League of Legends is the most-played game in the world. It s also, like Dota 2 and other lane-pushers, incredibly deep, with more than a hundred champions and layers upon layers of systems and strategy. The common wisdom is that lane-pushers are so complicated, they re often intimidating and drive away potential new players. At a GDC 2015 talk on Tuesday, Riot Games lead designer Ryan Morello Scott had an interesting take on that common wisdom. Short version: he thinks it s wrong.

I feel that paradox [between complexity and accessibility] exists, I think the way we analyze it in this industry is intuitive, but I don t think it s correct, Scott said. Let s think about games that are really culturally relevant to us as gamers. That we talk about 10 years later, we talk about them as big pieces of what we re thinking about. Counter-Strike. Starcraft. Halo series. Games like this. These games all have depth depending on what you want in a game...that s what makes popular games. All those games require mastery.

Our assumption from the get-go is that players desire mastery. It s one of our core pillars of League of Legends and Riot in general. With that, we think the paradox is actually: if you try to make your game broadly appealing first, and then make it deep, you fail. Because you re building on a weak foundation. There s nothing to hold up the house. Accessibility s great, but it can t be the foundation of your game. If you build depth first, and make a game that s rich in decision making, highly challenging, lets you master things over tens or hundreds or thousands of hours, then go, okay, we ve made it, great. Now how do we make it so it s more [approachable?]

I think those games, in the long-term, are much more popular...There are people here still playing Counter-Strike. That game is 16 years old. It s because the mastery is that rich. That s how much I think our brains are hardwired to want to learn, to want to overcome. If you deliver that to players in ways that are fun and satisfying, I think it's a mistake to underestimate that people are smart and want to learn. If you assume they don t, I think that s underestimating what players are capable of.

It s an interesting explanation for why games with especially high skill ceilings—games that reward mastery over a span of months or years—tend to foster passionate communities. Lane-pushers like League of Legends and especially Dota 2 are often criticized for their complexity, but Scott made a thoughtful point about two aspects of MOBA design: complexity and depth. They might sound like the same thing, but he had a good argument for why they re separate concepts, and why depth is more important than complexity.

Complexity and depth have a relationship, but they re not the same thing, Scott said. Complexity is a cost, and you try to pay the littlest cost you can in complexity to get the most depth. So if you [say] I want this depth, how do I care that the gameplay systems around the depth are the most understandable versions?

The gameplay mechanic of denying—killing your own minions to rob the enemy team of the experience and gold they d earn by getting the kills—came up as an example of complexity that didn t add much depth, which is why Riot removed denying from the game during its beta. Dota 2 players may disagree, but Scott had an argument for denial s removal.

I think some explanation is important on that, he said. It s kind of like, the analogy I would use is, you want to add skill to the game, you want to make sure skills will be tested. So think about last hitting. You re trying to manage minions. That s taking your focus, and while you re doing that you have an enemy trying to disrupt you. Denying is actually the same skillset you're testing. [Killing] minions, being pressured by an opponent, watching your health, things like that. It double dips into the same skillset.

So does it create a higher skill ceiling? Yes. But it tests the same skill twice. Is it also superfluous? We thought, yes. It adds complexity, but it adds low depth.…The cost to benefit ratio for how much it adds weight to the game s back, versus how much more it allows for the mastery of the players is low, so that s where we started making cuts.

So, high skill ceilings and games that encourage players to commit hundreds of hours to master them? Great. Except a skill ceiling that s too high can add messy complexity while gaining minimal depth. As difficult as lane-pushers are to play, designing them is clearly a whole nother level of tricky.

When the usual stream monsters and Reddit commentators speak about bad picks and poor champions, there's usually an unspoken caveat: these champions are bad compared to other options. Being an unspoken caveat, it's also one that's often forgotten—hyperbolic terms like "useless" or "does negative damage" gets flung around to the point that it's easy to forget the context in which they're judged by.

Which is why the new Nemesis Draft mode is so bloody brilliant. I love it. It pains me to know that it's only here on a temporary basis, because it's been direly needed as an object lesson to the player base—hopefully it'll be something they put into regular rotation.

The gist is: it isn't the champion's fault that you're losing lane or your teammates are doing badly. The worth of a tool is based first and foremost on its craftsman. Sure, a hammer forged from good iron might be better at beating things in than a rock you picked off the ground. But the difference between your grandfather's hammer with the well-worn wooden grip stained by decades of sweat and labor, and the sleek one with the ergonomic grip from a $1000 toolset, is that the former was used by somebody that knew what the hell they were doing. You, on the other hand, just spent $1000 to build a shoddy birdhouse!

Let's take some recent games for example. I keep seeing Taric banned—that is, with the Nemesis Draft, you ban to keep your opponents from giving you something that sucks instead of, as with normal play, to keep them from taking something good. In one game, he slipped through and fell right into my lap.

I proceeded to pick up double-kills in lane, and chase whole teams around with the threat of a Dazzle. I tanked literally everything they threw at me, heedless of any of their threats. Taric is terrific—at least from a team-oriented perspective. Radiance gives so many free stats it's obnoxious, and his passive basically acts as a Sheen, which (naturally) worked especially well after I built the actual item, laying waste to anybody I could Flash onto for a close-range stun.

Or Maokai, in another game. Who gives people one of the tankiest possible top laners in the game in Nemesis Draft? Did they think that the recent prevalence of Irelias, Gnars, Lissandras and other high-damage fighters in the current metagame meant that Maokai was weak? He spent months at the top of the charts as one of the most devastating presences in teamfights—and nothing about that's changed, merely the strategic emphasis at the uppermost levels of the game.

On the flip side, I really wish people would stop letting Sona through. As a former Bronzie that dug himself one game at a time out of the 800 Elo trench (back when Elo was still a numerical value), I know better than most just how effective Sona is, even after her rework—especially after I figured how to make her into an off-carry. Put a Sheen on her, and she's ripping health bars apart every time she hits her Q. No, dude, treat the songstress with respect—just because you think it's funny to force them to play a support champion as an AD carry doesn't mean she isn't really freaking effective as an AD carry, given the great interaction between her low-cooldown abilities and Sheen-boosted auto-attacks.

What Nemesis Draft teaches us is the wide gulf between champion reputation and actual capability. It reminds us that a lot of the roster's only weak because Riot hasn't gotten around to nerfbatting the hell out of Ahri, Gnar or Kalista yet. But when you remove the S-tier champions from contention, seemingly useless champions suddenly find themselves leaders of the pack, if only its players properly understood how to utilize them in combination with other rarely-used options.

Now, that's not to say that all champions below the common tournament-tier choices are equally valid. There are definitely some ways to abuse Nemesis Draft to really frustrate your opponents. But it's less about choosing specific champions and choosing common (and detrimental) champion qualities. For instance:

1. The Unruly Mob: a team without reliable Crowd Control is a team asking to die—repeatedly. Though giving them Fiora, Lucian, Ezreal and other high-damage champions seems like a somewhat self-destructive course of action, they're also champions that have traditionally required a lot of peel and front-liners to be effective. On the other hand, if your team was granted anybody with a root, stun or knockup, especially as an area of effect, all that damage can be (relatively) trivially shut down...assuming you were all smart enough to build tanky against such a team.

2. The Pacifists: The other end of the scale's champions that do have a lot of utility, but hit like wet noodles. Alistar, for instance, simply doesn't have the ratios or qualities in his spells to be anything but a front-line tank—though he is, of course, very good at that role. Same with Nami: she might be able to single-handedly orchestrate the outcome of a fight with good Aqua Prison and Tidal Wave placement, but the high cooldown of her spells and weak autoattacks makes her an especially poor choice as a mage or carry.

3. Purebreed Problems: What happens when a team only has magic-based damage? Well, if your team wants to win, it probably means that everybody, including the support, has a Null-Magic Mantle after their first recall, and Locket of the Iron Solari gets built extremely quickly. Similarly, you can feel the palpable frustration when an all-AD team runs headlong into multiple Randuin's Omens. Forcing the enemy team onto solitary forms of damage makes it braindead easy to build against them, giving you a statistical advantage even if you can't quite secure a strategic one.

4. Knives in a Gun Fight: There are few things sadder than an Udyr with a whole bunch of items that has to wade through slows, roots and Elder Lizard buffed autoattacks to get close to his target. With no gap closer, no ranged abilities and only a small movement speed booster, Udyr might as well be a kite played with by the enemy AD carry and mage. Though ranged champions tend to be designed with a glass jaw in terms of defensive stats, they nonetheless hold an inherent offensive advantage against melee-range counterparts, and one accentuated the more skilled the player is at positioning and movement options.

5. Urgot: ...okay, fair, maybe there is such thing as a truly useless champion.

Feel free to give them Elise too.

Riot Games has pulled back the curtain on the newest League of Legends Champion: Bard, the Wandering Caretaker, a mightily bearded "celestial vagabond" tasked with maintaining the cosmic equilibrium of Runeterra. Hey, nice work if you can get it.

Bard is unique, in that he's the first support who gains advantages simply by moving around. His presence causes sacred chimes to appear on the playing field, and collecting them gives him a brief burst of speed, experience, and mana. He also attracts Meeps, small spirits who travel at his side and throw themselves at his targets when he attacks, dealing extra damage; the more chimes he collects, the more Meeps that join him, and the more dangerous they become.

Mobility is obviously Bard's game, but he "fires out solid poke whenever he's in lane thanks to Cosmic Binding," Riot wrote in the new Champion's description. "Though it deals meaningful damage on its own, the slow (and if procced, stun) give Bard surprising ganking potential, particularly when used in conjunction with Magical Journey. Enemies will have to think twice about escaping through the bottom and top lane brush when Bard can pin them to the nearby wall with a well-placed Q."

Bard works particularly well with Caitlyn, Udyr, and Amumu, but he has trouble with Rek'Sai, Draven, and Leona. "Bard roams to empower both himself and his allies. Each chime he collects soups up his Meeps passive, and as he travels, he s incentivized to help out his other lanes and jungler with Caretaker s Shrine. Crucially, they don t require him to hang around, so Bard can pop into mid, drop off a shrine, then carry on towards top on the hunt for more chimes," Riot's Rabid Llama explained in the Champion Insights section of the Bard page. 

"Each chime gives him a short movement boost (so he can jet around at a decent pace), experience (so he doesn t miss out on too much minion xp), and mana. All this means he s rewarded for roaming, and can (hopefully) leave his marksman during laning without handing them a big fat death sentence."

Catch all the details on the Wandering Caretaker at LeagueOfLegends.com.

'Spectate Faker' is a Twitch stream that uses OP.GG to broadcast the matches of League of Legends pro player Lee "Faker" Sang-Hyeok. It's a simple enough concept, and yet the stream has sparked controversy—with Riot's president, Marc Merrill, saying it "reeks of harassment and bullying.".

The source of the problem is an exclusivity deal between Faker and streaming service  Azubu—one of a number of exclusive deals the site has with Korean e-sports pros.

Azubu went so far as to send a DMCA notice to the Twitch stream's owner, "StarLordLucian". New problem: Azubu doesn't own the content being streamed by Spectate Faker. While they have exclusive rights to broadcast Faker's streams, Spectate Faker isn't a direct re-stream of that perspective—rather, it's the view from a sanctioned third-party client. That footage is owned by Riot.

Here's where things get a bit complicated. Riot's own terms of service claim the following:

"We ll start with our golden rule you can use League of Legends IP as the basis for a fan project that you re giving away for free or that s only generating ad revenue ... as long as you comply with the guidelines outlined below for using our IP. As a matter of fact, as long as you comply with our Guidelines, we think it s great if you create awesome, free and original content for League of Legends fans."

"StarLordLucian" claims that his stream follows those guidelines. He describes Azubu's DMCA takedown request as "illegal," and claims the only ones with the the ability to end the stream are Riot themselves. To verify this,  the Daily Dot spoke to an actual real life lawyer, Bryce Blum, who largely agreed with SLL's statement. Blum's conclusion: "That content isn t Faker s to license—it s Riot s."

So far, Riot hasn't issued a takedown request. But let's go back to Marc Merrill's statement. The Riot head took to Reddit to explain his thoughts on the issue:

"If you can't see how this potentially harms Faker and/or anyone else in this situation, then that is more reinforcement that we need to take the appropriate action to protect players from this type of unique situation.

"As to the comments about our API, of course we want 3rd party devs to do cool things with spectator. But when people utilize one of its components to harm / harass an individual, then we need to potentially re-evaluate our rules."

As for "StarLordLucien's" position, he has posted on Reddit numerous times, most recently with the following:

"I know some people will disagree with this and bring up ethics, but I think this whole issue is about a lot more than Faker. It's about Riot not enforcing their own legal terms of service. It's about a co-owner of Riot Games being completely out of touch with esports and the spectator mode. It's about a company (Azubu) issuing a false DMCA claim for content they didn't even own. These are issues that will affect the future of the game and the spectator mode. All of this needs to be debated for the future of League of Legends and esports.

"Right know nothing my stream does is illegal or against the League of Legends terms of service. Riot can always change their terms. And Riot can DMCA my stream at anytime, as they have the power to put any League related IP or Project to an end.

"If Riot does DMCA my stream that will be the end of it, I won't counter them or try to make a new stream. But I won't be listening to anyone else from Riot or on Reddit lecture to me about morals anymore. To those people I say, I'm doing this stream because I can legally and it's allowed by League of Legends' legal terms."

Finally, and most recently, Faker's team, SK Telecom T1, have released their own statement through their Facebook page:

"First of all, SKT and other pro eSports teams have started streaming business last year to help ensure stable environment for players to play professionally. Not only has the streaming deal expressly helped with players' with their professional activities, it also has been a good medium through which a pro gamer's value is recognized.

"Unfortunately, some of the fans have been re-broadcasting Faker's (and other SKT T1 players') games through the spectator mode, and this has negatively affected players' streaming business. Faker, a member of the SKT T1, also expressed discomfort over the current situation where his summoner name and videos of his games are being broadcasted with no consent.

"SKT T1 team and its players truly appreciate the fans' fantastic support and interest. However, we would like to politely request the re-broadcasting of our players' games without our consent to be stopped."

Thanks,  PCGamesN.

By Joe Skrebels.

It s a sign that things have got out of hand when you start offering people rewards for staying within the law. Riot Games recent decision to give exclusive League of Legends character skins to those who hadn t been banned in the previous 12 months has the air of a dystopian future where good people get sent to prison because it s the safest place going.

MOBAs have long been the world s greatest source for creative takes on being told to kill yourself—the intense need for teamwork, huge time commitment and highly specific tactics becoming the crucible in which anonymous internet dicks are forged. But with the meteoric rise in popularity of online games as a spectator sport, companies are forced more and more to consider how to keep their chatlogs as clean as their bugfix lists.

Curbing bad behaviour is a constant battle for many game developers, says Hi-Rez Studios Austin Gallman. A lot of research has gone into finding the best ways to do this. I don t feel that anyone has really found the perfect solution yet, but having a punishment/reward system certainly helps.

It seems that the traditional system of temporary and permanent bans simply isn t effective enough in a world where games have the populations of countries, and where players—so often spun into a frenzy of competitive ire—have the mouths not just of potties, but of entire sewage reservoirs.

For a company like Hi-Rez, the problems become clear when you look at its games slate. With even automated ban systems requiring a human support response, and players from across the world working their way onto the naughty list, multilingual teams need to be assembled, growing as the games do. That s not to mention the different kinds of bad behaviour that different games engender.

MOBAs are typically more team-oriented, Gallman explains, so we ve seen a bit more in the way of poor behaviour with Smite than we did with our shooter titles. Cheating, however, is a different story entirely. Attempts at cheating were much more common in our shooters. Working on online games for so long has helped Hi-Rez cultivate a good sense, and an efficient system, for punishing infractions, but the tide now seems to be turning towards a more holistic approach.

Obviously, it is near impossible to eliminate bad behaviour in games entirely, Gallman adds, but cultivating a fun gaming environment is always step one. We also think it is important to model positive behaviour via Twitch.TV streaming of the game, game-oriented video content, and other community oriented activities.

Riot s Pavlovian be good, get nice things approach is the kind of positive-reinforcement ethos that the industry seems to be pushing.

Simply teaching people that good behaviour is the norm, and creating role models for younger players, could be the way forward for an industry plagued by literal problem children—and including Riot s Pavlovian be good, get nice things approach is the kind of positive-reinforcement ethos that the industry seems to be pushing for as of late.

But it s by no means the only way—at least not if you ask CCP. The EVE Online developer s famously laissez-faire attitude to its bewilderingly enormous universe extends to its punishment policies, too. In a joint statement, Sigur ur varsson and Dav Einarsson, men with the privilege of having the job titles of senior and lead game master respectively, explain: Many actions lie outside what we consider to be fair play, especially with regards to socially unacceptable behaviour, but EVE Online remains unique in the sense that we have an extremely relaxed ruleset that governs the way the game is played. Many actions that would be a bannable offence in other MMOs are often considered fair game in EVE.

Theft, corporate espionage, piracy—this is the stuff of EVE legend, and the kind of high-level assholery that would have you out of most online games at the push of a Del key. In the cutthroat world of Icelandic space warfare, however, it s par for the course. Accepted wisdom would have it that lugging around a cargo bay filled with real cash is just begging to have it stolen by either other players or the cruel pull of the void.

This doesn t mean that anything and everything goes—CCP s ban protocols are regularly updated and robustly enforced, and, as they put it, our players are extremely creative, and often find ways to breach our policies in ways that we could never imagine (see the above boxout for more on that). The key to EVE s relatively sedate community is acceptance of the fact that it s not the developers who will cause you problems should you ever step out of line, it s the other people playing the game.

varsson and Einarsson again: Promoting largescale warfare and violence in EVE Online is a large part of the game. Regardless of this, for the most part our community remains extremely close-knit, civil and friendly toward one another. This tends to be due to the fact that in EVE, reputation is everything, and can make or break your career in New Eden. With this in mind, our community tends to police itself. EVE players are free to use any in-game tools to wage war and aggress each other, and we recommend that they resolve their differences within the game environment.

While the methods taken are wildly different, there s a parity between Hi-Rez and CCP s approaches, and one that marks the biggest swing in policing online gaming. Whether it s by staying hands off or reminding players that playing nice is the way to have more fun, both companies are creating a status quo, boundaries that players can understand intuitively simply by participating in the community enough. To co-opt the words from some old book: give a man a ban and he ll be salty for a day. Teach a man what will stop him getting banned and, with any luck, we ll all avoid our grim prison-planet future.

[Correction: A previous version of this article erroneously placed CCP about 2,200 km east of Reykjav k, Iceland.]

Riot Games is dropping hints about a new champion for League of Legends with a cryptic tale told over a campfire on the top of a tiny mesa.

"Stories are not just history. They can be so much more," the preview site states, as nightbirds call in the background. "They nourish your mind and, if told well, can even fill your belly. Some tales are warnings, reaching across time. Others uplift our souls from the yoke of everyday burdens. We laugh at fools, cheer heroes, and curse villains until the fire burns down to embers."

It goes on like that for awhile, with references to the Frozen Watcher, the Fall of Shurima, and the Shattered Crown, none of which offers a whole lot of insight to casual observers. The final page is equally enigmatic, but it does at least offer a vague hint as to what's coming.

"There was a time, not long ago, when this constellation was absent from the night sky," it says. "Some call it the Mountain Shrines or the Great Caretaker. Those of us from the floating villages know of an older name, a name that speaks of a universal truth. The name we took from our kind: Bard."

Predictably, there is much speculation underway, but little in the way of consensus: Mount Targon or Ionia, Jungler or Support, Aatrox or Ao Shan, nobody knows, and for now, Riot isn't saying. There is, however, a Wonder Above wallpaper, dug up by PCGamesN, that I think we can all agree is quite lovely. Get it here.

See that unhappy looking fellow above? That's Nautilus from League of Legends, and he's about to be immortalised in the form of an artificial reef. When statues and plaques don't cut the mustard, only an artificial reef will do.

Riot's Oceania office announced the news today following last month's Ocean Week, which tasked Summoners in the region to contribute at least 275,000 points over the course of a week for a variety of prizes. To the surprise of no one, the champion-themed artificial reef stretch goal was unlocked in just three days, and Nautilus was voted from among three champions. Fizz and Nami lost out.

As for when the reef will appear, Riot is currently consulting with an artificial reef company and a marine biologist in order to determine where on Australia's east coast ol' Nautilus should go. That means it's probably a little way off yet, but once finished every contributing player will have their name etched into the reef. 

Since opening a Sydney office last year, Riot has expanded its League of Legends presence in Oceania, most notably with the new Oceanic Pro League.