Interview with Riot's Playtest Lead, Adam "Riot Afic" Cohen.
dignitas Novalas catches up with Riot Afic on his role as the playtest lead at Riot Games.
League of Legends, a game that tops the list as the most popular game in the world, has a team working on it at Riot Games. Working on the game are many developers and designers, creating the champions, items, and mechanics that are implemented in the game. Behind them, however, is a team of high elo Rioters referred to as the Playtest team. This team has a heavy focus on providing feedback to designers to test out design goals & balancing. Dignitas Novalas catches up with the lead on the team, Adam "Riot Afic" Cohen. A League of Legends player since the game's inception, Riot Afic knows the game inside and out. In this interview, he discusses the purpose of the playtest team, his opinions on upcoming reworks, and the changes in the preseason.
Adam "Riot Afic" Cohen
Hi Adam, for those who don't know you can you please describe your role?
Afic: My role is to lead the playtest team and help give us direction for what our team’s goal is. What our team’s supposed to do is provide feedback to designers that either challenges or confirms their assumptions. So a designer will put in a change, it’ll have goals, our job is to make sure those goals are or are not being met and then being able to prove it. It’s a lot of qualitative work and that’s basically our job. We also help with balance.
About once a year or so, you post on Reddit recruiting additional members to your team and the main criteria seems to be having relatively high elo in League of Legends. What other requirements would you suggest for potential applicants?
Afic: Elo is definitely the easiest way to “fast fail” people. If people are below the elo, it’s just not going to work in what we’re looking for so that’s a must. After that, communication skills, working in a team environment, and analytical thinking are the next three qualities to be able to exemplify.
Who are some notable people on your team right now?
Afic: Ya know, we’ve got them high elo players. We’ve got some former pros. People we got from the last Reddit thread; we have Penguin who was a high tier Anivia player, we have gleebglarbu, we got ROBERTxLEE, a few other people. But those are the big ticket names.
How does your team tie into and communicate with champion designers and system designers? What’s the value of your feedback?
Afic: It’s going to depend from designer to designer on how much they go out of their way to interact with us. We make sure that anything that does get put into League of Legends, we either agree with or agree to disagree. I would say that in terms of mechanics and whether or not something is healthy for the game, we have some input on that but in terms of overall balance, we have a lot more input on that. Our playtest, the games that we run, really help demonstrate the long term power of these changes. A lot of that information comes from our games. They are the most balanced. Designers, they span from low silver to high diamond. You put a bunch of those people in a game together; it’s one of them NA fiestas is how I would describe it. You can’t get any meaningful balance data out of that. Our games are very useful in terms of determining how balanced something is.
In your games, since they’re aimed to be more balanced, do you aim to demonstrate champion mechanic balance or do you guys look to find specific bugs? Or is it mainly just for balance purposes and do you rely on another team for bug testing?
Afic: We are mainly for balance and “feel” related things. A lot of qualitative feedback. The bug stuff is handled separately. That’s generally not something we do. If we notice something, we bring it up and it gets fixed but we don’t actively look for them.
Talk to me about your experiences with the upcoming Assassin Rework. What are your opinions on the overall Assassin Rework?
Afic: My opinions on the overall assassin rework, oh boy. I think they add a lot of fun things to the class at the cost of removing the degenerate patterns. So by that, what I mean is… Well, let’s use Rengar as an example. Rengar mains, don’t hate me for saying this Rengar mains, they are in love with a degenerate play pattern. They like the possibility of repeatedly deleting someone from the game without the opponent having any opportunity to do anything about that. While that may be fun for you, I would argue that it is disproportionately un-fun for the other person. So a lot of what the assassin rework is trying to do is trade that power to burst someone down in an unreasonable small window to other strengths where you have an opportunity to not get bursted down without looking at the screen for more than a second. And now, they have other things to get excited about to make up for it. That’s the basic theory of the rework.
I am concerned that the assassins in general are going to be weak. I think the goals, they make sense. They’re in the right place. I think the execution might struggle a bit. One of the problems and reasons why is because when we tend to make champion changes, players fail to adapt. What that means to me is that people are going to see this new Rengar and they’re going to try and play it like it’s old Rengar. They’re going to try and play this champion and try to one-shot someone and realize they can’t, lose the game because of it, and think the champion is too weak.
A good example where this occurred is Swain in the mage update. Swain’s core build path changed from maxing E first to maxing Q first. The win rate difference between players who maxed E first and maxing Q first was 13%, meaning if you maxed E first, let’s say you had a 40% win rate. That’s incorrect. Swain players who maxed Q first had a 53% win rate. That type of disparity is actually unheard of and there’s way to teach people, well there’s no good way that we’ve found to teach people, a large majority of players that “Hey! You’re doing something that’s actively wrong! Don’t do it! Do this instead!”
The Assassins might be victim to that. So Katarina for example, you might try to press all her buttons but she takes more finesse now. Talon, he lost some damage for his ability to roam faster but people are still going to try to play him as if he has that same damage. LeBlanc, they’re going to try to burst someone out. You actually have to stay in combat for a few seconds to access that burst now. We’re basically put into a position where we have to predict how poorly people are going to adapt to these changes. That’s a tough spot. So while long-term, these champions might be balanced, their short-term win rates might suffer negatively.
Ivern’s another example of this. His win rate right now is around 40%. His experienced win rate is well above 50%. My personal win rate is above 60% because I know how to play the guy. I know how to do the unfair things he can do. People don’t see that though. People try playing him like a normal jungler. People try duelling other junglers and, guess what, you’re going to lose. That’s not Ivern’s strengths. You’re trying to teach people to do new things with something that they could previously get away with. If they try to get away with it, things are going to react negatively. It’s going to take awhile to catch on.
Do you think that the Assassins will have significantly different build paths? e.g. Ekko being designed as a utility Assassin but being built tank?
Afic: There’s a good chance. It’s something we’ve brought up and identified awhile ago. We’ve taken some steps to mitigate that but it’s still possible. So going back to the theory, the theory being; the goal is to decrease the window of burst or well, make the window of burst over a longer window. So it takes two seconds to do all your damage rather than 0.5s. If you need to do your damage over a longer window, it is quite feasible, and we tried theorycrafting this a bit, that the optimal strategy will be to build a little tankier so you can survive for those full two seconds to really access your full damage. Things like RoA + Rylais might be ideal. I don’t expect the full tank builds. I think the AP ratios are too important and the bases are a little too low. And the repeatable crowd control isn’t there which is what Ekko had. But yeah, that’s a definite concern. I would consider the Assassin rework a failure if the optimal build is full tank. Except for Rengar. He’s shifting out of the Assassin role.
Art by Alexandra Vo
Do you think he’s going to be more of a bruiser or a tank, in regards to Rengar?
Afic: Bruiser-y/Diver. He’ll still want some damage.
Do you think a Titanic Hydra + Tank build would be something good on him?
Afic: It’s certainly possible. The classic “Heavy Cat” build of old might be ideal. He really likes cooldown reduction, even moreso than he does on live. That’s because he doesn’t need as many stacks. You get your empowered abilities more frequently than you previously would with similar cooldown reduction.
But it doesn’t quite have the same effect as live, right?
Afic: It sounds like you haven’t experimented with crowd control immunity long enough! When you’re tanky, sticky, you do consistent damage, AND you’re CC immune, that’s basically Olaf… except on a basic ability.
Are there any interesting iterations of the reworks that ended up getting scrapped?
Afic: Oh yeah. Lots of them. I think the champion that went through the most was Katarina. We tried so many things. There was a moment where her shunpo, it was super telegraphed and had delayed damage, and the tradeoff for that slowness was she was immune for a half second. She was like “Hey, I’m going to appear here! By the way, I’m immune until I do.” The goal was to prevent seeing her jump all over the place and not being able to track the champion. You could track her but didn’t matter because she was basically Fizz. I personally, as an AD player, don’t want another Fizz in the game. Call me old fashioned. But yeah, that was pretty out there.
LeBlanc, her ult casts, instead of creating a clone, used to stealth her. It was like a Wukong clone. It would leave an image of her and she would be invisible and cast the ability. That was, well, to describe it in a word, frustrating.
Of all the assassin reworks, which one do you think will have the most success in solo queue - whether it be through the champion fulfilling its goals the best or being too difficult to play against at launch?
Afic: I think Katarina is going to have the most success in solo queue. The main reason is that so much of her damage is going to be based on her being able to pick up a dagger and slash people with it. It’s going to take people awhile to get used to playing against that. In our experience, it’s fairly easy to avoid the daggers. They bounce relatively predictably. If you see the circle on the ground, you stay away from it. Good job. Katarina just lost a lot of her damage.
However, we’ve also spectated other designers games and other teams play who are lower elo and… God damn, if those circles on the ground aren’t just like magnets to people sometimes… I can’t tell you how many times I’ve watched people at Riot just ignore the circle. Not even walk near it, just ignore it. Then Katarina jumps to it, gets a crap ton of damage. When that champion has easy access to reliable damage, she does scary things.
Let’s move onto preseason changes. The jungles getting changed. I like the idea of pushing players away from a boring, 2-dimensional clear. Also, trying to force players to go for more “off the top of their head” action. From your perspective, do you think this is a necessary change? Do you think that the jungle has been stagnating?
Afic: To answer your question, I actually don’t think the jungle has been necessarily stagnating. I think there’s still - well, the example I’m going to use is C9 vs TSM in the finals. Jensen was playing Taliyah. He shoved the wave at level 1 and then he went to invade enemy raptors. What TSM did instead is, they assumed he would do this because he was becoming predictable. They changed the starting path and they ended up first blooding Jensen when he went for the raptors. They started at red instead of starting at Krugs. That, to me, shows that even at the highest level that the jungle is still evolving and play patterns are adapting. I think that for some people, people do the same thing every game. I’m not convinced that these changes address that issue. If people find something that works, they’ll still do the same thing. Maybe they will be more capable of changing up their patterns or changing what they do. But if they’re going to do the same thing, they’re going to do the same thing.
The biggest difference here is the increased spawn times. That’ll have a few interesting implications. One, I think carry junglers will have slightly less access to gold. Although we’ve put a lot of work into making it even-ish. My gut does tell me that it’s going to be slightly less, however. You’re going to have more time to do things that aren’t just killing camps so you’ll be able to gank more. You’ll be able to get deep vision and all that stuff. Invading and stealing camps will be more impactful. If there are fewer pieces of the cake and each piece is bigger, if you take one, it’ll have a bigger impact.
What about the removal of jungle buffs? (Smite buffs)
Afic: I have seen some of the data and, as someone who has played jungle, I actually really enjoy the decision making that goes into it. Between sometimes aggressively smiting wolves, like - that was actually a pretty cool play. Unfortunately, the majority of players did not catch on that. It is a pretty complex mechanic with minimal pay-off in the grand scheme of things. League of Legends isn’t balanced for any sole group of players. Regardless of how much decision making and optimization there was, the perception of it was minimal at best. That to me is a problem. Although it sucks from my point of view, I get it. For the millions of people who play the game, it’s going to free up space for things that are more interesting, obvious decisions for people to make.
Let’s talk plants. How do you personally feel about them?
Afic: It’s interesting. I know the first bit of feedback we had was that it seemed a little gimmicky. I don’t think there’s any sidestepping that issue. But part of what they were initially doing was exploring options, seeing what types of systems they could add, what impact they would have, and looking for better systems.
With plants, we went through a bunch of iterations and honestly… the games were just more fun with it. Maybe it’s because we weren’t taking them as seriously because we were playtesting and elo wasn’t on the line but those plants create some high moments. Some of the most entertaining plays I’ve had involved running; like I’m playing mid lane, the enemy jungler comes behind me, I run into the enemy jungle because that’s the only way I can stay out of range, then they chase me into their own jungle, and then I blast cone over the baron wall. Some of the play emergent gameplay could be perceived as random but I knew exactly what I was doing and it felt great. Gimmicky aside, I personally have a lot more fun with them in the game.
So you would say it just adds another dynamic to the jungle?
Afic: Yeah and it’s not super common or anything. It’s not game warping. But when stuff works, it feels real good.
What do you think of the upcoming item changes and additions? Do you think that they’re going to be beneficial to the vast majority of players?
Afic: The quick answer to that question is, yes. I do think that these items are going to be beneficial to the game. Going into more detail, first I want to outline the reasoning behind them. You can watch any time Meteos has talked about League of Legends. In the history of forever. He’s going to complain about Aegis. That item has been must buy on someone on your team for WAY too long. The aura on Aegis has been removed. That is no longer a thing. You should no longer feel that, if you need to survive LeBlanc, someone on your team 100% has to buy an Aegis. That mentality should be no more.
With that being said, that leaves a big void in terms of support itemization. When every support in the game was forced into buying this item and is no longer forced to, there’s a lot of flexibility. So the purpose of these items is to give them AN option. Knight’s Vow is one of the options for the tankier guys, presumably. Redemption is supposed to be the more “enchanter”, “healer”, traditional support character item.
Starting with Knight’s Vow, it does what it’s supposed to do probably too well when used optimally. It’s a tough one to talk about. If you’re putting damage resists on an ally, that’s going to be difficult to perceive. In testing, we did some things Kog’Maw, well we tried to abuse it. There’s a good chance that it’s going to be the next OP thing but it CANNOT be worse than Aegis. No matter how you slice it, this item will not be for everyone. The stats are too inconsistent for everyone to want it. The goal is to create more options for supports so you don’t feel pigeonholed into sightstone, boots, Aegis every game.
The other item is Redemption. That item is a lot of fun. If you’re playing any Janna, Nami, Soraka ish champion, this item’s a lot of fun. Taric loves this item. It’s optimal use will realistically just be for the out of combat heal which is a little sad because it has so much more potential but it’s an option I could definitely see myself getting in some non-trivial percentage of games on traditional supports. It will also help fill the locket void.
Edge of Night, that one went through a bunch of iterations. It does what it’s supposed to do (with the spellshield active). It can feel a little unfair at times depending on certain matches. First example that comes to my mind is Singed. Singed has one way to break spellshields. You can pretty easily force start a fight and not let him do that. The main thing is, assassin's know when they’re going to go in so they use this item pre emptively. Say a Zed uses this right before he goes in. The Ahri that’s waiting to use her charm behind herself when the Zed comes out of ult? That’s no longer going to work. It’s a nice buffer against certain champions with crowd control. It makes it harder for the enemies to reliably peel you. It’s a way of helping assassin's be slightly more consistent in accessing their damage.
Rylais. I think that item breaks champions. Certain characters that are supposed to have no consistent crowd control can get it in an extremely stat efficient way. Brand, Malzahar? Those champions without reliable slows look very different. When Brand can land his stun reliably, which by the way he can do on a 40% slowed target, his damage output increases like woah. If you’re not missing your stun, that means you’re reliably proccing your passive, bombing people. It breaks some fundamental strengths and weaknesses of champions. I firmly believe that in some capacity, it needs to get nerfed.
What do you think about the changes to stealth? That is, Camouflage and Invisibility. Additionally, what do you think about the implementation of Control Wards (replacing Pink Wards)?
Afic: Long term, it’s going to make the game easier to balance. How do you balance a champion like Akali? She can be REALLY, REALLY, REALLY strong. By the way, if you buy a pink ward, she’s useless. I think things like that are unrealistic to expect to be able to balance around several elos. This is a type of systemic change that should help normalize the difference between high elo and low elo. You don’t want to say something like “Buy a pink ward! Git gud.” You’re either not allowed to play this champion at a high level in a meaningful way or, at a low level, people aren’t going to be good so should that ruin everyone’s experience? That’s the type of question we have to find the answer to and a change like this will help and be the best of both worlds in some capacity.
Let’s talk about my two favorite champions in League of Legends; Ryze and Nidalee. But let’s start with Ryze first. Ryze has gone through multiple reworks and in his current state is once again being picked in competitive. A lot of high elo players have been making Ryze seem very, very powerful once again. What do you think the solution to Ryze is in his current state? Do you think that he’s an overpowered champion? Is it possible to tune him without another rework?
Afic: I think that this version of Ryze may not be tuneable but it is certainly more tuneable than the previous version of Ryze. The difference between the previous version of Ryze versus this iteration played optimally… Yeah uh, when Faker played the previous version of Ryze, he did unfair things. If you knew the correct combination in which to press your buttons, you could 100 to 0 someone at level 3. That is inherently unfair. It happened all the time. Turns out a lot of people, as easy as it seemed, could not pull off the combos. That made balancing Ryze a nightmare. With current Ryze, which Faker is still dominating with, that gap is much smaller. When current Ryze is played optimally, yes he’s good. When current Ryze is played suboptimally, he’s not as good but he’s still good. The gap is definitely much closer in that sense. Fortunately, the other thing you can do with Ryze, is that since a lot of his output can be team orientated cause of his ult, you can more meaningfully nerf his damage or make his damage more consistent. Then you can push his teamfight power or competitive power towards his ult which I would argue is a more interesting style of play than if he’s just bursting people down 100 to 0.
Did you and the playtest team expect a lot of the interesting usages of Ryze ultimate? (e.g. using it to move minions to tower, using it as a fake ult and then running into them from another direction)
Afic: Yep. We figured this one out months ago. For a long time, we’ve said “If you want to fountain dive, Ryze is the champion to do it.” What we would do in our playtests is that, when the game was ending with super minions in the base, you ult all of them into the enemy fountain and the fountain targets the super minions. Then you get your penta kills.
It was really funny watching C9 Jensen play it. In our playtests, we kept trying to use it to initiate and every time we did, people see the circle on the ground and throw their shit. It was basically like “Here’s a donut! Fill it with jelly!” Everyone would teleport in and die. C9 did that to SKT and I looked at my coworkers and was like “Hey! It’s just like in our playtests!” The optimal use for this ability is not to initiate. It’s either to escape, to flank, or to clean up fights. We’ve seen it all. We’ve seen the minion lane swap. We’ve seen the ult zhonyas. We’ve seen the “drop an ult for an escaping teammate.”
Did you see the “ult and don’t actually take the ult, then run at them from a different direction”?
Afic: Yep! The fake ult. I’ve been victim to that one before Ryze was released. Absolutely. Funny enough, people are still not using the ult optimally. If you’re standing on the left most side of the ult when it ports, you’ll be on the left most side when you teleport. So if you’re chasing someone, you can position yourself where you’re going to appear. Watching pro players, they’re not actively doing this. People still have not optimized this ability. They’re still doing it super wrong. There’s a lot of wiggle room in bringing down Ryze’s damage and increasing his team play capabilities.
Nidalee. The other problem child. Nidalee just received nerfs in 6.21. The nerfs… they didn’t do anything. Myself being a Nidalee player I looked at the nerfs and said “OK…”. Then I went into a game and immediately replicated my clear pre-6.21 except it was maybe one or two seconds slower. What’s the reason behind such a tiny shift in power? Is it because of changes coming up in the pre-season? Did the team not want to gut Nidalee?
Afic: To understand Nidalee, we need to take a step back. The first thing I want to say about Nidalee is that she’s another one of those champions where the difference between high elo and low elo is ACTUALLY crazy. When played optimally, this champion gets two levels ahead of solo laners. When played poorly or suboptimally… or well, by anyone who hasn’t put in hundreds of hours in practicing this champion, this champion is unplayable. You’re actually setting your team up to lose by picking this champion. Some of the more recent changes have actually closed the gap. Let’s say her high elo win rate is 54% and her low elo win rate is 43%. I’m sort of making up these numbers but the numbers aren’t that far off. The difference between the elos on this champion was ridiculous. After some of the more recent changes like taking away the auto attack resets, nerfing some of the traps (e.g. snaring mobs, and buffing the base damages, her high elo “experienced” win rate stayed the same but her low elo “experienced” win rate went up dramatically. Basically, the gap between high elo and low elo is much more manageable.
With all that said and done, now we can talk about the more recent nerfs. The intent is for it to just be a numeric hit to her win rate. Now that the gap between her high and low elo win rate is much more controlled and acceptable, we can just bring down her power in a necessary way. Maybe these changes didn’t do it. It’s hard to say. But that was the thought process. Initial clear aside, when you numerically are pulling power of a champion (and don’t fool yourself, these are nerfs!), these are to move the needle. The goal isn’t to drop her win rate by 4-5%. Plus, if you want to nerf her by 3-4% win rate , you typically aim for half of that and then make the change. Over nerfing and then buffing and then nerfing again… It generally doesn’t go over well. It’s just unrealistic to have expectations that every change is going to change the win rate by exactly this much of a percentage or change it in exactly this way. This type of approach… In this case, it’s probably to just go with a little and see how much of an impact it has before being more drastic.
Having played League since beta and now being the Playtest Lead and having dealt with all the mechanics in League, what mechanic do you think is the most unhealthy for the game? What would you remove if you had the power to?
Afic: Grave smokescreen.
So blinds?
Afic: Graves in particular. I think Quinn’s is much more acceptable but it’s still bad. Key reason being, when I look at abilities, I think about the goal. “Is this fun?” Fun often comes at the cost of someone elses fun. So I could say “I’m going to play an assassin. It’s really fun insta-gibbing someone as an assassin.” Well, that’s really shitty for the person getting insta-gibbed. There’s some line where it can be good for me and bad for you. Let’s talk about Graves smokescreen. I know I’ve been hit by it. I’m going to assume you’ve been hit by it. When that shit hits you, you feel hopeless. Not only do you not know what’s going on and are forced to make the incorrect play because you have no idea what’s going on, your teammates don’t know you don’t have the information so you just look like an idiot to them while you’re running around. You’re a chicken with your head cut off. What are you doing? You don’t know. That is a lot of frustration that you get on the receiving end of smokescreen. Something like that can exist but it better be super, super satisfying for the other player.
Turns out, Graves players don’t even realize half the time. You don’t even realize how much of an impact smokescreen has. You kill someone you hit with a smokescreen and they run into you? From your perspective, you’re just like “Oh, this idiot ran into me. What a dumb ass.” You hit someone with a smokescreen as you were going to die but they stop chasing you because they couldn’t see you? You’re just like “Ha! What an idiot. He could’ve killed me!” You don’t realize that you just ruined their day by pressing W. All you understand is that you got away. The satisfaction versus frustration… The ratio between the two is completely out of whack. I hate it. I’m tilting thinking about it.
What do you do in a situation when you communicate to a developer that you feel something is unhealthy in the game and the developer disagrees? Do you agree to disagree or do you keep pushing to get the change implemented?
Afic: The obvious copout answer is “It depends”. It’s a lot of factors. Some of the factors would be like what phase of development is it and what are the goals. The first question I’m always going to ask is “What are the goals?” If something is not meeting its goals, that is much easier to dispute than if it were meeting its goals and the goals are just bad. The best way I can typically be persuasive is that I just say “As a player that has played this game for years, my expectations are this. This is failing to meet my expectations in these ways.” When I press these buttons with a promise that this is going to happen, if it is delayed in this sense. If it is misaiming or the visuals aren’t matching. If it is failing to match my expectations, that is typically very impactful feedback. You ask the question why five times. “Why is it doing this?” and they respond. “But why?”. The purpose is to just get deeper and deeper. You really want to get at the root of the problem. If you can coherently describe why something feels the way it does in the most basic of terms, you’d be hardpressed to find a designer that wouldn’t respond to that level of feedback.
Since your team has a lot to do with balancing changes, do you take much feedback from the League of Legends community whether it be from Reddit, the forums, or social media?
Afic: Yeah, absolutely. The main reason I say we do take that into account is because perception is reality. A champion like LeBlanc can have a 45% win rate right? That champion still gets banned because when that champion is strong, it feels super unfair. There’s confirmation bias. There’s a million reasons to explain why that happens. If people think something feels unfair, it doesn’t matter how unfair it numerically is. It’s probably not good enough.
So yeah. We do take that into account but we don’t take everything into account. The saying is “You want to be data informed, not data driven.” I could look at LeBlanc and be like “Nope! She’s 45% win rate. Time for a buff!” but players would flip if you overbuff a champion that’s perceived to be strong. Graves, for example, has a 48% win rate. People think he’s the most OP champions ever. Reality is that he’s just not. He just feels that way because he does some pretty unfair things. When he gets ahead, it just feels terrible. But yeah, we take feedback into account but it’s not the sole factor in balancing changes.
Is the goal of Riot to balance for the main playerbase? How often would you say you balance for competitive?
Afic: We often balance for the majority of the playerbase. That being said, we often balance for competitive as well. If you look at the patches leading up to Worlds, almost all of the changes were focused on balancing competitive outliers. I’d say that the number one goal is “Don’t ruin the game for the majority of players.” With that said, at times we will prioritize creating the best experience for competitive. It’s super complex. There’s a million factors that going into play. Basically, what we try to do is get the best of both worlds whenever possible. When we can’t, sometimes you get a champion like Lee Sin that has a 46% win rate across all elos and is still played in competitive.
Alright! Thanks for the info Riot Afic! That's all for the interview!
Make sure to follow Adam Cohen on his Twitter for League of Legends related content!
