S3 Championship Tickets, New Icons Info, Sivir, Possible Removal of Improved Summoners in S4, Update on Skarner, Sale, and More

Important: Free Champion Rotation

S3 World Championship Tickets Available Soon

Originally Posted by Riot (View Source)

Season 3 is coming to a head and teams around the globe are jockeying for position on the road to the World Championship. Soon, you'll be able to secure your tickets for playoffs and the championship itself. The playoffs open at the Galen Center on September 15 and run up to the October 4, when two teams will battle for the Summoner's Cup at the Staples Center. Tickets will go on sale August 15.

Jump to lolesports.com for the absolute latest on ticket info and to keep tabs on the latest League of Legends esports news.



We have announced the schedule for the Season 3 World Championship. Tickets will go on sale August 15th at 12:00pm PDT / 21:00 CEST.

Group stage will be held on September 15, 16, 17, 19, 20, and 21 at a live studio in Los Angeles.
Quarterfinals will be held on September 23 and 24, also at the Los Angeles Studio
Semifinals will be held on September 27 and 28, at the Galen Center (same location of S2 World Finals)
Finals will be held on October 4 at the Staples Center.

For more information, check out the announcement on lolesports.com.


New Item Icons

Originally Posted by Riot (View RedTracker Source)

Hey there!

I’m sure you’ve noticed some of the items have had their art updated and replaced. As the aesthetic of League becomes more cohesive and solidifies in quality, we’re working hard to update all of our old art. In particular, the Visual Design team is working hard to innovate and improve on our user interface and visual identity. We have already been raising the icon quality bar on our champions and summoner icons, and it’s about time items got some love. Because these icons are so important to our players, I thought I’d walk through some of the decision-making process involved in developing the new aesthetic.

The primary goal of the reworks was to improve clarity, and then to build thematic connections between the item visuals and the other aspects of their design. Familiarity was a big concern as we began working on these—how far can we push the designs until they become too hard to adapt to?—but a few icons had very little in their visuals that could be held onto. When we were looking into which icons to re-design first, we wanted to seek out the ones that presented some of the biggest challenges to redesign, the ones whose themes seemed seriously out of line with their visuals, or whose current art didn’t localize well. From there, we wanted to find ways to build them into the rest of our game’s world in a way that felt consistent and logical.

Let’s look at how each one was chosen and how we developed them:

Abyssal Scepter

Old Icon

Abyssal’s older art was a bone on a stick. It was pretty lacking in terms of design, and nobody really felt that the whole “abyss” theme was captured by it. The first explorations toyed with keeping the skull, but it just wasn’t exciting enough. Somebody on the team suggested pushing more in the direction of something extremely warped, exaggerated, and unusual, but retaining the feeling that this was the skull of some terrifying creature. The more we pushed, the more we realized a hybridization of the two would provide the most unsettling result. A vaguely human skull at the end of a staff made out of its own spinal column, wrapped in worn strips of leather. The bone went from bleached white to dark, decayed brown, which also brought out the silhouette and distanced the design even further from appearing too human. The final touch came from one of our splash artists, Alex Flores, who really pushed the S shape in the staff to where it is now. In the end, we had this:

New Icon

Catalyst the Protector

Old Icon

Catalyst’s upgrade had a few very specific goals going in. The first was to change the rune to match the kind of shape language we would see in our game today. The shape we used was designed by one of our other Visual Designers, Zach Roberson, and it definitely feels like it belongs with the world we’re building. The second consideration we had was to make it purple to match the idea of blue and red crystals combining into one new item. The result was something with a really unique color palette and silhouette, brought up to the quality we want to see out of every new icon:

New Icon

Deathfire Grasp

Old Icon

More than any other icon so far, Deathfire Grasp’s redesign pushed the limits of what we could do with changes. The original icon felt like such a missed opportunity, and had too much overlap with Abyssal Scepter’s visuals. We wanted to separate the two, and make each feel more unique. And then there was this great theme, right there in the name “Death Fire Grasp.” It sort of stuck out in our earliest discussions as the obvious thematic tie, but being obvious isn’t always the best way to design. So we tried a few iterations and different ideas, but in the end we kept coming back to a staff with some sort of hand or claw shape on the end of it holding a spirit flame. Once we decided where we were taking it, we felt like the next step was to bring it into the world. Where would you find a staff holding a captured soul? The Shadow Isles. Looking at champs like Hecarim and Thresh, we tried to give the staff some of the same flair, all the way down to the color pallette, blue-green effects and lighting, with warm iron for the metals. The new design was a huge departure from the original, and we had a lot of internal discussions about whether we were pushing too far. In the end we agreed the added clarity and visual theming made the icon more distinct and easier to associate with its thematic. And here we have all of that process in its final form:

New Icon

Glacial Shroud

Old Icon

Glacial shroud also really missed its thematic. We started looking at this around the time we were developing the Freljord event, and with all this lovely new art for our wintery domain, it made sense to make Glacial Shroud more “glacial.” The first step was to remove the lion. After abstracting the details of the face, we saw some shapes that reminded us of the knot shapes on Howling Abyss, so we adjusted them all to fit in. The blue ring matched the True Ice of Freljord, so we inlaid a ring of it onto the medallion. Next was taking the gold and toning it down to the kind of muted color palettes we see on Trundle, Sejuani, and the other Freljord champs. Finally we took those wispy shapes and hinted at a frost-magic fabric the pendant holds together as a brooch, giving us a very literal shroud to play with. The new icon went from being something a bit out of place, to something that could really fit in to our world:

New Icon

Haunting Guise

Old Icon

For Haunting Guise we set out to actually make something a bit spooky. A clown face was a specific nightmare, but we wanted something that would feel appropriately haunting to anybody. The shape of the mask we ended up with had a lifeless, cold feel. The red and white hinted at something Ionian, and the green magic matched some of what we were seeing out of champs like Karma and Master Yi. We drew on lots of our Ionian champs for reference, but the biggest help on this icon came from the Visual Development department, where they really helped us nail down that Ionian style. The result:

New Icon

Hextech Revolver

Old Icon

Hextech isn’t extremely well-defined to our players right now, but there’s plenty in the various champions to suggest at what it could be. Again, Visual Development played a big role in helping us achieve a look the felt appropriately “hextech,” and helped steer the direction to something more specifically Piltovian, which felt like a natural fit for the item’s origins. The gun replicates a lot of the aesthetics seen in Jayce, Vi, and Graves, where magic and technology play off one another to form unique implements with an unusual design principal. The gun doesn’t function without the magic, and the magic is useless without the gun to harness it. It’s a symbiotic relationship:

New Icon


Old Icon

Right out the gate, Thornmail felt like Noxus. Bearhug your enemies with spiked mail? Totally Noxus. Again Visual Development helped us find that perfect mix of utilitarian design and aggressive, militaristic aesthetics that makes it fit so well into our world. Darius, Draven, and other Noxian champs heavily influenced the design, right down to the spiked shoulders, dark grey metals, and the red background (shying away from the pinker hue we were seeing before). The process on Thornmail really seemed to flow as soon as we knew it was Noxian, and I think the result really shows for it:

New Icon

Vampiric Scepter

Old Icon

Vampiric Scepter is another Noxian artifact, this time drawing from Vlad’s aesthetics to really drive home the vampire aspect inherent to the item. It took a few tries to get here—some of the early iterations involved a bat with scythe even—but the result is something a lot more mature than the older, goofier icon. We managed to preserve and abstract the face and scythe shapes, and give the metal a really unique gold treatment to match the rendering we see in Vlad’s splash. The final touch was making the magic feel like is was reaching out and sucking in energy from the space around it, drawing on the strength of others. The final icon is a lot more menacing than before:

New Icon

And that’s them! The Visual Designers here at Riot worked really hard to bring out the best in these icons, and I think it shows. Lots of thought, time, and energy were poured into these to make them the best icons we can make, and all the other teams who dog-piled them really helped us go that extra mile. A huge shout out to everybody involved!

I’ll be in this thread to address any comments or questions, so if you have anything to say just let me know and I’ll do my best to respond. It’s great to have these out, and we hope you all enjoy!

Abyssal Scepter: So basically, you turned it from a skull on a stick to a skull on a spine. Okay?
Fair point. I guess it's mostly that the old one lacked any real design and wasn't super interesting. The base was there, it just needed to be pushed. :P

I like it. The only one I don't fully like the change on is Vampiric Scepter. I would have liked to see more of the skull/face in there.
I tried to keep the implied face in there with the fangs and eyes, but keeping the skull ended up making the whole item feel super clunky. Kind of a shame, because in theory it's pretty cool, I just found it really difficult to execute on and still make look like the Vampiric stuff we see with Vlad. Thanks very much for the feedback though!

I think the new icons are brilliant! And I like the whole explaining the process behind each design change too!
Hope you get round to updating all the remaining ones soon so we don't have the clash of art styles...
Will you be updating summoner spell icons too?
I assure you, no icon will go untouched (unless it released more recently--Orb of Winter, Spectral Cowl, and Seeker's Armguard are already in-style, as are all the newer champs since Annie's rework). Look forward to them soon!

love the new icons.
how long does it take to redesign one? And how many people are involved?
Are there plans to change the other icons too?
It varies from icon to icon. Ability icons usually take about two days to bring to final polish, since we paint them at a much higher resolution than seen in game. Item icons can take a bit longer, because a big part of the icon is doing actual redesigns and running the concept by people to make sure the shape language and colors are working. The average time for these item icons was about 4 days apiece, though they were staggered a bit. The overall work time on these was about a month in between other icon projects, and lots of testing and reviews were done before we decided to put them in.

Also, keep your eyes peeled for more coming up in the future!

The only one I don't like is new Deathfire Grasp, and that's becaus the new icon is so vastly different from the old, not just in terms of the actual item itself but even in the colours used for it, that I can no longer identify it at a glance. All the other new icons I can give them the briefest of looks and still immediately know what it is (which is important in a game when you need to do your shopping as quickly as possible), but for Deathfire Grasp, I find myself constantly have to search and research the shop to find it because the icon is too drastically different. I'll get used to it eventually, and for new players coming into the game this will never be an issue, but I find the drastic visual change is definitely a huge annoyance that slows down my shopping at the moment.
This was one of the icons we were most worried about changing, but the old one had very little to work with. I tried a few variations keeping the pink and grey, and even had several iterations with the skull, but they were all mostly as confusing as the original was (which isn't terribly clear, though I know it's been around a while so it has become hugely familiar). In the end this one provided the clearest read, and we wanted to introduce the more dramatic changes slowly so that the overall transition would be less jarring. We have a few others planned to be reworked to about this level, but the number isn't very high and we hope to slot them in gradually over time.
Agreed. A month is INSANE. I could make more icons than this in a day. Bit more if I had to go through discussions and touch-ups, but definately not a MONTH.
As I said, "The overall work time on these was about a month in between other icon projects," which means we had a lot of things going on simultaneously. Keep in mind we're trying to hit all the new content too, guys. :P

RiotRadioblur , I was wondering , Are you in charge of updating champion's abilities icons?
Because , I had a really big doubt. Why didn't you change the ability icons for some champions when they got a Visual Upgrade? I know you did for Annie , Karma , Master Yi , Katarina. But why not Ashe , Nidalee , Soraka?(just examples)
Soraka , for example , really needs new icons to match her new look and all. Her old ability Icons just look.. really outdated and not really pretty.
I know this doesn't exactly belong to this thread , but I would really appreciate It If you could give me an answer.
I'm doing most of our icons internally right now, but the whole Visual Design department shares a lot of that responsibility so it doesn't really fall to any one of us in particular. I can say that updating relaunched champion ability icons was a movement I started when I got here, which was just before Annie was relaunched. We're looking at ways to get back and update all the old stuff, but there's a lot of new content rolling out all the time that eats up a significant amount of bandwidth. I don't want to promise anything, but I assure you it's not off our radar.

That being said, were these Icons worked on by the Vigil hires I read about? The style reminds me so much of the Darksiders art, and for me that's not a bad thing. I was a fan of Vigil long before I became a Riot fan and it makes me happy to see the style live on in another of my favorite games.
Actually all of these items were primarily spearheaded by me with some seriously huge assists from all the other departments. There are quite a few Vigil folks around the office, but none of them had much of a hand in this rework pass to my knowledge. That said, we might be seeing some down the road with some of them jumping in for support, just as long as the end result continues to feel distinctly "Riot."
Meanwhile, Nome implements Monflair Spellblade which gives resistances(As all swords do!) and has a ranged only tag (Because long-range-sword!) ^^
Don't get me wrong I'm not ranting, I'm just trying to sarcastically point to those inconsistencies. They really bug me - You can look at Thornmail and you'll know what it does:
It's armor, it gives resistances - check
It has thorns, it hurts you when you touch it - check (sorta)
Name and item-design match so well - Why would you break this principle just because you want to recycle an item there?
So, just a humble request: Please not only update the visuals, have an eye on those items which match/don't match their icon and adjust those accordingly.
That's actually a great point. Definitely something to think about going forward. Thanks for the feedback!

Updated Sivir 'Out Soon'

Originally Posted by Riot (View RedTracker Source)

She looks cool. She will be out soon. The skins are still being worked on but Sivr Pax is looking legit not much changed, just streamlined the design a bit for readability.
dis ansar is satisfying
lol, I know it was too short of an answer. I had to edit it and add a bit more.

But I will say this much, She's not the next VU.
Attachment 749610
That image made me lol for real. So a while back one of the Relaunch guys (can't remember who, might have been Qqroon or Ironstylus) explained about high risk reworks vs low risk reworks. Master yi is an example of a low risk rework. We knew what he was going to be, is original look and concept was cool, the execution just needed a bit of polish to get to the current standard. It was mostly just visual work. Reworks get more risky as more aspects like gameplay design and lore have to be involved. So we wanted to do a few low risk characters before we hit the really work intensive Relaunches (like Sion). Sivir was mostly visual work, but there are some story elements that the writing team have been working on to make her tie into the world a bit better, and make her a major player in the Leagueaverse. That being said, the reason she's not the next one out is simply because we started another one before her.
its twitch right
It's not twitch, but Twitch is coming along nicely. I am really digging him. I have the sculpt done and the team has approved moving him forward.

Also what does it mean to have a sculpt done? What steps are there in the process of giving a champion a VU?
The sculpting phase is when I translate the 2d drawing into a 3d sculpture like I did for master yi and trundle. https://vimeo.com/68653110
It's a phase where the design is getting locked down and streamlined for the actual game.
After sculpting it gets dropped down in resolution and we take the shadow info from the high res sculpture and use it as the starting point for the low res game model's texture.
doubt we can get any form of teaser for the new vu
and will it be in the next pbe set?
The team likes to announce these types of things very formally with lots of great support from all the different facets of Riot, from the web team, community,events, and marketing teams. If I were to leak images or teasers, that would be a waste of a lot of hard working peoples time.

Our goal is to have one rework done a month. Now having one done and having one released are two different things. Yi had been done for a few weeks, but SGU had priority, we didn't want to steal any even a little bit of it's thunder, also there were lots of platform techy stuff things that would have made the patch too big. So we held off until SGU was home free. So what I'm saying is, we try to be flexible.Just because a rework is done, doesn't mean it's time to release it. We try to time it with other stuff that can boost overall player engagement.

No hints!! If I even give you the slightest of tidbits you guys will figure it out! You are good like that...

Possible Removal of Improved Summoners in S4 Masteries

Originally Posted by Riot (View RedTracker Source)

... many of my friends who are pre-30 can't really work out what to do with their mastery pages, or how to use them to optimize their play.
Hope this helps.
P.S. thanks for coming out to EUW don't let the mob scare you, I for one am very glad to have you here helping us give input.
No problem. I've been hanging around the last few years on the EU forums (almost 2.5 years now eeek!) though I've been quiet recently due to time constraints both work related and non-work commitments.

As for helping your friends out who struggle optimizing masteries how useful would a recommended masteries button be? Basically after you select a champion you hit the recommended button in the masteries tab and it fills out the tree for you?
- Least favorite; This is a major complaint of mine, actually, about a minor (and, to a degree, insignificant, some would say) detail. There's a total of 10% CDR spread throughout two mastery trees. However, instead of an equal amount in both (5-5) you opted to go with 4-6. "Sorcery" annoys the living hell out of me, because you often end up with that terrible number, 39% cooldown reduction. And I detest it...
For Gods sake, swap over that 1 CDR point when we transition into S4, so we won't have to add one point in CDR via Glyphs, or find some other obscure solution to add that one missing point. It seems like an unnecessary distribution of CDR. Champions who value CDR will still spec into both and acquire the full 10%, supports will only lose out on 1%, but will still have a psychologically satisfying 5% to work with, and carries who value CDR won't have to make gimmicky builds just to reach the cap beyond the 39% CDR.
It's on my list. Current plan is to remove the Improved Summoners and shift it to be 5% and 5% in Offense/Utility. It'll take a little more shuffling and tuning beyond that but I agree its a concern just due to the extent it bothers some people.

Update on Skarner's Changes

Originally Posted by Riot (View RedTracker Source)

Hey guys, sorry for the long silence. Stuff is going really well for Skarner over here, and I may have more updates soon. We have done a bunch of playtests on the new kit and are getting lots of good feedback.

Coupling the attack speed with the Q has been super well received. It feels very natural and his ramping threat/brawling feels stronger than ever.

W speed increase has also been very effective in making him get into fights more consistently but not without some effort still.

E is a mixed bag, and we are still testing around with a few different versions of the spell. Some liked the on next hit version of the ability, some were lukewarm to it. I'll update if we come to any resolutions. I have seen a few good ideas on here for the E as well that I might try out.

The changes to R have been 100% awesometown. Everyone feels like it is more consistent and meeting the players expectations when they activate the ability.

I too would like some new Skarner skins... I'll see what we can do.

The Process of Reworking a Champion

Originally Posted by Riot (View RedTracker Source)

And I want to know around how much time is needed to rework a champ if BroDude can answer this, are you playing with the old texture and modifying it or are you changing everything like making a new champion and not even working with the old texture? The "ugly" champions really need to get a VU, it starts to look like a hybrid of a new and old game when you have a new champion and an old champion on the same lane, like in WOW where you meet a dwarf with a worgen or a panda! xD
On a true Rework/Reluanch- we throw away almost everything and recreate it with a fresh model,higher fidelity textures, better rigs, cleaner scripts(on the backend), updated vfx, new animations, beefier sounds, and a brand new voice over. It's partly why we need time to work on this stuff since we are making it from scratch. We try to make it match the quality and scope of a newly released champ.

Lyte on Player Behaviour and the Tribunal

Originally Posted by Riot (View RedTracker Source)

About leavers/AFK in ranked:
- Reduce LP loss for the team with leaver/AFK.
- Increased LP loss for the Leaver/AFK.
Ever thought about it? (Probably yes), Then tell me why it wouldn't work in conjuction with the leaver Buster system? Doesn't, with both of these solutions instored at the same time, allows players who had to struggle in an unfair matchup lose less and the offender being hurted a lot more? Thus not inclining someone to leave the game for the team and reduce LP loss?
Thank you, have a good day
These are interesting ideas that we've discussed in the past. The problem with reduced LP loss for the team is that it encourages the team to bully or harass someone to leave the game if they believe there's no longer a chance to win. In many ways, a feature like this would encourage some number of neutral players to 'go toxic' in frustrating games and vent on a player, hoping that they would leave.

Increased LP loss for Leavers/AFKs is less of a problem; however, in Ranked Modes, we don't see many intentional leavers or AFKs--most of the leavers in Ranked Modes are likely due to legitimate reasons such as emergencies or outages. Everyone in League of Legends has disconnected at least one game in their career--should we really be punishing these players even harsher?

It might be worthwhile to punish these players harsher if we reduce intentional leavers/AFKs by a certain percentage; but, it's something we'd have to model and calculate out and see if the change is worth it.
I think the notion that pre/post game chat logs only affecting the most toxic players is highly flawed.
A major factor here is popular belief.
A substantial amount of people won't report because they know full-well that pre/post-game chat logs aren't included.
And equally so, there are toxic players who will restrict their unsavory verbal abuse to the pre/post-game chat.

Correcting these problems would have the misunderstandings resolved immediately, resulting in very swift reductions in pre/post-game verbal abuse. (That includes Champion select arguments!)
This means the "X" your data suggests may very well be 5 to 10 times greater.
Additionally, if you want to talk about pragmatism, then I want to hear your thoughts on the effort-required-to-solve. Effort-required seems vastly differently here. (imo)
Solving chat logs, a purely technical project, is equal to A, while Champion Selection, an extensively psychological, technological, and UI-related project, is equal to 10(A).
Thinking pragmatically, it would make more sense to pursue the 5(X) for A instead of 10(X) for 10(A).
There's a slight nuance in our thinking.

The Tribunal is a harsh punishment system, aimed to reform (and if that fails, remove) players from the game. To us, this is a last resort system. Because the system was designed as a last resort, we only want to utilize it against the worst of the worst players because it really sucks to tell players they can't play a game they love. If at all possible, we'd like to reform these players, or shield their behaviors from other players while still allowing them to play the game.

A lot of players might not report pre- and post-game chat because they know it isn't included in the Tribunal; so, it's true that if we included pre- and post-game chat that reports would increase. However, reports increasing doesn't necessarily mean that we would tune the Tribunal to punish more players. We've been recording pre- and post-game chat logs for awhile and have run numerous analyses on them. In many cases, players that are toxic in pre- and post-game chat are also toxic in-game and are always punished by the Tribunal--it just takes a bit longer than the average toxic player. We've also observed that the number of players who are solely toxic in pre- and post-game chat is so low, that the incremental value of adding the chat logs just isn't worth it for now.

The key here is asking how many players are getting away with toxic behavior, by just being toxic in pre- and post-game chat? For now, the answer is too low for us to justify resources. To get back to your original question, we do consider "effort to solve" though. Effort to solve Champion Select is much greater, but the eventual value is much greater as well.
I just want to throw my experience behind Restricted Chat working with a quick example.
Yesterday I had a game with somebody approximately named "FVCK TH3 TR1BUNAL".
Great. I'm in the third of my promotional series having gone 1/1, and I get saddled in with this clown who has obviously been banned.
In champ select there's some light banter between folks, nothing out of the ordinary, but it makes the pit in my stomach grow.
We get in game, and he (as Lee Sin) says "I'm on restricted chat, so I'm going to use lots of pings."
Yeah, big surprise there bucko.
As the game progresses, thee's a couple moments where one of our lanes didn't follow up on his engage, and there's a line of charged chat between them, but it quickly cuts off, in my opinion due to inability to escalate the argument.
Lee Sin proceeds to play well, communicating with pings and we push through to win and get me promoted a division. After the game I compliment his Lee and he returns with how I was a great support who saved him more times than he could count.
I've been in piles of games where that first argument simply explodes into insults back and forth, ending with a demoralized team just waiting to take shots at each other and a wholly expected game loss. Restricted Chat very visibly salvaged this game for me.
I think the issue isn't playing being a toxic person. There are certainly folks out there who are just jerks through and through, but I believe most cases are just people not knowing when not to hit back. Not every insult needs retaliation, not every grievance should be escalated into a full scale argument. Restricted chat helps players censor themselves when don't have the judgment on their own.
This is what we were hoping with Restricted Chat Mode.

Thanks for your story.
To what extent do you see mathematicians having a role in the gaming industry beyond statistics, and are there any particular things you see as being valuable to study if one wanted to keep that option open? I'm starting my master's this fall, and will probably focus on mathematical analysis but I have plenty of room for extra courses on the side. If I don't go into academia, I could always go into the financial industry but would much prefer to avoid selling my soul.
Welcome to science since the 17th century.
In recent years, the industry has really branched outside just games and looked for experts in relevant fields. Game studios have been hiring mathematicians and statisticians, economists and psychologists, and even anthropologists and sociologists. It's a really exciting time to be in the games industry because in many ways, the data that you have access to in games is beyond anything in a lab. In fact, I truly believe that for decades to come, some of the most fascinating findings in many of these fields will be from games.

For the player behavior team, we've been using a lot of predictive modeling and machine learning for some of our systems. These techniques are only going to become more valuable with time in this industry, and by the time you graduate, the industry will be fully supportive and 'bought-in' to the idea of hiring outside expertise into games.
I think you're wording this poorly.
Do positive players win a higher percentage of games, or do they just play more games and therefore win more games?
I've noticed the warnings you receive are not based upon getting reported a lot over a series of games, but just getting reported by a large sum of players in a single game.
Like, getting reported by 5 players in one game rather than reported by 1 player every game for 5 games.
Why did you choose for that instead of the latter?
Because the stats average across the entire playerbase, usually factors like # of games played are normalized. So in the stat I mentioned, the positive players win more games (but play about the same number of games as the other two populations).

Behavioral Alerts do take into account both the # of reports per game, and the # of reports across games. We might not have optimized the tuning of the numbers yet though.
More like players who are positive (more sportsmanlike) just won more games and thus weren't angry.
Like your loading screen tip says about players who are more negative - they win 23% less games right ? Guess why they're more negative AND lose more games. Yes, one causes the other, but guess what? It's the reverse of what you claim.
When someone is throwing your match, of course you are more negative.
The first few times we ran these stats, they were just correlations. So you are right, we couldn't decipher the causal direction of the relationship and it could just be that players become toxic when they are in losing situations.

However, in the more recent versions of the analyses, we've attempted to control for these factors. For example, how does a positive player respond in a losing situation? How many reports do they receive on average? We then compare these baselines to the toxic players and see if we can figure out the exact direction of the relationship between winning and sportsmanlike behavior. With some of our more recent work, we're getting confident that being more sportsmanlike does win games. Teamwork OP right?
I'd love to hear your thoughts on the fact that negative reinforcement has been repeatedly shown to be not only ineffective, but also detrimental.
Any ideas on stopping the pointless and ultimately ineffective negative reinforcement, and using positive enforcement for perceived "good" behavior, instead?
We definitely would like to focus on positive reinforcement of sportsmanlike behavior; after all, if you make it easy for players to be sportsmanlike, they tend to be sportsmanlike.

However, negative reinforcement hasn't been shown to always be ineffective or detrimental. We've seen success with negative reinforcement in many areas--we just prefer to shift to positive reinforcement instead.
First of all, please do not remove or punish players playing 4-man premades.
I have a few friends I play with and end up in 4-mans a fair amount. But our lone teammate rarely knows it... In the majority of my premade games, only one person will really talk in chat to the solo player. So people who say "every 4-man premade is toxic"... may not even realize they're playing with a 4-man premade most of the time that they do so.
The only times people have asked if we were a premade, and these are rare:
1. Organized in champion select without talking. (One person chooses an ADC, then with 15s left, switches as another wants ADC - that sort of thing.)
2. Playing in a fashion that seems a bit more organized than is normally seen in solo queue. (Nothing fancy, just playing cohesively)
3. The random person is raging and I will defend the choices of whoever is being raged at. This is not 'we all gang up on the random person'. This is ME explaining why my teammates did X. Which I do when playing solo as well. And by defending a person - people assume you went into queue with them often times.
3b. If we had two or more ragers in a row, the next neutral-positive person may end up getting much adoration by the team and receive 4 honor for just being a decent guy. Even if they play poorly.
Anyhow, my question for Lyte...
Are you going to do a player behavior panel at PAX? If so, what day?
This would be on the top of my list of things to do if it's on Sunday or Monday (only two days I'll be going).
As far as I know, we won't be having a player behavior panel at PAX Prime this year; however, I know a few Rioters might participate on other panels (with other industry veterans) that might be related to player behavior.
Hopefully this isn't too off-topic, but I'm really curious to hear team PB&J's opinion on this.
Recently I saw an article that XBox live was going to a 'Prisoner's Island' type system, where the worst of the worst would all play with each other and leave everyone else alone.
My first reaction was "Hey! Thanks to Lyte and Davin I know what that is, and why that would suck to have in LoL!"
Do you think that's a good solution for XBL, or do you think it'll run into the same problems League would have?
It's difficult to say--I don't have access to their data or have a full understanding of their ecosystem. We're just glad to see more studios challenge the player behavior problem, and we hope they share some of their lessons so we can all learn about the most effective ways to solve player behavior in the future.
There is a folk theory that the "V" tiers of ranked are highly toxic, in part because they are filled with people who are safe from moving down.
Does your data support this? Anything on the horizon to address it?
Also... hurry up on the champion select issues!!!
RiotSocrates has done some research in this space, and he's the one to ask about any changes to Leagues.

We're working our fastest on Champion Select
Lyte, do certain champions cause more reports in a game. Example: teemo or an unpopular low win rate champ?
It depends how you normalize the data. In fact, champions like Ashe get the most reports and the most honors but it's mainly because Ashe is one of the most-played champions in the game.

A long, long time ago, pre-stealth rework Eve received the most reports in the game; however, the interesting twist was that Eve champions were not punished more often in the Tribunal than any other champion. It was interesting to us because players were acknowledging the context of Eve, and pardoning players a bit more frequently to off-set the increased number of reports that Eve got back then.
BTW has nobody mentioned smurfing? Specifically the huge amount of smurfs in ranked soloqueue. I've watched friends play a lot and it isn't uncommon at all for them to be matched with or against smurfs several times in a row. It's pretty easy to tell who is smurfing or duo boosting (you can check the duo boosting and the obvious smurfs just pick a stompy champion and clearly outplay everyone in the game). I wonder what the benefit of this is (considering Riot has encouraged this there should be a benefit, right?), as the downsides are fairly obvious.
Smurfing isn't officially endorsed by Rioters.

It's something we are actively doing research on and trying to address. We agree with some reasons to smurf--such as to make a new account to play with a brand new friend; however, we'd like to make it easier to do that instead of encourage players to smurf.
What about between different nonranked game modes? It definitely seems like the social landscape between different modes is different. For example people don't seem to honour as much in ARAM as in Dominion, yet I experience less toxicity in both when compared to normal and ranked SR games.
Also, I've been wondering about the Optimus experiment. I thought the point of the experiment was to find the best tips for priming players into a less toxic mindset, and that "fluff" tips like naut's dance or draven's axes tips were simple controls. Yet, when the new batch of tips were brought in, most of them were "fluff" tips that would seemingly have no effect on toxicity.
Is the experiment still ongoing? Or do those "fluff" tips have a value of their own, like providing a backdrop and get people to look at that space for when true priming tips come up?
The Optimus experiment was to identify the most effective tips to shape player behavior positively; however, there's still room for 'flavor' in a video game, especially if the effects of the flavor are neutral in the behavioral space.

However, we're working on sharing more research from Optimus in the future and do have some interesting insights about the 'flavor' tips.
Lyte, i have judged a few cases on the tribunal and was wondering who determines the severity of the punishments like warning, timeban, permaban.
Is that completely automated or do certain things proc some riot employees to weigh in on the punishment?
Punishments are usually based on frequency of toxic behavior; for example, someone who is toxic in a majority of their games and has visited the Tribunal multiple times is more likely to be up for a longer timeban.

However, because there is a review process with Riot Player Support, sometimes cases get escalated to higher punishments based on the severity of the toxic behavior as well.
Hey Lyte, as a player who has recently been banned for toxic behavior and leaving games, I can say that the restricted chat does help, it makes what you type more meaningful, and if you care about winning you aren't going to waste it on calling out your teammate.
I'm not sure why I've been such a ******bag as of late, but I'm sorry, I'll try to clean it up.
Thanks for your feedback, I'm happy to hear that Restricted Chat Mode helped you focus on the game--and winning!

We've heard a lot of feedback that with Restricted Chat Mode on, players are winning more games simply because they are focused on the game and not on negative chat.
Is there a transcript available for this?
I don't believe so, but there's a nice written summary of the talk here:

Lyte - is there any way to add some type of icon or badge to players who have restricted chat? I communicate a lot during games and if the player is not responding, I get extremely frustrated and quite often will report for refusing to communicate with the team. Also, does restricted chat prevent champ select chat or just in-game? I've never seen anyone in champ select say "Hey guys - i'm restricted to chat in game, so if I don't respond, that's why".
We were very careful not to give indicators or badges that a player is in Restricted Chat Mode; this is because we don't want to brand players as 'toxic' and set them up for failure before the match has even started.

Some players still choose to use their messages to inform their teammates that they are in Restricted Chat Mode; however, many players simply choose to hide their status and be sportsmanlike members of the team and we're OK with that. If we labeled these players or otherwise branded them, they might be harassed even if they were going to be sportsmanlike in that particular game.
I guess "recent" is relative being that this thread's last Red post was over 3 1/2 months ago.
I know you don't have a solid idea of what is going to happen, or maybe it's a date that you don't have. But I feel like maybe we should have an idea of where you're likely to go with this, even if we don't get a date. Can you give us any update? Is role selection the direction we're heading or maybe preference tags? I know I'd like to know your thoughts, even if not finalized.
We spent quite a few weeks on research, and have begun pouring our development resources into our Champion Select solution. We're actually having a few meetings over the near future to see what we can share with players.

I'm super excited to maybe even talk about some of the research and why it inspired some parts of our solution.
Lyte, currently I'm taking a break from League because I don't like the way I've been responding to it lately, but I have a question that I hope you can shed some light on. I'm a plat player, and one thing I've noticed at this level is that nobody seems to honor. For example, when I play with my low level friends, everyone seems to honor. Teammates will honor you for just plain playing well, or whatever else. In platinum, nobody uses this feature at all. I'm not trying to state that something is wrong because I'm a great player who deserves honor etc, far from that.
Rather I don't see the system being used in situations that if they had taken place in a low level game, would have resulted in honor being awarded by another player. Everyone at this point just seems to not care anymore. I miss the moment of reward I feel after a game where everyone acted appropriately and we pat each other on the back with honor. That seldom, if ever, happens. And it's not as if there aren't games with good people at this level, just that the feature isn't used.
Have you noticed a difference in the amount the system is used at different levels of play? Are you guys planning anything to encourage people to use the system? I know that's a tricky subject that would be hard to prevent abuse, but I don't see any other way to get players at this point to use it. Or is my experience just abnormal and the people I play with just don't happen to use honor and/or people just don't feel I deserve it.
We've definitely noticed that Honor distributions vary greatly across different game modes and skill levels. We have a few ideas in mind that we want to research and try for Honor, after we're done with Champion Select.
Since the longer a player plays the less likely toxic players are to stay in the system since the tribunal will weed them out, and since you guys made the change recently to have people with similar numbers of wins matched against each other, doesn't that mean that people who've played a lot of games are less likely to experience toxicity than people with less games played? People always talk about how the lower levels are full of toxic players banned from their mains, but isn't the opposite also the case?
This is a really cool insight, and is partially true. Very, very veteran players should see fewer extremely toxic players, which is the demographic that the Tribunal targets; however, our research suggests that the majority of toxicity comes from the neutral and sportsmanlike population potentially having outbursts due to bad days or other RL incidents. Veteran players will probably still see the same amount of toxicity from these 'types' of incidences.
Anything going to be done to stop afking in game? Does reporting them get anything done if only 1 or 2 people do it?
LeaverBuster is the system that handles Leavers/AFKs, and yes, reporting them is effective even if only 1-2 players do it.

The player behavior team hasn't done a pass on the LeaverBuster system in awhile, it's something I'll look into after Champion Select.
Done Editing: I have extensively gone through this thread and here are my replies for everything that interested me :P
There's a lot of material here (and in the link you posted), so I just wanted to quickly say that I don't have time right now to give the thread justice.

I'll loop back when I have more time! Thank you for the feedback on the systems.
This isn't super related to the behavioral model. You guys give great talks and it's always interesting to see what you're up to. However, none of the videos have transcripts available or closed captioning. Are there plans to incorporate this kind of thing for the players in the community who may be deaf or hard of hearing?
We'll talk to some of the organizers at future talks to see if this is possible.

For now, I know Gamasutra had a nice summary of the talk here:
what are the odds of unbanning permabanned accounts with permanent chat restrictions if their only offense was verbal abuse?
Our policy is that permanent bans are permanent. We will not go back retroactively and apply different punishments or unban accounts.
What can you tell us about the MMR system. How sophisticated is it? A rioter mentioned that the current system had so many deviations from the Elo system that it couldn't be called that anymore... but how can this be when the only metrics you are looking at are ostensibly wins/losses? It seems to me you have to be tracking other stats, otherwise there's no way the system could be this accurate. If playing with a duo queue I'm surprised at how close the games turn out to be, and when playing on a smurf it only takes a game or two to be placed against other smurfs.
The system is pretty sophisticated because it's continuously evolving as we collect new data. Over time, we've added custom features to the algorithm based on unique situations in League of Legends such as how do we handle duo queues? Premade 5s? How do we handle situations where players are queued with a friend who has a large skill differential?

In many of these cases we are taking into account a lot more variables than simply Wins/Losses.
This just in: most normal league of legends games are already "prisoners island". The real reason you don't want to add low priority to LoL is because of the revenue you generate from permabans.
When we permanently ban players, we don't want their time or their money. We want them to spend their time elsewhere. I can't be more clear than that.
On this note then, given how possibly delayed the system is; is it currently possible for certain players to receive these notifications long after they actually passed a threshold where said person has been sent to the tribunal?
It's possible, but extremely rare.
So why not just IP ban those punished? The toxicity is a direct relation to being free to play. There are ways to circumvent IP bans, sure, but leaving the comfort of your home and going to a friend's house or Starbucks isn't exactly convenient. Secondly, if you get IP banned at your friend's, there will be some real world punishment from his friend. (If the friend wants to get his IP unbanned investigation measures should be fairly simple).
IP Bans aren't effective for a variety of reasons. For example, there was a story a few years back of an IP Ban that was handed out and it turns out it was at a college dorm--we received something like 200 tickets from players that were affected but didn't deserve it.
Hey Lyte,
I have a few questions for you about Honor and positive reinforcement initiatives in general:
Do you feel like Honor is accurate enough to use as the basis for other incentives? I'm currently working on an analysis of the system where I'm recording as much data as I can that's relevant to see what correlations there are referring to the accuracy of the system. So far it seems like there's almost no correlation between Crest ownership and good personality, though obviously my sample size/testing methods are extremely limited. You're welcome to check it out here if you'd like: http://forums.na.leagueoflegends.com....php?t=3722962
Do you feel like Honor is accomplishing what it was originally set out to? A lot of the feedback I've been getting so far seems to be saying that people either don't care about Honor due to how difficult it is to keep a Crest or feel slighted because of inaccuracy and so just refuse to use it on principle.
Are any other systems in the works for positive reinforcement, or is the goal to be using Honor as the basis for everything?
Depending on which Crest you observe, there's a correlation between Crest holders and games that have 0 reports--that was interesting to us.

Honor needs some work, and we've already done some research into the space and how we want to polish and upgrade the system. However, it's currently being used a healthy amount on all servers and had a very positive effect on player behavior in League (while also allowing us more resolution to analyze player behavior because we only had the 'negative' spectrum with reports before). We'd like to improve the correlation between Crest holders and sportsmanlike personalities, and also implement a few features into Honor to encourage players to aspire and strive for Crests.

To us, positive reinforcement is more of a philosophy that guides our system design. You'll definitely see more systems use concepts of positive reinforcement in the future.
I'm sorry, i havn't read the whole topic but what about some free runes (except quintecences) for players who earned a honor ribbon ?
It's an interesting idea, and I do want to hear more about what sportsmanlike players would find valuable to their game experiences or their accounts (beyond the obvious like "just give us free skins everyday!")
Dear Lyte, will your team present a new infographic soon about the Tribunal again?
Oh and over at the TBI forums we had an interesting thread about posting guidelines: http://forums.na.leagueoflegends.com....php?t=3732940 made by Dreyvan I would suggest this to be stickied? =3
Also over at the other thread, the forum regulars there wondered if the sticky http://forums.na.leagueoflegends.com....php?t=3598477 title could be changed to "Tribunal Justicar Handbook"
Changed the title of the thread to "Tribunal Justicar Handbook."
Honestly my issue is whenever I bring a friend to the game who's pretty green, if I pair up with him we'll be put against teams who could completely destroy him. So then my team complains to him and he gets totally discouraged.
Have you tried bringing your friend into a few Co-op vs Bot games first? Our data suggests that trying that for a few games with new friends is a much better experience.
Does/did your team ever disagree on the direction of player behavior? How did that argument get resolved?
There was a time we made Player Behavior face off in a Normal Summoner's Rift match, and my team won in a landslide.

Even though the above did happen, we have a smart team that knows it's about give and take. We generally have long discussions and meetings where we pour over the research and determine the best course of action for the team--surprisingly, we haven't had many disagreements on what the team should do.
Isn't that what checks and balances is for?? You say that reduced LP loss would cause some players to harass others to try to get them to leave. That might be true but if you truly believe in the tribunal then that's not a problem because harassing to prevent LP losses will inevitably get you banned. One system keeps the other in check and from being abused.
If you worry about corner cases, you will never be able to implement any idea. In 100% of the games where someone leaves, 4 people who didn't do anything wrong get punished in some way. Even if they win the match, they still made it harder to win and wasted their time. If the concern the leaver had is a legitimate emergency, then they would gladly take an additional 10-15 lp loss in exchange for the other 4 people they let down not losing full LP.
You can create rules to it to prevent abuse also. Most of the time people experience 4v5 matchups its from the very start. So make a set of rules like: When the game is over (from /ff or nexus falling) if the losing team had less than 5 players for OVER 50% of the game, they lose XX% less LP. If the game is going badly, and someone forces someone else to leave they would likely still lose 100% LP and get reported to the tribunal.
My opinion.. rules are rules. "I robbed the bank because a little girl needed a important medical operation." Sure, it's an important reason, but rules and rules and actions should still have consequences. Just because someone has an emergency doesn't change the fact that you just ruined 20-45 mins for 4 other people that were counting on you.
It's not about believing in the Tribunal system.

What was the original goal of reduced LP loss for teammates? Was it to reduce Leavers/AFKs in games, or reduce the impact on teammates when there was a leaver or AFK player?

In both cases, reducing LP losses for teammates doesn't actually solve the problem. It creates more. Regardless of whether the Tribunal will handle the new abuse that the feature would introduce, the feature doesn't actually reduce Leavers/AFKs (and you could argue it would increase the number of them if teams start harassing each other to get someone to leave) and the feature doesn't actually reduce the impact on teammates because now teammates are not dealing with Leavers/AFKs--they are creating them through toxic behavior.
This is more a request that I'm not certain is able to be addressed, but I'd like to hear your thoughts on it.
I feel that everything behavior based is very hidden/secretive. I've been playing over a year, I've held every Honor Ribbon at one point (green showing now), I've never been to the tribunal, and I've only gotten 1 Behavioral Warning (Which freaked me out a ton by the way).
I still have no idea what the color of the ribbons mean and which is the most rare. I think green is a combination of a few? I think Yellow or Blue are the most rare now? I don't know if I have those and only green is showing up?
I also worry about the "if you report too much, your reports won't have as much weight" mentality. I don't know if I was just raised in a very strict and respectful family, but if someone is even harassing someone for feeding top or picking an abnormal pick though they're still trying their best, I will basically report the harasser. I never see ANY value in harassing someone (though of course we all have our bad days.) My mind set is that if I report them, and they're just having a bad day, then it won't go anywhere. But if I report them for something and they do it all the time, then I will be helping the cause. Does that mean that my reports have less weight to them? (I'd say on average I report 1.. maybe 2 people per game. On average)
To reiterate my point, I just don't know anything about this stuff. Other than this forum post (which I'm reading every single post in detail because this all intrigues me), I feel there really isn't much of a FAQ or really much information about any of the perimeters of the systems you put in place.
Thanks for all you do,
Thanks for your feedback Leetnanas. I agree, we could do a better job explaining the systems--it's just something we have to constantly work on and improve.

We've been trying to put more FAQs on the forums such as the Tribunal FAQ and "Understanding Tribunal" thread, and we're trying to do the same for Matchmaking and Honor. If only I invented a machine that gave me infinite time...

A Legendary Weapon Comes to Life

Originally Posted by Riot (View RedTracker Source)


You've been hounding the master weaponsmiths at Man at Arms for a while now to recreate an iconic weapon from League of Legends. They've heard your requests and asked us to choose which weapon to build. We thought it'd be better to ask you instead, so we're opening up the polls so you can decide which of Valoran's mighty weapons will be forged in our world.

Some of the most realistic and mind-blowing recreations of video game, TV and movie weaponry, including Link's Master Sword, Sora's Keyblade, Cloud's buster sword and even Wolverine's claws, come from these talented craftsmen. You can check out how each weapon is made by watching their popular YouTube series.

Along with the actual weapon, Man at Arms will be producing an episode of their series dedicated to bringing the weapon you choose into reality.

Let us know which weapon you'd like to see crafted by the Man at Arms team in the poll above this post, voting closes this Friday!

Champion Skin Sale: 8.06 - 8.09

Originally Posted by Riot (View RedTracker Source)

Give your champions a new look with these skins, on sale for a limited time:

  • Battlecast Urgot 675 RP




Add these champions to your roster, on sale for a limited time:





  • To post a comment, please or register a new account.
Posts Quoted:
Clear All Quotes