoming test for XP gain in Russia
- Lyte Ask.FM Roundup
- No Lissandra Passive Changes planned
- The value of Last Hitting
Looking at upcoming changes for 5.18When asked who is on the balance "hit list" for the upcoming 5.18 patch cycle, Meddler shared:
"Some of the stuff we're looking at for 5.18:
On watch/needs investigation
- Skarner/Darius/Garen (where are they post 5.17?)
- Devourer (too strong?)
- AD Ezreal (too weak?)
- Lulu (too strong solo, too weak support?)
- Vayne (too reliably strong?)
Odds are high some other champs will end up being worked on too, and that some champs above will end up being taken off the list as work on the patch continues. The above are some of the key things we're looking into though."Meddler noted:
"What we're current testing:
- Q grants AP for champion assists as well as kills
- W starts at the beginning of Veigar's cast time, not the end"
As we nerf duo lane Morde we will be looking to put some power back into solo Morde. I doubt he'll get back to a 50% win rate though.Meddler explained:
"We're not trying to remove Lulu from solo lanes, we do think she can be too strong there though, particularly in organized play. Additionally, despite that strength, support's still a position a lot of player's do choose to play Lulu in. We feel Lulu's overall both fun and healthy as a support, so would like to try and put her in a slightly better spot power wise there as a result."[UPDATED] He continued:
"Changes we're testing below, with the goal of making her a bit more vulnerable in some match ups, while preserving her survivability against AD damage dealers.
- Base stats
- Base Armor increased :: 19 >>> 23
- Base Health reduced :: 553 >>> 518
- Pix, Faerie Companion (Passive)
- Pix can trigger Tribute (from the Spellthief item line) once per Lulu basic attack (instead of never triggering it at all)"
When asked if there are any plans for Zeke's Herald, Meddler noted:
As far as I know we don't have any plans for Zeke's at the moment. Win rates around 60% aren't unusual for items that aren't the first or second item a champion buys, since they're not found equally on winning and losing teams. Supports that are already winning are more often the ones that actually get to build a Zeke's in the first place, so you'd expect it's win rate to be well above 50%. Same sort of thing happens with various other large cost, non rushed items (Deathcap for AP, Last Whisper for ADCs etc).Meddler noted:
"We'll start by looking at plant AI, see what can be improved there. After that depends on how that investigation goes."Scarizard also added:
"Zyra's a tough one for us to figure out. We're looking to have some changes to help her out in 5.18, but nothing on the large-scope level that she might need due to it being the patch for Worlds.
This isn't set in stone, but almost every conversation i've heard or participated in about Zyra has been focused on making her more about plants and plant interactions as a means of having 'zone control' that isn't simply 'here's a large aoe'"
As for Vayne, Meddler noted:
"TLDR: It's possible Vayne's too strong, so we're looking at her. 'Everyone says' doesn't necessarily equal true though.
There are a lot of anti Vayne threads on Boards, with you there. That indicates that, at the very least, a lot of players are finding Vayne really frustrating. Frustrating doesn't necessarily mean overpowered however, though the two are correlated. Champions sometimes generate a lot of frustration feedback even when balanced or underpowered (e.g. Blitzcrank really annoys some people regardless of his balance state and waves of anti Blitz threads crop up every so often). It's also important to remember that the population of Boards isn't necessarily representative of the entire player base (it's a subset of the english speaking population from certain servers).
Having said that though it's very possible Vayne's too strong at the moment. Her win rate in solo queue is quite high for such a popular champion. That, particularly when combined with the frustration she's causing, means she's worth investigating. Doesn't necessarily mean nerfing Vanye's the appropriate outcome though. It's possible we might find that Vayne's too dominant at present because the champions that should be keeping her in check in lane are weak and need to be buffed. Or because items she scales especially well with are overpowered. Or because other meta conditions (lots of HP stacking champs for example) are very favourable to her. Or even some combination of all those factors and too strong a kit, hence the need for some investigation."In a reddit discussion concerning Meddler's Vayne comments, Scarizard elaborated:
I'd like to expand on this point, because i think Meddler's intent may be misunderstood; It's not so much that Vayne 'doesn't have a glaring weak point' - It's just that her weaknesses period aren't really manifesting.
Not every champion should be 'strong early, weak late' or 'weak early, strong late' (though this paradigm is often commonly seen among ADC's). However, 'okay all around but not amazing at any one specific thing' is how we end up with champions like Brand and Vel'Koz - they're fun and feel unique, but in the end are indistinguishable from a lineup of mages because they don't excel in a specific area (why not take someone that has a strength that stands out?)
Now, stepping further towards Vayne - 'okay all around, but not amazing at one specific thing' isn't true in her case. She's pretty top-tier at skirmishing and cleaning up, but her relative weaknesses (waveclear, sieging) clearly aren't presenting themselves at the moment. Reliability here simply means 'this champion's supposed to be a risky, but if you just pick it there's high likelihood you can take over the game anyway'. I can think of a few other champions (Gangplank, Fiora) that might be too reliable for how they're intended to play out atm, but that's my opinion.
I touched on a lot of different concepts here super lightly/quickly, so feel to hit me up if you wanna dive deeper. the deal w/ Vayne though is we're not exactly certain on what a changelist looks like for her (or if we're even truly going to do anything about her for 5.18, or all) so Vayne changes aren't set in stone. That said, from the conversations I've had if we were to change her we'd be looking at ways to detract from her high-end as little as possible and instead focus on 'what can the enemy do to react to her or stop her from going off?' this could present itself in a lot of ways, but like i said, still far out.
just wanted to drop by and add my perspective - i don't think we're against having some 'safe' picks generally, but well-rounded champion is a tricky (but not impossible) place to be as our roster expands
"Let me preface this with this: no one is thinking of lowering Vayne's range. The general feeling is she's lost a lot of her risk factor - I'd agree that people pick vayne because she's fun - but i'd argue it's also because she's super reliable and strong; that said, we're not looking to make her disappear entirely. As for other marksman, there's a lot of stuff you'll see post-worlds focused on the role and its general power/satisfaction ratio, though i'm not really looped in to the specifics, nor would this be the avenue to preview those.
For the interim, we had buffs to Lucian and Twitch in 5.16 and we're currently planned for even more tuning to the 'lesser' marksmen in 5.18."
"Simply put, yes.
The design teams as a whole have been looking for ways to make Marksmen more maintainable in the future on a number of different fronts - no ETA on when you'll see the beginnings of it, but it's a definite priority (though nothing before worlds)."
CertainlyT on Mordekaiser Rework and Solo/Duo lane.
"Hey MaliciousMetal,As for why they didn't foresee solo lane Mordekaiser being so much weaker than Duo, he noted:
Mordekaiser definitely isn't performing in solo lanes to the level we expected. Some of this is players adjusting to his new playstyle, which favors his jungler playing closer to his lane (strong 2v2 counterganks and Morde doesn't lose experience). Some of this is adjustment to his new build paths (I'm pretty convinced that Gunblade rush isn't nearly as optimal as it is common). However, a lot of it is that Morde is just much weaker in a solo lane than I expected. We are beginning to test some alternate paradigms for his W that include self-castability to hopefully give him a substantial boost in solo lanes. Look for changes here in 5.18.
That being said, Morde is not likely to return to the "immovable object" capable of permanently keeping a shield large enough to slough off entire trades. To the extent that is what you enjoyed about Morde solo lane, you're out of luck. He should retain great carry potential, amazing in-fightturn around capability, and best-in-class capability to completely turn around a game if gets the right ghost.
As to Mordekaiser being unbalanceable, I don't think that live Mordekaiser demonstrates this. Many champions require teammate to bring them to peak effectiveness -- peelers such as Janna, engagers like Amumu, and fragile carries like Jinx are unlikely to win a game without ally support. Indeed, this is an important part of League -- champions should find their greatest success when properly paired and coordinated. Champions who lack synergistic strengths are generally relegated to isolation/split pushing. Unfortunately, solo pushing side lanes is a narrow space to occupy. Due to their agnosticism toward the other nine champs in the game, we tend to see one or two split pushers dominating the rest at any given moment in the meta. I think there's a balance to be struck when it comes to the W that current errs too heavily on the side of dependence, rather than synergy."
"No, it wasn't obvious to me. Perhaps I'm just dense!
It was obvious that players would have to play differently in order to succeed, such as working more closely with their jungler. Our internal testing indicated that this was possible. I mis-estimated the ability of players to consistently do that in Live games. Again, I'm not anticipating that Morde will have a magical "50% win rate" in solo lanes in solo queue after changes unless players make efforts to make use of his kits tools. Spam pushing and rushing Gunblade before boots are probably not that though.
It's actually quite difficult to know how a champion will perform on the Live environment... you guys are collectively much better than us at this, crowdsourced optimization over tens of thousands of games is an extremely powerful tool for solving complex problems, and likely always will be no matter how many talented individuals we task with theorizing balance."He continued:
"It's tough. We also listened to the "very informed" testers who told us Skarner was weak. Feedback is always contingent upon the choices made by the playtester. In Mordekaiser's case, I felt the playtesters were not making the right choices but that over time they would begin to make them. I was incorrect.
In the end, players will still have to adapt to a play pattern that opens more windows of meaningful response for the lane opponent. My task is to give players the ability to be able to judge whether they are making progress toward figuring the play pattern out. When a champ is underpowered in a role, that's often difficult because even if you do everything right, you might still end up failing."He continued:
"My job is to balance the competing interests of all 10 players in the game. It's also to keep the game dynamic. This means I have to balance feedback from multiple sources. In this case, I had a lot of feedback from people saying Mordekaiser was a nagging problem on the game as a whole to balance against feedback from the fewer but more passionate people saying they wanted their champion unchanged. Based upon the totality of the circumstances, I made the decision I felt was better for the overall health of the game. I don't regret it, though obviously it is difficult as a designer to change something that many of your players enjoy."
When asked to explain the purpose of the 5.16 Mordekaiser changes, CertainlyT explained:
what i want to know is, why did the morde changes go through in the first place if everybody that played morde didn't want them.
Makes no sense to pretty much destroy a champion and replace him with a new one but just give them the same name.
Mordekaiser is not even close to the same champion anymore and it fucking pisses me off
Good question. I suspect you will not like the answers, but this was my thought process:
First, many people liked the changes. It is challenging when designing for a group to please everyone without simply making the champion more powerful. It is challenging even when designing for an individual to upset their habits, to freshen the decisions they make, without producing frustration. Nevertheless, it is vital for a long lived game to freshen the challenges players must overcome to succeed.
Second, Mordekaiser was not in a healthy place. Many lane opponents lacked meaningful ways to interact with him after level 4 or so. This problem didn't seem big because Morde was an unpopular champion, but on a per game basis he was still one of the least interesting champions to fight. The attempted solutions we looked at internally compromised important facets of his mechanical identity (e.g., low mobility, no CC, huge damage, huge durability). These changes produced what to me seemed the best balance of game health and identity maintenance.
Upcoming test for XP gain in Russia
Here's Riot S0undwaves with more information:
"Hey all,Riot Soundwaves continued:
We're looking to make some upgrades to League’s progression systems long term. Right now we’re thinking about the level 1-30 experience. Specifically, we’re taking a look at how long it takes to level an account all the way to thirty.
The numbers here are pretty disparate: the grind from 1-30 takes hundreds of games, and League’s matchmaking systems can accurately place players pretty swiftly. While we still believe there should be a pretty big buffer between new players and fiercely competitive modes, we think the number of games between 1-30 could be a bit too big. To test that hypothesis, we’re running a test in Russia only, doubling XP gains on the server from 9/1-11/20. After the test concludes, we’ll use the data to inform improvements to the leveling experience on a global scale.
We chose Russia because it’s a younger, growing territory for League, making it ideal to test systemwide changes on. If anything, the test might help Russian players bring some of their friends into the game.
Shortly after the test wraps, we’ll circle back with you, share some of our learnings, and let you know how they could shape future changes.
Let me know if you have any questions!"
"Yep. Just to be clear, we're not looking this as a single solution to progression. Hopefully the learnings from this will help inform our longer term plan."
When asked about IP gains or runes costs, Soundwaves noted:
In a thread from a lower level player concerned with the difficulty of last hitting and inquiring why it is necessary to the game, Meddler commented:
"We won't be changing anything regarding IP gain or rune costs specifically related to this change. It's certainly something we're cognizant of though and we're looking at potential solutions longer terms if we adapt a quicker path to 30"
Lyte Ask.FM RoundupLyte's ask.fm!
[Link] Any plan to give us feedback when we report someone and a punishment was given?
We've already started a conservative version of this feature called Report Feedback. You'll get a pop-up in the client when you are the report that "triggers" the punishment. We've designed it to be the report that "triggers" the punishment for now because we're going to be adding Intentional Feeding, Chat and Ranked Restrictions and more to this Report Feedback System and want to be sure not to spam players with too many pop-ups. As many as 20% of players were not interested in these pop-ups, although quite a few players were interested in getting ALLLLLLL the pop-ups, so we have to balance a fine line here.
Some players have suggested just adding an options menu to toggle whether you want these pop-ups or not, but adding a toggle for almost every system or feature like this would be a pretty awful experience and result in 30 pages of Options Menus.
[Link] So, players who got banned can't contribute on the tribunal rigth? But players who got banned in the previous season but not in this one can? And can players who improved their behaviour contribute on the tribunal?
Players who are currently in negative standing (so had a game ban recently, or active Chat and Ranked Restrictions, etc) cannot participate in the Tribunal voting system.
[Link] When will team builder's requirements be lowered?
We're making quite a few changes to the new Champion Select experience in Team Builder Draft that will make it more low level and new player friendly. When we launch this, we'll most likely implement some of these upgrades to the experience for Team Builder "Normals" and reduce the level cap of these experiences to Level 3 but these details are still being discussed.
[Link] Will Ranked Team Builder Draft mode be out for the 2016 season so that players can get the roles they want to climb ladder?
We'll be talking more about our launch plans in the next 2 months, we're just working hard on the system right now to make sure we can deliver something players will love.
[Link] Hi Lyte, since "the system" also stores the pre/post game chatlogs - is it positive/neutral/negative to call a (preferred) lane? And is "the system" able to understand different languages or is it just english? If only english, did you plan to extend it to multiple languages?
It's positive to call a preferred lane; however, it's an unspoken rule among players to follow Pick Order. So, if 1st Pick calls Jungle, and 5th Pick calls Jungle, we agree that 1st Pick should get Jungle. The system is able to understand different languages!
[Link] Simple question. Can Riot be more Transparent with PB statistics? We need credible data not what you feed us. We need research done by reputable companies. Because your statistics are misleading.
First, very few companies release full blown studies on their work to "prove" their statistics. Secondly, we're extremely transparent compared to most companies about our data, and you can see numerous peer-reviewed publications about some of our work conducted by external universities. For example:
1) Regulating Anti-Social Behavior on the Internet: The Example of League of Legends (Kou & Nardi, 2013) from UC Irvine
2) Exploring Cyberbullying and Other Toxic Behavior in Team Competition Online Games (Kwak, Blackburn & Han, 2015) from QCRI, Telefonica Research and University of Washington.
3) Predicting Crowdsourced Decisions on Toxic Behavior in Online Games (Blackburn & Kwak, 2014) from University of Washington.
4) Governance in League of Legends: A Hybrid System (Kou & Nardi, 2014) from UC Irvine.
We're also officially collaborating on some work coming out later in 2016 with MIT and University of York and both are coming out in peer reviewed journals.
[Link] hey Lyte, I really like what you do <3 But, there is something i'd like to know about the report system, if someone report you for "Feeding" when you got a bad game, does it can involve into a chat restricted or something ? Potato > Tomato or Tomato > Potato ヽ༼ຈل͜ຈ༽ﾉ
No, our new Intentional Feeder Detection will differentiate between intentional feeding and just having a bad game so if you just had a bad game nothing will happen to your account. As usual with new systems, we'll start off conservative with the settings and ramp the system up over time when we get more and more confident with the accuracy of the system.
[Link] Hi Lyte ! Thanks for answering so many questions ! I've got one myself : for the people that got low-priority queues BUT got a message saying they've improved on that point, will they still get the rewards ? Thanks ! :D
You should be fine, in the end, I don't think we're doing anything with Low Priority Queues and Ranked Rewards.
[Link] Doest riot do IP-Adress bans or do they just ban accounts?
IP Address Bans aren't very effective, and just have the potential of collateral damage on other players. For example, if you IP Address Ban a university, then every player in the dorms tend to be screwed. Or, if you IP Address Ban a PC Cafe, then every player in the cafe can no longer play League. In these cases, 1 player's behavior can punish 100s of other players for no reason.
[Link] How do I know players with extreme behaviors like intentionally feeding will actually get punished?
In the future, you'll get Report Feedback for intentional feeders too. So, you'll see a pop-up that they were punished for their behaviors. You'll also be able to see some feedback about what happens to players in the new Tribunal voting system, where you can log in to vote on in-game behaviors like chat or game toxicity and help decide whether the behaviors are OK or not.
[Link] You have given rewards for good behavior in the past and given punishments for toxic behavior. In your opinion, which has been more effective?
What's interesting is that there isn't a silver bullet to solving online behavior. Punishments were great for aggressively going after players showing toxic behaviors, but positive reinforcement was great at nudging neutral players to be a bit more positive. We did see some signs that positive reinforcement "hurt" the motivations of already positive players a bit, so it's something we have to be aware of as we design future reward incentives.
No Lissandra Passive Changes planned
While it has been a while, Meddler also answered a question regarding his previously discussed plans to change Lissandra's passive and commenting they have no current plans to change it:
"No current plans. We did talk about possible passive changes a couple of patches back, best idea suggested though was the change to the self cast ult, which ended up testing really well, so we went with that instead."More on Lissandra's 5.16 self cast R changes can be found here.
The value of Last Hitting
In a thread from a lower level player concerned with the difficulty of last hitting and inquiring why it is necessary to the game, Meddler commented:
"We feel last hitting's good for the game for a number of reasons, including:
It's an important expression of skill
Last hitting, when you're by yourself, is relatively straight forward. As soon as you add in enemy players however it becomes a lot more difficult. Last hitting well, while at the same time harassing the enemy when they try to last hit, watching the rest of the map, looking for opportunities to all etc can be really difficult. Shifting from a model where 'last hit = gold' to 'any hit = gold' or 'proximity = gold' removes or reduces a bunch of laning skills we believe are good for LoL.
It drives action
Last hitting in lane creates moments where at least one champion is vulnerable as they move forwards more than they otherwise would, briefly stand still when they otherwise wouldn't, use mana/CDs to farm instead of saving them or harassing etc.
It offers ways to make champions more distinct
Because last hitting's important the tools a champion has to last hit with or interact with other champions as they last hit are particularly important in lane. To look at an example a notable amount of Ezreal's distinctiveness comes from how he interacts with minions in lane. His Q is a safe, cheap tool that makes it easy to pick up some farm from a distance. On the other hand he lacks AOE wave clear apart from his ult, so he can be denied quite a bit of farm if you push him to his tower."He continued:
Im surprised you guys take last hitting so seriously. Albeit, I never thought that someone would end up complaining about it, cause I thought it was just a mechanic of the game that everyone took for granted..........like buying items. I'll admit, it is surprising to see this detailed of a response to something so fundamental of the game, very few people ever think about it unless they have kill stolen from them.........and then they forget about it 3 seconds later.
Digging into the fundamental stuff that it's easy to take for granted can be really useful in game design. The basic actions that make up the game (movement, last hitting, leveling up, buying items, death timers etc) are the foundation that everything else gets built on. If you don't get that stuff right you could have the best designed characters possible and you still might not have a fun or interesting game. Modifying them can be risky as a result, but there are a number of problems (e.g. excessive income disparities between roles) that are sometimes better solved at that foundational level than with individual pieces of content like items."