Red Post Collection: Statikk Q&A Today, Tentative Heal / Exhaust changes to PBE, vladcole on "Why does it always seem like riot is understaffed"

Posted on at 8:05 AM by Moobeat
[ Update: Lyte just tweeted out - "Today, players that show extreme toxicity (intentional feeding or racism, etc) will be instantly 14-day or permabanned in #LeagueOfLegends" and "If this #leagueoflegends initiative is effective, we'll be rolling it out more permanently. Every permaban case is reviewed." more details as they become available ]

This morning's red post collection features a reminder that Statikk is hosting a Q&A later today, ricklessabandon mentioning he has Heal nerfs and Exhaust changes head to the PBE for testing, vladcole stepping into a reddit thread regarding players feeling Riot is understaffed, and more!
Continue reading for more information!

[ Reminder ] Statikk Q&A is today!

Just a reminder what Statikk, a design on the live gameplay & balance team who also worked on the Sivir, Evelynn, and Kassadin reworks, will be hitting the boards to answer YOUR questions later today!

Be sure to riftwalk by the Q&A thread to get your questions in and , as always, I'll have a summary of the Q&A out in a future red post collection!

Tentative Changes for Heal / Exhaust coming to PBE

 In a thread asking if Barrier will ever be buffed to compete with the latest version of Heal, ricklessabandon noted he has a few summoner spell changes coming in the next PBE update:

"not sure if this is relevant or not, but heal is potentially getting nerfed in the 4.13 patch (the changes should be in the next pbe update). 
i also have a change to exhaust, but that's it for summoner spells in 4.13 (and again, they're candidates—like anything else on the pbe, they will be subject to change/removal)."
He continued:
"the bigger question is whether or not we should even have both barrier and heal on summoner's rift. we might look into summoner spells more in the next pre-season, but until then we're trying to work around the existing spells when we can."

Homeguard Sorcerer's Shoes hotfix early this week

As you may have noticed, buying the Homeguard enchantment for Sorcerer's Shoes has been disabled on live since late last week due to a bug:
"Hey all, 
Due to a bug with Sorcerer's Shoes giving double the magic penetration when upgraded with the Homeguard enchantment, we're going to be disabling this item combination (specifically Homeguard enchanted Sorcerer's Shoes, not the shoes or the enchantment alone) to minimize its impact on the fewest possible players. 
We're working hard to find a fix and plan to hotfix it as soon as we can early next week. Sorry for the inconvenience!"
When asked why it is taking so long to fix, ricklessabandon replied:
"it took me a few seconds to fix it, which happened after a few minutes of checking the file histories of a couple things. i think the scripting side of things (before testing or anything) had a fix candidate before i started seeing people talk about it on social media, and definitely before we disabled the item. the reason that the fix isn't already live is because of the process that happens after that (which was made more complicated in this case because we were preparing the 4.12 patch for deployment for regions that didn't have it yet). one of those steps is making sure that the candidate actually fixes the issue, so it's pretty important stuff that we don't want to skip. 
i think the current plans would have the fix out to north america early this week, though i don't know exactly when as that's handled by other teams. the moral of the story is that it's really important to make sure things work right the first time, so no one has to suffer through a disruption in normal play because of bugs like this. :<"

Feature and Fixes for Locked Camera

In a thread asking about any sort of upcoming changes for locked camera, Nome noted:
"We're actually looking into some features and fixes for locked camera (includes space bar hits) right now. By right now, I mean, right now. We have to be careful with major since Worlds are rolling up soon--and that can affect our release cadence--but I feel pretty good about saying that we'll be putting out updates to the camera in the upcoming patches."

VladCole on "Why does it always seem like riot is understaffed"

vladcole has stepped into the ring to answer a redditor's question "Why does it always seem like riot is Understaffed", explaining  his point of view on Riot's hiring practices and clearing up some common misconceptions that where popping up in the thread.
"This thread has gone on for a while now without contribution from an insider, so I'm going to take the scary step of wading in and trying to share some perspective on a few of the common misconceptions that are popping up in the thread. 
So with regards to this comment, I can't speak for Valve, but I'll say that we actually have a philosophy of hiring what we call "athletes" over specialists. An athlete who can medal in multiple events is generally better (and worth more to us) than the specialist who can do just one thing very well. 
We hire in this way because the future is uncertain. By having teams that are capable of many things, we're able to address new challenges and new opportunities as they come up. 
Inside Riot, I've probably heard the following quote a dozen times, "It is not the strongest of the species that survives, nor the most intelligent, but rather the one most adaptable to change." (This quote is commonly misattributed to Darwin ... some professor somewhere actually said it). 
Also, it's true that we lack experience running a game at this scale. Most of us are in the biggest job we've ever held in our lives. That's not because we're incompetent, but because we value aptitude and attitude over experience when we assess hires, based on our belief that smart, adaptable people will beat experience every time. Also, there are some aspects to the scale of League that make it hard to find someone with relevant experience. There's literally not another network engineering/game engineering team in the world that's had write code that can serve more than 67 million monthly active players who collectively hammer servers several billion hours per month. 
This isn't to suggest that we're ace at all we do. As the OP rightly points out, there are a ton of gaps between what players would like us to do and what we're actually able to get to."
He continued, replying to someone who believes money is why Riot fails to hire more staff:
"I stated that prior comment too strongly. We have specialists where it makes sense. We prefer to hire generalists. 
"Everyone knows that money is the real reason why they don't hire more people" 
This couldn't be further from reality. I give a presentation to new hires at Riot every two weeks. The last three of those classes had close to 30 people in each of them, many of them very experienced, very capable industry professionals. We've been hiring as rapidly as we can over the last 5 years without sacrificing our high standards for talent.":
He continued:
They'd rather take the failures and related costs on the chin than get some more full time employees.
Not the case at Riot. As I noted on a prior response, I give a presentation to new hires at Riot every two weeks. The last three of those classes had close to 30 people in each of them, many of them very experienced, very capable industry professionals. We've been hiring as rapidly as we can over the last 5 years without sacrificing our high standards for talent. 
We have hundreds of positions that we're actively recruiting and interviewing people for (see them We are growing, but of course we're trying to do that in a prudent way -- it would be dumb to expand the cost base too far ahead of the revenue base -- all businesses need to grow with an eye on sustainability and Riot's no different. 
Regarding office space, that's not a blocker to hiring. Our note about the new office space was posted here: . The bunk desk thing was a joke. Sorry the humor fell flat. 
edit: added quote at the top to make it clearer what I was responding to."

vladcole then dove into a long discussion on work that players value, don't value, and more in response to why summoner's feel so little gets done "
"Lemme try to take this on. Without (yet) disputing your charge that we're doing very little, I'll suggest a framework. 
The output of any game development team or company equals the sum of: 
(a) player-facing features that players value
(b) player-facing features that players don't value
(c) game improvements that players can't see/touch/taste
(d) business infrastructure upgrades for non-player "customers"
(e) business services rendered to non-player "customers"
(f-z) probably much more, but this is good enough for now. 
Your charge is that there's not enough (a) based on the headcount you cite. One way to look at it is that your perception is all that matters here. If we're disappointing you, then we're disappointing you and that's all that matters. That's a real problem. As a company that cares very much about not disappointing players, it hurts when players don't perceive value in what we ship, or notice the improvements that we're making. 
But what I think is really going on is that there's a lot of (c) and (e) that we don't get credit for, and actually a fair amount of (a) that tends to not be top of mind when threads like this are generated. So lemme give some examples of each of the items above, and talk a little about why I feel this thread's misjudged us. 
(a) player-facing features that players value 
This includes things like the new summoner's rift update, special game modes, champions, skins, and even stuff that appeals to a smaller portion of the playerbase, such as the Pentakill Album , the Jinx music video , events (from big ones like PAX and Gamescom to smaller ones like GaymerX) and so on. I'd argue that the velocity of player-facing improvements is pretty high, but not as high as we'd like it to be (because we know there are efficiency gains we could make, and because, as others elsewhere in this thread have noted, we've got some legacy code issues that make development in some areas much slower than it should be). 
What's interesting here is that some of this work is region-specific. So, updates to the PC bang loyalty program (South-Korea specific) aren't visible or appreciated by North American players. No American gamer really cares or should care about what happens in South Korea, but the PC bang owners care a whole lot about the accuracy of the billing communications. With offices in 13 locations (and growing), we've got several hundred people outside the US focused on building services and features that are market-specific for players in the more than 145 countries who play League of Legends every day. We're not asking American players to care about all of that, but I'm just bringing that context in here because it's relevant to understanding why we have the number of people that we have. 
(b) player-facing features that players don't value 
Not everything we ship is a hit with players. So there are occasional features that are shipped that are met with a yawn. I think we've got a decent batting average, but we could of course improve. More interestingly, there are a number of features we ship that result in invisible improvements to the game. Our initially gnarly codebase is getting improved iteratively, day after day. These improvements to code improve the velocity of future feature development efforts, decrease the chance that we'll inadvertently create new bugs, improve the uptime of the game servers, etc. 
An interesting example of this, and someone inside Riot may shoot me for jinxing this, but as of tomorrow, EU players will have experienced no unplanned downtime (by our strictest measurement standard) for 30 days straight. That's the first time we've ever achieved that milestone, and it's only possible because of the massive investment into code stability and network infrastructure (including the Amsterdam data center). I'm not saying you should give us a cookie here, but I am saying that most players don't notice when things aren't broken. Server uptime and reliability is expected, but it's a pretty remarkable improvement over how instable EU used to be. 
(c) game improvements that players can't see/touch/taste 
Most avid players read the patch notes, and are aware of the changes to live balance. But we're also constantly adjusting a number of systems that make everyday play more enjoyable and reliable. These changes aren't documented via patch notes, but have a positive impact on the player experience. A good example of this is a tweak to the matchmaking algorithm. Because matchmaking is a black box, algorithmic enhancements aren't generally noticed or felt. The impact might be a 1% improvement in the overall fairness of matches, something that no individual can see/touch/taste, but that really matters across hundreds of millions of play sessions. 
(d) business services and infrastructure upgrades for non-player "customers" 
This is an area where the lack of visibility is even greater. It takes a team to develop and iteratively improve the Riot Games API ( ), but this project is really important to a number of third-party developers who create experiences that millions of players use on various websites every day. 
Another example is something as this page: . Writing a legal policy that allows content creators to do cool stuff with League of Legends on YouTube took lawyers, community team members, and business development effort. But this effort to create a reasonably permissive legal stance towards Riot's intellectual property is why League of Legends content flourishes on YouTube. These sorts of things don't just happen as a matter of course -- they take real thought and planning and effort. 
A final example in this category is the candidate to Riot. For this summer's class of 39 interns, we received 5,781 applications ([6] ). By the time this summer's over, it will have taken Riot a collective effort of many thousands of hours to serve those people -- from reviewing resumes, to conducting onsite interviews, to planning events for them, to handling hiring paperwork, to setting up payments (subsidized housing) and payroll... the list goes on. An internship program is a massive investment in time and organization resources into the future of Riot Games that doesn't result in a ton of player-facing improvements today, but that just might result in some big and delightful player experiences in the future. 
Supporting top tier talent takes a cast of business services people. We've got a great IT department that helps keep all of our computers running. We've got a facilities team that puts out fresh fruit and cereal and keeps the coffee machines humming smoothly. We've got a legal team that creates the contracts that allow us to procure everything from property leases (13 offices worldwide) to contract engineering services and software licenses. We've got finance people who handle the massively complex work of doing taxes in all of the countries where we do business (and much more), plus a tremendous team of people handling education efforts for Rioters so that we're investing in future learning. Almost every day at Riot, there's a class being held to help everyone level up their skills. 
All of this stuff takes people to run -- and for the most part, we're doing a lot more than organizations many times our size. (It's not that those other guys suck -- it's that we're just behind the curve in terms of growing to meet the needs of all of the players and non-player customers that have asks for us. 
I've gone on way too long here... time to come up for air."

Returning to the discussing about staffing, vladcole replied to the original posters follow up questions:
Ty for your contribution. However, I'd just like to clarify my point. I did not try to imply that Riot is truly understaffed, I'm sure you already have thousand of employees. What I was trying to express was the "feeling" that Riot is understaffed and this feeling mainly stems from the way you (as a company) seem to manage/handle/allocate your resources around projects. It's not really a "Why don't you hire 10 more people?" (which could still be a good question in certain cases...I believe that some major changes/improvements to the game shouldn't be left in the pipeline for so long only cause teams are focussing on other stuff) as to a "Why does it seem like the 10 people you have already hired seem to actually be out of sync/rowing in different directions". Basically a well-coordinated team of 10 people will work at a 100%, the one that isn't coordinated (everyone does what he wants, doesn't focus/plan etc) works at a 50% and therefore from the outside it actually looks like they only have 5 people (and end up looking understaffed) when in reality they probably have all the people they need.
We actually are understaffed in many key areas. I've got an open role I'm trying to hire for that's been open for over a year. I've interviewed two dozen of the best people in the world for the job, and haven't found someone yet that's just right. My experience is pretty common for many hiring managers at Riot. We'd rather run perpetually understaffed than throw bodies at the problem. 
The other issue here is that we've made too many public promises too soon. If we could just do a better job of not promising stuff before we deliver it, we might not disappoint you so frequently. We need to get better at emulating other highly successful companies like Valve, Blizzard and Pixar, all of whom say "we'll ship it when it's ready and meets our quality bar." 
And yes, your point about coordination is a really important one. I have some visibility into our organization philosophies, successes, and failures and am confident that we're a company that's better organized (for where we are in our lifestage) than most others I've worked at. No organization is flawless, of course, but I see the company's leadership team working on this particular problem every single day. 
Getting everyone rowing in the same direction is incredibly important for any company to do, but especially important for a company that works the way we work."
He continued, expanding on the hiring practices:
>we value aptitude and attitude over experience when we assess hires 
That's funny, because I was just telling my girlfriend the other day about how Riot's problems stem from the fact that they clearly have no idea what they are doing, and here you are happily admitting that you have a hiring philosophy of getting people in who have no idea what they are doing.
That's one way to look at it. We actually take more of a glass-half-full approach to this. We like to be the sort of place where we trust people to do stuff they've never done before. That's the sort of place I want to work. That's the sort of place most of us want to work. It'd be a real drag to only be trusted to do a thing after you've already done it for X years. 
What's more, people who have never done a thing don't come at it with preconceived notions about the "right" way to do it. Preconceived notions are often wrong, and experience can often lead people down a path that doesn't make sense. 
Take, for example, some of the very experienced venture capital investors who took a first look at investing in Riot Games, back when the company's founders were running around looking for someone to bankroll their dream. Some of these investors were very rich, very successful, with sterling resumes and 20-40 years of experience investing in and building game companies. The investment community is smart. 
And yet, some of them gave some horrible advice. One suggestion, for example, is that we should have built League of Legends to run in a browser. This was based on the reasonable (but flawed) idea that if you're going to make a game free to play, you need the lowest possible barrier to trial. If someone could fire up a browser and try the game, then that's going to be good for the game, right? 
Wrong. Brandon and Marc (being core gamers themselves), knew that core gamers would hate the latency and UX sacrifices that would come from trying to play a game like League of Legends through a web browser interface. They knew that gamers would prefer a one-time, multi-gigabyte download if it meant that they'd get a responsive, immersive experience. (Let's forget, for the moment, the challenges of our current client. We're intensely aware of its shortcomings.) 
This is just one example of the many examples that I could cite regarding experience. In fact, the history of business is filled with examples of the dreamers who upset entire industries because of their lack of experience. It's the guys who don't know any better who often try something that everyone else -- with all of their wisdom and experience -- thinks is impossible."
Vladcole also elaborated on hiring outside specialists
Why the fuck wouldn't Riot at least hire a bunch of specialists as outside advisors, helping the projects and bringin tons of experience, without the need to have them on staff forever?
We do hire outside advisors, especially for stuff that's outside of our core competencies or that's better handled through a short-term engagement rather than ongoing operations. But when there's an activity that's central to the art and craft of making games, we'll hire the specialist (if necessary) in house. We have a number of specialists... so we're not against hiring them. 
Did you see the note I made elsewhere in this thread about the specialist we hired from Netflix -- literally one of the world's top experts in management of massive, massive databases? That's the sort of hire that we're making on a regular basis. We've hired top game designers, top concept artists, top producers, top writers, and more from all over the industry. 
But as you say, none of those hires matters if you're feeling that we're not getting shit done. Really sorry that you feel this way about us right now. The only thing that's going to make you (and those who feel the same way) happy with Riot is for us to address the stuff that pains you. 
Interesting note about items. I don't know enough about game design to respond to that point, but I've made a todo note for myself to get educated on the issue so that I can understand whether you're talking about a failure to deliver or an intentional design choice."

As for why other companies such as Valve can run their game with much fewer employees, vladcole replied:
"I can't speak for Valve at all and it'd be a mistake for me to ignorantly conjecture about how they work. 
I'll agree that they're doing some things better. Steam is an incredible platform that I at least once a week to play other games. I see many Rioters using Steam and enjoying all that that platform has to offer. We've got tremendous respect for what Valve's accomplished there, and in many ways everyone working in the PC games business owes a debt to Valve for what they've done to keep the PC platform a vibrant haven for core gamers. 
But you're comparing two very different organizations to each other and assuming that the operations of each are similar enough for the metric of "number of people" to be comparable. Check out the really long (too long!) response I made elsewhere in this thread: In that response, I try to answer this question, though I'll admit that I'd need a lot more words to adequately get to all the nuance."

When asked about the age of the Riot office, vladcole noted:
"I don't go around asking people's ages, but I can guess. Here's my SWAG: if Riot's ages were plotted on a bell curve, you'd probably have a mean of about 32, a mode of 30, a left tail that probably gets as young as 21 (not including interns), and a right tail that gets to the mid-late 60s (not including non fulltime contractors and advisors, many of whom can skew older). Plenty of grey hair around here. I'm 39 in October, myself."

No comments

Post a Comment