Jordan Rapp is a former elite rower and triathlete and current masters cyclist who has turned his sights on virtual sports and gaming. After working for several years as a game developer for Zwift, Rapp is now an engineer at Apex Legends.
Dr. Seiler hosted Rapp in November 2022 as a guest lecturer presenting on the intersection of esports, traditional sports, and future frontiers. They cover Rapp’s athletic background, his transition to game development, how esports are similar to traditional sports, how they’re different, and how aspects like doping differ between the two.
For more in-depth knowledge on Zwift and working out on bike trainers, check out our Indoor Cycling Pathway at the link below.
Reposted on Fast Talk Laboratories with permission of Dr. Stephen Seiler.
Dr. Stephen Seiler 0:00
Okay folks, it is 5:15 and as you have already heard, some of you are here. Jordan Rapp is my guest today. I don’t know Jordan very well, but we’ve met via Twitter where a lot of sports science geeks and athletes and coaches kind of are in my little world and I’m in their world and luckily it sounds like I think we’re going to finish our guest series that you have listened to with a highlight because Jordan is so deeply embedded in your world of gaming and also my world of sport, of endurance sport from before. So it’s a happy intersection. So with that, Jordan you can tell a little bit more about yourself. You’re Princeton engineer, rower turned triathlete. Developer, game developer and so forth. So lots of hooks to place you on here. Take it away.
Jordan Rapp 0:58
Okay so, here I’ll share because I have a presentation I put together for of you. So yeah, I think ideally right it’s interesting I think that there isn’t so much difference between the world of gaming and the world of sports and I hope we’ll get into that. So if sports are just a skill and so we can talk about what that means. So first who am I? I have been a elite endurance athlete for over 20 years. In college I was a national level collegiate rower at Princeton University, a lightweight and then transitioned to triathlon and was a professional for 13 years. Six-time Iron Man champion and the 2011 ITU long-distance world champion and then since retiring in 2017, I compete as a master’s cyclist. Placed top 10 at the Dirty Kanza, 200 mile. U.S. National Masters Pursuit Champion. Bronze and silver medal at the UCI Masters World Championships, also in the pursuit and then gold medal in the team pursuit at the masters level my background is in…
Dr. Stephen Seiler 2:11
Can I just ask how old are you now Jordan?
Jordan Rapp 2:13
Dr. Stephen Seiler 2:14
42. All right.
Jordan Rapp 2:16
Yep. I have a bachelor’s in mechanical and aerospace engineering from Princeton and I am actually currently enrolled, fitting digital class, online classes. I’m getting my master’s in computer science from Georgia Tech and I’ll finish this summer, also in online program. Yeah, I think it’s amazing. A lot of downsides obviously to covid, but I think the rise of and sort of acceptance of online education as sort of a meaningful and great way to learn is awesome and in addition to all the racing I did, I worked as an engineering consultant and field test engineer for Zipp/SRAM/Quarq. So it’s a bike products family and then also for Specialized Bicycles, Diamondback bicycles and then more and I now work as a technical designer software engineer at Respawn Entertainment on Apex Legends, but have been there for two years and then prior to that I worked at Zwift in the same sort of roles for three years.
Sports and Skill, What’s the Difference?
Jordan Rapp 3:20
So, let’s talk about sports and skill and why there really isn’t too much difference between games, well certain kinds of games and sports. So when I think of skill, I think of basically that it’s a proxy for neuromuscular coordination. So cycling is clearly I think, we think of as a sport, also a skill. Bike handling, the training, all of that and I think an interesting one and I had my summer job for six years was I fixed vintage and historic race cars and I got those. So I actually have still a bunch of connections with the racing community. Driving a race car is most definitely a sport and we can we’ll talk a little bit more, but I would say even though it was shown for a while on ESPN, I don’t think of poker or chess as a sport because I don’t see the same sort of muscular, it’s all thinking. It’s obviously a skill, but it’s not sort of a skill for our purposes here.
So the muscular component is essential and so bike handling, a steering wheel, mouse and keyboard, controller, gamepad, joystick. You’re talking fighting games, things like that. They all require coordination and are highly specialized skills like train through many many many many many hours of repetition and endurance is clearly still relevant. I think you definitely have the idea and I’m not saying that professional gamers are the pinnacle of health and yet, Apex Legends tournaments, other major gaming tournaments can be six, eight plus hours and so it’s again very focused endurance like hand and wrist type of stuff, but I think they definitely, the big esports teams have started to recognize most of these guys. They all have dietitians, strength conditioning coaches because it’s overuse injuries in the hand and wrist are massive and so I think posture, overall health.
Playing a game for 10 to 12 hours a day, which a lot of these guys do, is it’s definitely not easy. It’s weird to think of it as endurance, but it’s specialized endurance and I think sort of going back to Zwift and indoor racing. Riding a trainer removes certain skill components of cycling, but it adds others. Sprinting on a trainer is not the same as sprinting on the road, but sprinting on the track is also not the same as sprinting on the road and sprinting on a BMX bike with tiny tiny chain size right at massive RPMs is also not the same. So I think specialization of skill is obviously a thing and I think people will gravitate to movement and skill patterns that suit their specific style. That’s true that the best Zwift racers are not necessarily the best road racers and then the best players in Apex Legends are not necessarily the best War Zone players or Fortnite players. Obviously if you have skill in a certain thing like a good bike rider is a good bike rider. A good FPS player is a good FPS player, but as you get higher and higher up the pyramid and into more and more elite level, specialization becomes even more important, which sort of gets into the importance of specificity.
Importance of Specificity
Jordan Rapp 6:45
Racing a bike is good practice for racing on Zwift, but racing on Zwift is the best practice for racing on Zwift. It’s deliberate practice and you need to know and I think that’s where we get into… you think of Zwift, there’s a draft. There’s physics, but you can’t feel them, but you learn to understand them through other cues and I think that’s where, again you can see that time on the specific platform is massively critical, interesting sort of side note on that is that you can often find sort of opportunities with others to sort of make use of that specificity in other ways.
So actually there’s a bunch of race car drivers that are quite active in the endurance sports community I know through triathlon and I used to write a column that was sort of similar to this talk for Lava Magazine for oh I guess about five or six years, sort of finding this neat intersection between technology and sports and one of the ones, I talked with a very high level NASCAR driver named Landon Castle and so when he watches footage of races, whether it’s himself driving or other people, he will often do it on Zwift because he said if I watch the whole race like it’s useful, but if I watch the whole race in a sort of state that it’s very similar to how I feel when I am driving the car, it’s much more useful.
I asked him, so for typically say a four-ish hour race, a 500-mile NASCAR race, which is like three to four hours, he would say, “My heart rate is typically about 160 plus beats a minute,” which is pretty high and he’s losing over a liter an hour of fluid and so basically he will ride his bike at a similar level so that when he watches the footage, it makes it more real. It’s much more of what it’s like when he’s actually driving. I think you all, if you have experienced that sort of the tunnel vision, like closing in, it makes it much more applicable and so I think that it’s interesting where you find sometimes specificity in unexpected ways.
Dr. Stephen Seiler 9:07
Well I got to. Let me just… There’s lots of things you’re saying that are kind of interesting to add on to, but there is research on the brain that shows that if I for example am watching a sport that I have some familiarity with, there are some parallel processing going on where when they do movements the same parts of my brain that would be controlling those movements will get activated. So I’m in a sense kind of mirroring their movements in my own brain as I watch. Does that make sense?
Jordan Rapp 9:42
Oh yeah. So I trained for a long time as a triathlete. One of my closest friends and long time, I was probably more his training partner than vice versa, was Simon Whitfield. The 2000 Olympic gold medalist, 2008 silver medalist and I remember we both came across some of this research and I think for swimming especially, we were big believers in watching good swimmers because for most triathletes, I would say swimming is probably the thing that comes… they’re good cyclists and good runners, sort of similar physiology right and then they become good enough swimmers. I mean there’s some exceptions.
Dr. Stephen Seiler 10:22
It’s just technologic. It’s just a different movement. There’s no crossover.
Jordan Rapp 10:28
Yeah and so I remember we both, I especially did not have a ton of swimming background. I watched a ton of video footage and even when I was a rower, my coach would just gave me a bunch of video of rowers and just said ‘just watch these’. YouTube makes it amazing now. When I was rowing, it’s so early days of internet, I have some footage and it’s captured at like 320p or something like that. It’s like this tiny blurry and now you can get super slow-mo and incredible footage. Just hours and hours and hours, but I remember watching Michael Phelps just swimming on loop because it was to try and get some of that just absorb a little bit by osmosis.
Dr. Stephen Seiler 11:17
Right. Well, another issue is you mentioned this, I guess there’s… in Norway we have this model of sports where we really want young children to sample. To do different sports, to build a kind of a motor coordination library in a sense or build out their skill and then specialize late. So it’s kind of a late specialization model, keep them sampling, doing different sports, multiple sports for quite a few years and then at some point you know teenage 14 15 or 16 they start zeroing in on the sport that they’re best at, but they have this motor skillset that is easy to then start to specialize and it sounds like gaming has some of the same. You have the general abilities that you build, where a gamer from one game can pretty quickly be at least decent in another game very fast just because they have this general gaming ability and then the specialized skill sets built on top of that. So it seems like also there esports and sports are very similar in that kind of model. Is that reasonable?
Equipment Makes the Man
Jordan Rapp 12:37
Oh totally. I think one of the things that’s most interesting and we’ll talk about a bit more, is players who play a game at a very high level. Apex we have you can play either with a mouse and keyboard or you can play with a controller. Whether you’re playing on console or not. So these are very different input methods and yet there’s definitely a perception among the community that because of the way that we’ve tuned the game, that currently at the elite level, players who play on controller have an advantage and so you have some very, very high level, some of the best in the world mouse and keyboard players who have decided that controller is an advantageous. It would be a sort of analogous to the cycling world if you had sort of people that were like, “Okay I’m very, very good at time trialing, but time trailing is sort of not relevant.” But sort of breakaways are now really a big thing in the peloton and so time trialing and breakaways are kind of similar, but obviously very different. And so you’re sort of like, “Okay I’m gonna become a breakaway specialist.”
And so you have some of these players that were mouse and keyboard players that have decided, “Okay, I’m going to become a controller player.” And you would think that’s totally different and yet they operated an extraordinarily high level then there’s sort of this very slight dip and then they’re basically back at almost the exact same level with this totally different input and I’ve been playing on a controller my whole life and yet it doesn’t matter at all that I played controller.
Dr. Stephen Seiler 14:21
You not getting advantages, they are.
Jordan Rapp 14:24
Yeah, he has 20 years of keyboard experience and like a month of controller experience, but because he has so much more gaming experience, he’s clearly better than I am and it almost doesn’t matter and I think it’s you can see that all of the other skills, the general skill bucket is so big that the specific skill then becomes essentially, the difference in the specific skills it doesn’t make any difference because all of the other skills support that specific skill and already allow you to leap off at a very high level. I think I saw that with triathlon. When I started racing triathlon, the fact that I had been an elite rower, it was like ‘oh, this is basically the same thing because I have this huge general skill, but I’ve seen…
Dr. Stephen Seiler 15:12
A racing skill set, there’s chat going on, so you got to check the chat out because these guys are…
Jordan Rapp 15:20
Yeah. We’re going to talk way more about aim assist and controller players because I think it’s one of the things that’s most interesting from this and so we’re sort of starting to tease in there with the idea that equipment makes the man, but vice versa. So you have machine specific physiology and I think this is something I saw at Zwift where we talked a lot about, you need to normalize the trainer for obvious reasons. Measuring devices, which is what a trainer is, they’re not accurate enough to be precise. So everybody needs to ride the same trainer and sort of all those trainers need to be calibrated in the same way, but nevertheless certain riders will perform better on certain trainers and we saw this, that certain riders who are very, very good riding on a like a Tacx Neo or actually sorry riding on a Wahoo Kickr, we’re much less good on a Tacx Neo. The Tacx Neo is the most sort of ‘realistic’, but the there’s an interesting question of whether realistic meaning similar to on the road, but yet it’s not clear that necessarily makes for the best racing.
Dr. Stephen Seiler 16:40
I have a Tacx Neo bike, the full. Is that disadvantage in me or…
Jordan Rapp 16:48
Dr. Stephen Seiler 16:49
Jordan Rapp 16:51
Yes because it’s the most accurate and realistic to on the road, but it’s interesting because it’s not clear that if someone is and so the way that I would say, is we had this when we were doing evaluation, that there were people who could make their trainer report that they were generating 1500 watts and yet they were almost certainly not generating 1500 actual watts if you were talking like on a real ergometer, but the question then became did that matter because they weren’t cheating, but they had basically figured out how to game the system and of course that’s part of gaming. It’s figuring out which, so in Apex certain characters are overpowered. In Zwift, certain trainers are overpowered and so it becomes this thing of like is that a good thing or a bad thing? Do you want to encourage that, and I think sport has always I think struggled with this and they sort of have this uneasy relationship where with cycling you have the Project 96 where they had all these super aerodynamic bikes and then they banned them all and now they’re sort of slowly allowing more and more of that technology.
The rubber suits in the pool were super interesting. They exposed that there were certain people that benefited more. So there was a German swimmer named Paul Biederman. In 2008 he was ninth in the world in a 200 and 21st in the world in the 400 free and then in 2009 in a rubber suit because that was the year of the rubber suits. 2008, 2009 was the jump. He won the 400 free World Championships and he broke Ian Thorpe’s world record. So you have a guy who was 21st in the world, breaking the world record of the arguably… Thorpe was the best middle distance freestyler of all time. Better than Phelps. His PR in 2008 was 3:47 and then he goes seven seconds faster in 2009 and then they banned the rubber suits and Biedermann sort of fell off the bat and people criticized him which I thought was ridiculous because it’s it’s not his fault that FINA made these decisions, but it was for whatever reason his physiology. He benefited more from the rubber suits than other swimmers did.
Dr. Stephen Seiler 19:25
The shark suit the thought was that it was the external surface that was changing the water interaction with the skin and it was given an advantage, but the in retrospect they found out that what was actually happening was it was compressing the body and it was improving flotation. So they were sitting high water.
Jordan Rapp 19:44
Dr. Stephen Seiler 19:46
So the whole idea of the shark skin engineering and all that stuff was bullshit. It was the buoyancy.
Jordan Rapp 19:54
It totally was. It was like when they did the first Tesla. It was when they did the first sub 2 and they all the nutrition and the shoes and everything like that and then I think if you did the math it was like 95% of the actual like gain that Kipchoge had came from the aerodynamics of basically the Tesla and all those pacers
Dr. Stephen Seiler 20:20
Jordan Rapp 20:21
So like they talked about all this stuff, but it was basically an exercise in aerodynamics, but shoes. The Super Shoes are also very interesting because I think they’ve made a massive impact in the marathon, but I think the sport where they’ve made an even bigger impact is in triathlon. So typical triathlon, sort of marathon time was for an elite male was we’ll say it’s about 250 and you could see the Super Shoes came out and it just dropped five to six minutes. 244 is the new 250. So you think on a relative basis compared with regular marathoning, it’s way more, but it makes sense because the cushioning in that response for a race that’s eight hours, where sort of just pure fatigue and muscle soreness is much more of a limiter. The shoes are going to make a bigger difference and so I think it’s interesting how you see that the technology ends up, certain athletes will benefit more. I would bet that heavier athletes, your triathletes are going to benefit more from lighter athletes because again there’s more actual shock absorbtion.
Dr. Stephen Seiler 21:26
We’re less socialize. Well if you move your conversation, these issues over into the gaming world are there mechanical like innovations in the controller units that are making the same kind of changes or somehow allowing for the talents of the athlete to express themselves more? Is that happening?
Controller Mechanics and Drift
Jordan Rapp 21:50
Yeah. So the arguably I would say pretty convincingly, the best player in Apex in the world right now is an Australian a guy named Genburten, that’s his gamertag and so he has two controllers that he comes with and so we have a very detailed inspection process to make sure that people aren’t cheating. That there’s nothing in there because obviously with electronics you can make it so the controller actually makes you better. So we’re checking for those, but his controllers all have a very, very slight drift. He has one controller that has a slight lateral drift and he has one control that has a slight vertical drift and the reason our theory anyway is that in our game aim assist only kicks in when we detect an input. Some games like are passive like you could just have your hands off the controller and it will track. For Apex, you must provide some input, but what the drift does is that even if you’re not actually providing input, the controller…
Dr. Stephen Seiler 22:56
It’s seen as an input, okay.
Jordan Rapp 22:58
It is and so it will cause aim assist to always to kick and so this gets interesting. Should that be allowed and he’s clearly not cheating. You could argue the other way. The drift is problematic because he basically has to manage it, but he certainly is taking advantage of what the game provides to him.
Dr. Stephen Seiler 23:21
So he is very cognizant of that drift and what effect it has for him.
Jordan Rapp 23:27
Yeah although it’s interesting to use the word cognizant because I don’t know if he’s aware, what level is he aware at because it’s obvious the game is so fast that there’s no thought process, but I think it’s more that he’s cognizant at some level for sure because he perceives
Dr. Stephen Seiler 23:51
He’s at least cognizant that he hasn’t chosen to buy a controller that doesn’t have the drift.
Jordan Rapp 23:56
Right or more that he probably buys lots of controllers and is looking for basically specific drift a certain amount of drift or that he plays with a controller until it gets that drift and then he plays with it until then the drift becomes too much and then he throws it away and gets a new one.
Dr. Stephen Seiler 24:13
That’s like shoes then. So they’re literally wearing this stuff out or there’s a sweet spot for how it performs and then they get rid of it because it no longer performs adequately. Is that correct?
Jordan Rapp 24:24
Yeah and mouse and keyboard even the specificity of the mouse itself. How sensitive the mouse is or not. The switches on your keyboard. How much you actually have to depress them before you get response. All of these things are very individual and players will hunt down certain behaviors and they will have things that they want and you see some players with split keyboards where they want their left and their right hand angled differently and the size of the keyboard, the keycaps themselves how they feel tactically. All of these things, there’s very, very clear individualization a given player’s best setup is clearly not the same as another players.
Dr. Stephen Seiler 25:15
Well I can tell you are in the world of these students now because they are chatting it up on these different details. So I’m pretty sure they can confirm, but I think you’re our guest lecturer that is most into and knowledgeable of their world. So I think they’re giving you some props just through the chat of what different comments on what you’re saying. Very good, so thumbs up. You’re in their world man. Now in a way I cannot be.
Future Olympics and Zwift
Jordan Rapp 25:57
I mean, but see it’s interesting. I think this is where I think we talk about the next Olympic sports. I think almost everybody says that at some point video games will be in the Olympics and I think Zwift is a very comfortable sort of first step for a lot of the IOC and UCI. There’s a lot of UCI sports World Championship, but it’s interesting because it started out being very much like this is cycling on the road, but in a virtual world.
Dr. Stephen Seiler 26:29
Gamified a little bit.
Jordan Rapp 26:31
Yeah. There was an interesting thing and so this is that it’s over indexed right now on pure physiology. The skill aspect of Zwift is not yet there, but…
Dr. Stephen Seiler 26:46
There’s no steering, although there is an experimental steering thing, but it’s not an advantage. Yeah.
Jordan Rapp 26:52
But you definitely see the draft. I know from the data that the best Zwift riders for a similar performance will put out 20 to 25% less power for a similar result than inexperienced Zwift riders because they know how to feel the draft in the absence of feedback like wind in your face because they actually they ride so much. I think the timing and use of power-ups I think this was a big thing with the UCI was will you they keep the power-ups in the game for esports because it’s like that’s not cycling, but it’s very much Zwift and they kept them and I think it was great and one of the things that I’m most proud of my contributions is that I introduced and basically did the development on the sort of the last couple new power-ups. I came and they had the existing power-ups which was the aero boost, the feather which reduces your weight and then there were some others, but then I reprogrammed they had a what’s called the burrito, which was basically made you undraftable, but initially it basically only affected riders behind you and so if you and I were riding next to each other and I enabled the reader, it made me undraftable, but it didn’t really matter because they were just drafting you and so I made it under the hood change that it basically broke drafting within an area of effect and then basically we made it reintroduce that to racing and then the other one
Dr. Stephen Seiler 28:22
Made a more powerful power up, a better power up.
Jordan Rapp 28:26
It was the ghost power-up that made riders invisible. It was interesting because it got panned really hard when I like when we first roll it. People were like that’s gonna be totally irrelevant and then you saw people that actually figured out how to use it in useful ways and there were some races that were actually decided based on clever use of the ghost power and I think Zwift’s biggest opportunity is to basically become more of its own discipline. That it’s like Zwift racing it’s not a simulation. It is not road racing in a virtual world. I think in order to become an engaging cycling discipline and specifically Zwift I know has goals of and I think the UCI has goals of esports being on. Yes and the neat thing about the ghost was you were draftable and you could draft, but no one could see you.
Dr. Stephen Seiler 29:24
For 10 seconds. Isn’t it like 10 or 15? 10?
Jordan Rapp 29:27
Yeah, 10 seconds and we’ll talk more about sort of tuning. I think you know making it so that these things are all more tunable. How strong they are? How long they last? All of that stuff. I think that’s another thing that Zwift has not done much. Where we talked a little bit early on about with Apex, you always want certain characters to be better, but you want that to be ever changing and I think with Zwift there hasn’t been yet enough sort of rotation within the power-ups of making certain things. I mean certainly the aero power-up was way too strong when I first started and one of the first changes I made was to have… The undraftable power-up, there’s no drafting for anyone within a five-meter radius. So it’s meant for solo guys that want to make a breakaway because yeah.
Dr. Stephen Seiler 30:16
It’s giving you a few seconds to try to break away because for those who don’t cycle, the effect of the draft particularly in Zwift is very powerful. So it is really hard to break away in Zwift compared to I would say on the road. Would you agree?
Jordan Rapp 30:32
Although, it’s interesting I think one of the other sort of less obvious contribution that I made was that when I started, all of Zwift’s physics were hard-coded and so they were always the same and one of the first projects that I led was they are now they are parameterized. So in every race the physics can be different to basically enable certain behavior and so we did a lot of experimentation and the physics that you race on in Zwift is almost certainly different than what the World Champ championships gets raced on, but like I think that that those physics should be changed much more often than they are because I think that’s something like in Apex, we’re constantly tuning. I think basically we know that you cannot find balance. You don’t actually want to find balance. Basically you want to find an imbalance that is still fair, but is engaging and then you want to change that imbalance to something else, but I think Zwift is sort of on the hunt. They want to find balance and then keep it. I’m like no, no that’s not that’s not the way that you make games. The way that you make games is that you’re sort of… They’re always broken, they’re just broken in different ways and it’s fine.
Creating Change Affects Any Population
Dr. Stephen Seiler 31:48
I gotta ask you Jordan because is the psychology of the endurance athlete who’s very process oriented wants predictability in the way things work, you know what I mean? I always found there’s a certain mentality that’s a generalizable mentality to endurance athletes that’s very process oriented and we don’t like surprises. Is that the same, you know what I’m saying? Do they want what you’re asking for in the gaming world or do they want it to reproduce?
Jordan Rapp 32:20
No, no I mean people in the gaming world people for sure want it, but I also think people want it in the cycling world. I think you see this is why the Tour de France route changes every year. Why the World Championships venue changes every year and maybe it’s in part also why like track cycling is I think, which is super exciting, has in some ways maybe floundered more than it should because there isn’t enough variety. I think that’s where you’re starting to see like things with the Champions League that they’re trying to say okay this part of track cycling is predictable, but here let’s show you all the ways in which it’s unpredictable with a lot of these overlays and sort of the data stuff.
Dr. Stephen Seiler 33:01
Well and what’s happening I think track cycling it’s like a tournament that they’ve compressed the time dramatically so that now it’s just a much better viewer experience and it’s more fatiguing for the athletes. So it’s there’s more unpredictability about how they recover and so anyway I think the champions league is quite interesting in track cycling, but I think yeah it’s more like what you’re six-hour gaming experience.
Jordan Rapp 33:27
It is, but I think that speaks to the fact that endurance athletes I think want consistency and fairness, but I don’t think they want they don’t want sameness and I like what Herman said ‘it’s an imbalanced balance’. Yes. That’s what we’re after and I think that’s where Zwift has the most opportunity as a platform to grow. that was a lot of what I tried to work on was basically giving them more tools to basically make an imbalanced balance.
Dr. Stephen Seiler 34:02
What’s the current membership of Zwift? Do you have any idea?
Jordan Rapp 34:06
I think it’s well over a million. They’re private companies, so they don’t share and I don’t like to ask now that I don’t work there because it’s like you know
Dr. Stephen Seiler 34:16
Right, so how would that compare with say Apex Legends in terms of.
Jordan Rapp 34:21
Yeah so I mean Apex is substantially bigger right. I think we are like 50-ish million users, active users and well over 100 million registered players and stuff like that, but yeah I mean here’s a good segue. I think when we had our recent tournament, the ALGS Championships which is, theres a two-million-dollar prize purse and we had over 600,000 like viewers on Twitch which is basically which is the dominant video platform right now and so this is sort of brings us back to right the controller discussion. Controller has aim assist, but sports rules have always favored and promoted certain play patterns and I think you see like the NBA is a great example here. You can see over time from 1998 to 2018, percentage of field goals by zone. So they move the three-point line backwards and yet three-point shot percentage still went up because basically teams recruited better three-point shooters and essentially teams made the calculus that it’s higher risk of course to take a three-point shot, but you get 50 more points. So like that’s a really good payoff and so I think teams basically just did the math and so the game itself has become biased towards three-point shooting.
It’d be interesting to see whether or not the NBA sort of makes decisions to try and bring it back down or do people like that, but I think for all the complaints around, oh yeah controller, that sort of implies that there was like a rightness and a wrongness and I think it’s clearly not right. It’s like what does the game decide and so the NBA I think consciously or unconsciously has made the decision that they will set up a structure for the game that will reward three-point shooting. Some of it may have been intentional and some of it may have been unintentional. I think that sport like big sporting leagues are often less intentional than game because game design is a very clear discipline. Again sports I think it’s more reactionary. I don’t think the NBA sort of sets out and says like oh ‘we want to have an increase or a decrease’. I think they’re sort of like we want more three-point shooting or we want closer games or games that are further apart.
Dr. Stephen Seiler 36:58
I compare sports with it’s a Darwinistic environment. I’m into evolutionary theory and so forth and if I watch a track and field competition, I can just look at the athlete and with about at least 80-85% accuracy I can tell exactly what event they do just by looking at them. There is such a Darwinian kind of effect on if it’s a high jumper they’re going to have this morphology. If it’s a sprinter it’s going to look like this and da-da and so I think any rules change, any kind of thing will fairly rapidly just filter through the entire population of performers and start tweaking what bubbles out of it and in rowing back in the when was it ’80, I think ’84. The women up into ’84 so the women were racing a thousand meters instead of 2,000 meters and that’s a three and a half of it a minute goes up to six minutes in that range, well that was enough to completely change the morphology of the athletes. The women, they were they were bringing in these former power athletes and putting them in the boat and then they became endurance athletes, pure endurance athletes. So these tweaks and that’s a big tweak, but little tweaks still have a have effects on any population of performers. Changing the rules of the game will change all kinds of characteristics of the athlete.
Jordan Rapp 38:35
Oh, I mean I think backstroke was one of the most interesting. Once swimmers discovered that going underwater was faster than going above the water and then before FINA put in the 15-meter rule the best backstroker was the person who could hold their breath the longest and they talked about having the submarine wars where essentially the hundred backstroke was you ended up basically taking two or three strokes where they would go almost the whole length of the pool underwater, take a couple strokes flip and then go almost the whole length of the pool underwater and FINA was like well this isn’t backstroke.
Dr. Stephen Seiler 39:08
This what you did. They’re just solving the problem, whatever problem you gave them.
Jordan Rapp 39:14
Yeah and so I think you can see that in game, in traditional video games that is a much more sort of known quantity, like oh yeah we’re gonna do this and it will potentially do this, but there’s also going to be a bunch of unknown effects whereas I think sports is much more ‘well we should change this for this reason’. I think the knock-on effects I think are less clear, it’s more of a bureaucratic process and I think there aren’t enough sort of game designers working in traditional sports, which is why I think esports is really such an exciting field because you have all these people sort of thinking about game systems and now you’re merging that with sort of the spectacle and the excitement.
Dr. Stephen Seiler 39:57
So they have the game theory understanding more than…
Making a “Meta”
Jordan Rapp 40:01
Right, which is what the term for in gaming is we call it a meta, where certain characters are meta and then you have different weapons and then certain weapons are meta and then the combination of maps and weapons and legends and then in Apex it’s a team game, so like composition of three different legends together and the fact that each of those legends must be unique. Obviously in Apex if we were to say, okay ‘you can pick any characters and you could even pick three of the same characters’ that would be have massively different effects than you have to what’s the best composition of three different characters and then the maps change it, but this sort of interplay where this is what we’re going for.
It’s interesting when you talk with map designers of, “Okay this is what I wanted, this is why the map is this way.” And then when you actually put it in the hands of pro players and then how they are so good and they figure out all of these things so fast because I mean in part both because of how much they have played and then how much they do play. They’re playing 10, 12 hours a day. They just figure things out, where you thought like I had no idea that was a thing. I had no idea you could do that and I think it’s interesting. The controller thing is a big one, where this is a sort of unofficial top-25 ranking of the current players in Apex and the ones with dots are all on controller and so this is this argument that controller is too strong, but some of it of course is like I think is self-perpetuating that controller certainly is stronger in certain play styles and then if people believe it’s stronger then you’ll have more people playing that way.
Dr. Stephen Seiler 41:47
Yeah I guess one percentage of all players are using controllers because that’s going to impact.
Jordan Rapp 41:53
Yeah, I mean it’s maybe less applicable only because you have on consoles, if you play on a PlayStation, you play on an Xbox you must play on a controller and so you don’t have choice on every platform and so that skews it. Yeah, so it ends up becoming a little bit of a self-fulfilling prophecy and so you could say players play on controller because they think it’s stronger, which then makes the game basically be played the way that controller players prefer to play it, which is controller is clearly more advantageous in close quarters and less advantageous at long distance because it’s less precise. So it favors movement over precision and so close quarters versus far distance and so you think like is controller overpowered or is it just that players believe controller is overpowered. So they play a controller style which then of course favors controller because more fights end up happening in the way that controller players prefer to take engagements.
Dr. Stephen Seiler 42:58
Well, you just made me think of something. Well we need to take a break, but we’re flying here the time is flying. It’s already been an hour, but I think everybody’s having a good time listening. Normally we take a quick break at the top of the hour, but I had this one question. Now let’s see if I remember what my question was. I think I forgot it. So guys do you want to take a break for 5-10 minutes? It’s six o’clock now or do you guys want to roll on and finish early? Any feedback on that because this may be kind of a conversation that. Yeah, toilet break would be appreciated. Let’s do that. Biology break and we’ll come back in like in 10 minutes okay and then we’ll try to finish early. I’ll take silence as an agreement. All right, this is fun Jordan. I’m learning a lot and I think they understand that you know the game or you know the world you’re working in, so that’s cool.
Jordan Rapp 44:11
Yeah I would never have expected to sort of end up where I did, but I mean I played games you know my whole, I was a gamer before I was an athlete.
Dr. Stephen Seiler 44:21
Okay. Yeah, yeah.
Jordan Rapp 44:23
Then I think I played a lot of video games growing up. It always…
Dr. Stephen Seiler 44:29
Yeah you’re enough younger than me that probably you were in a different set of, I was so old that the games that I played in my youth were two-dimensional. There were no.
Jordan Rapp 44:40
Yeah I mean mine too. I mean you know Super Mario. Yeah so to OTAR. I work at respawn. I’ve worked there for two years on Apex for that whole time. So yeah, but I mean it’s interesting.
Dr. Stephen Seiler 44:58
Is Apex, is it a new player in the market, relatively speaking?
Jordan Rapp 45:04
Yeah I mean, well yes and no. I think you compare to something like Call of Duty which has been around for much longer. League of Legends. I mean league is maybe different, not a great example because league is not a first-person shooter, but say compared to Call of Duty which is a first-person shooter or OverWatch. OverWatch and Call of Duty are probably the more established ones, maybe along with battlefield, but Apex now I mean we’re going into our fourth year so it’s hard to think that like we’re a new game, but I think if you could look like over the historical arc of first-person shooters compared with like Counter-Strike or Call of Duty.
Dr. Stephen Seiler 45:45
You say you’ve played Nintendo as if that’s going to make you connect to me, but see I’m even before that or Super Mario. I did play I did buy a Nintendo, but I was playing Space Invaders and Pac-Man.
Jordan Rapp 46:01
I had an Atari right, that was my first.
Dr. Stephen Seiler 46:04
Yeah that’s what I meant. I had an Atari.
Jordan Rapp 46:07
Like single a joystick and one input button. Playing a little bit more than pong, but yeah. I mean
Dr. Stephen Seiler 46:16
I was so good at Space Invaders because it was just once you learned it, there was no, I mean it was just a pattern that you just figured out. I mean I don’t know how many hundred screens I could go through before I just gave up. I was just tired, I was exhausted. Those are early games that’s what it was just once you figured out the patterns it then became just an endurance competition.
Jordan Rapp 46:45
Yeah, no I mean it’s interesting. I certainly grew up in the era of you play against the computer and I still much prefer that. I think it’s interesting right to work on a first person shooter, a game that is humans versus humans because I don’t love…playing against another person is always hard and I think it’s nice. One of the things that I like about gaming is that you can turn your brain off a bit and relax. A game you can set it on hard mode or whatever and different games have different sort of ways of tuning that sometimes it’s stupid and sometimes it’s quite good, but I do prefer playing against the computer more than playing against other people because other people are very smart whether they’re sort of intellectually smart or just like game smart, it doesn’t matter, but it’s always hard. The games industry I hope you still see great sort of player, what we call PVE, AI games I think you know God of War and that type of stuff, but it’s interesting that this is sort of the current like what is engaging to people is now and I certainly think if you’re talking about Esports, nobody really wants to watch player versus computer. I think certainly Esports wouldn’t have happened without this sort of a rise.
Dr. Stephen Seiler 48:19
One thing about Zwift that’s different and it does play to our worst human features. It’s also a training platform. It has group rides and so you can just solo work out. So it has that traditional training component to it, however the problem with Zwift or the challenge with Zwift is our own egos in that we join group rides which are supposed to be that well my goal today is to keep my lactate low, my heart rate’s here and I’m going to do this for three hours and it’s going to be. That is very difficult to achieve in Zwift because of this, you tend to want if someone cycles past you tend to want to jump on their wheel and slowly you get this acceleration process that happens. So I don’t know gaming, the gaming, electronic gaming doesn’t maybe have that because you don’t have the, there’s no such thing as just training is there?
Jordan Rapp 49:27
Yeah I mean at a certain extent no. I mean although, I think games are looking for ways to bring that about and so you think of something like Apex or you imagine a game like Call of Duty is maybe a better example because it has an actual campaign mode where you can play and the movement is all the same and the weapons are all the same, but you’re playing against the computer. So that’s the question is that training or not because that gets to the point of we’re talking about specificity. Is that actually the same? I mean it’s interesting that you say that there isn’t the question of balance for casuals versus professionals. I think there is we just don’t really think about it. W is a basketball basket 10 feet? It’s clearly better for probably most people if it was like eight feet, but we just sort of have accepted that oh regular people just play on the same court. A basketball court is this size and the baskets are this height, but it’s basically it’s sort of geared very much towards the professional game, rather than or you could say that maybe the professional game is basically bound by the limits of what we have decided is sort of acceptable at college or highschool.
Dr. Stephen Seiler 50:52
I think also there is just a motor performance aspect of it, is that if you change the rules of something tremendously then it’s one issue, but if you just change it a little like if you just sync the goal down a few centimeters or you make some adjustment based on height, it actually is a much more challenging motor adaptation because then you get interference problems. If you have athletes that are used to a certain height and then you switch you change it just a little you’ll just blow them apart because they’re so used to a certain thing. So I think that’s part of the reason that standardization has just maintained.
Jordan Rapp 51:34
It’s interesting where you do see departures though, so there’s a couple interesting examples. One is gearing on Junior bikes. UCI sets a maximum gear size so that kids can’t race on gears that are too big, which of course then changes the dynamic of racing because you can’t sort of just power away because you’re capped a little bit and so I think that’s an interesting one where the cycling has sort of change things on the youth side, but the one of the things that I think is the most interesting example of changing is that high school and college football, American Football play with the same size ball, but the NFL, the ball is bigger. I mean and noticeably so and yet there isn’t sort of much talk about whether or not a player oh will they have issue catching an NFL ball or will a quarterback.
Dr. Stephen Seiler 52:35
I didn’t even know the ball was bigger. That’s interesting. In soccer, which with the World Cup starting, we do know there’s ball size differences male, female; junior versus senior. So they do use that as to try to respond to growth.
Jordan Rapp 52:55
I think that gets into your motor pattern thing. Is a ball that is proportional to the athlete better or is it better that you always have the same ball so that you then basically know how a ball of that size reacts, responds.
Dr. Stephen Seiler 53:11
I would suspect your adaptation would be smoother whereas these kids that every time they change age group when the ball changes then they’ve got to go through an adaptive process whereas if they were always using the same ball they would be a more smooth function.
Jordan Rapp 53:28
Yeah I mean baseball. I think baseballs are always the same.
Dr. Stephen Seiler 53:30
A baseball is a baseball, so there’s different ways of solving the problem or trying to solve it. In rowing I rode when they were just switching from the making blade to the hatchet and the transition to the different and one of the things that happened was back injuries.
Jordan Rapp 53:52
Yes. Oh, yeah I mean hatchet’s way more load.
Dr. Stephen Seiler 53:55
Now junior rowers or at least very young they they’re not allowed to use the hatchets in a lot of countries, so they try to limit that as a protective measure.
Jordan Rapp 54:07
Yeah, I mean water polo. I mean ball size. I mean it’s interesting because I think there’s way sports, it’s just less obvious because I think in gaming we talk about all this stuff because in a large part because it’s so dynamic and literally the changing in it is you change lines of code. It’s much easier to basically change the size of a virtual ball than it is to okay all water. Polo balls, all soccer balls all over the world are now half a centimeter smaller in diameter. That’s millions of dollars to change.
Dr. Stephen Seiler 54:46
That’s a good point is gaming you can tweak things. It’s a constant experiment to see, I guess you can see results in in days or weeks in the data, in the metadata. How some small changes impact.
Jordan Rapp 55:02
Yeah and of course you also have, not only can you make the changes faster, it’s much easier to instrument. To basically gather data on those changes. So you can respond and react. Yeah I mean it’s true you could certainly break your own game if you make too big a change.
Dr. Stephen Seiler 55:22
Interesting. That’s fascinating just an interesting experimental arena for those of us who are scientists who are thinking oh this would be fun to be in control of a world just make small tweaks and then measure the effect of it. So all right, well is everybody back. I don’t know if I see every, I can’t tell if people are actually back, but I think we’ll just kind of start rolling again.
Twitch.TV and the Rise of Streaming
Jordan Rapp 55:50
I mean I think we have a good segue because I think sort of on that data, I think the Champions League in track is a great example of starting to sort of do a lot of esports type things, but in an existing sport. How do you change track cycling without changing track cycling? I think you start to look at things like how do you make the data part of the game and I think you certainly saw again maybe positive or at least sort of neutral effects of COVID was that sort of this type of engagement of sort of streaming and this instant feedback and a lot of the data being a part of the game really came about. So you can see I mean twitch essentially doubled in size overnight because of COVID right and like has just kept growing.
Dr. Stephen Seiler 56:54
It was good for Zwift, I’m sure.
Jordan Rapp 56:57
Oh yeah massive. I mean there certainly is fall off now, but I think it’s clear video games and the idea of spectating video games, it’s a little bit like getting a degree online like the idea that you would spectate a video game is totally normal now in the same way that oh yeah you’re getting a degree online. I think totally normal. You’re starting to see right like high school, collegiate, esports programs are coming out, it can be lucrative and I think it certainly has changed the way that even regular sports. I think you saw something I don’t know if you remember when they had the NHL where they had the puck that had the streak on it.
Dr. Stephen Seiler 57:41
Oh yeah, yeah. It was terrible it was a failed experiment, but they tried. In the NFL we have a lasting effect and that is the yard, the gain marker. The first down marker. We’re so used to it that we assume that athletes are seeing that same yellow line on the field, but of course they’re not
Jordan Rapp 58:01
So I think that certainly in that way twitch has changed not only esports, but I think it’s changed the way that people think about spectating sports. I think the changes that you see in the Champions League in track cycling probably don’t happen without that ecosystem sort of just exploding over the past couple years because I think data as an integral part of the spectating experience, that’s just an expectation now. Players want that and then viewers want that. Baseball was always great with this. I used to be a big, I never really played baseball, but I love baseball because it’s such a data-rich sport and my favorite way to sort of spectate baseball was on the radio because when you don’t have anything to see you have the announcers they have to fill the time and they filled it with basically like statistics and data and now of course that’s all visual. So I think all of this, yeah Herman the streak on the puck was crazy. So basically the NHL decided that they would do a vapor trail on the puck and so the faster it went the brighter the streak was and it was the idea was that people on TV could see the puck and it was, I don’t know that that was ever actually a problem and yeah people hated it because it was very messy and not very well done I think probably now CGI is so much better that they could.
Dr. Stephen Seiler 59:41
Well, it was in real time. It was not yeah post-production, it was in real time.
Jordan Rapp 59:46
Yeah I know. It was not good, but I think it could be better and I certainly think people learned a lot from that and I think now you see sort of learning from twitch and I think one of the big things that we at Respawn did in partnership with ALGS was actually partnered with Twitch to basically do this thing called multi-view where you could watch four different screens yourself on Twitch at once which was crazy. So you could watch the main broadcast that you could follow all your own individual teams, but I think it is this virtuous cycle where people are watching and then the way that people want to watch then ends up changing the game itself. I think that’s another thing that is I think it would be interesting to see how sort of the esports influence on traditional sports, how that continues to grow because I think it’s certainly we haven’t seen an end and I think sort of as a last kind of talking point.
Digital Doping and Anti-Cheating Solutions
Jordan Rapp 1:00:44
I think we should talk about cheating because esports, but hackers and cheaters can totally ruin a game and I think because the consequences are so high, the video game industry and when I was racing I was the athlete liaison for IRONMAN Tuwada and so I would say the difference in seeing how seriously video games take cheating versus how seriously I would perceive the sporting world takes cheating is dramatically different. I think, but a big part of this is because of how obvious it is. Video game hackers are obviously cheating, they have perfect aim and so what happens is they undermine sort of the spectacle of the game.
Dr. Stephen Seiler 1:01:34
How do they do that? How do they cheat.
Jordan Rapp 1:01:38
I mean there are way more ways to cheat when it’s just bits and bytes. I mean there’s aimbots that will basically use machine learning to actually center your crosshairs and stuff like that. You can actually edit values in memory. You can send bad data. I mean the ways in which you can…
Dr. Stephen Seiler 1:01:56
Okay, but you need to have programming coding skills and so forth I guess.
Jordan Rapp 1:02:01
Oh, yeah. I mean, but I mean in many cases you just buy it. There’s online marketplaces to buy cheats in the same way there’s online
Dr. Stephen Seiler 1:02:07
Just like buying steroids, okay.
Jordan Rapp 1:02:11
Exactly, right. Traditional anti-doping, even when it was lax, isn’t as obviously ineffective. Even you look at somebody like Pierna Reese, hematocrit of 60 plus. It’s not like he wasn’t riding a bike, he was still had to ride a bike. It was still hard, he still lost sometimes and so humans are still human. Even if they’re on drugs whereas sort of humans in video games that are cheating, they’re no longer recognizing people as players.
Dr. Stephen Seiler 1:02:37
If you see it very clearly, okay.
Jordan Rapp 1:02:39
Yeah and so the consequences are so much higher, but I think you’re starting…
Dr. Stephen Seiler 1:02:45
What happens if they get detected? Do they get banned from the game or how’s that?
Jordan Rapp 1:02:49
Yeah, I mean it’s an ongoing battle. Trying to detect them faster and faster and faster and then trying to prevent them, but yeah there’s so much for this and it’s a huge industry on both sides. We have a whole dedicated team, we basically have our own sort of internal use adawata and then there’s sort of this whole balco on the other side selling the drugs and all that and it’s this escalating battle back and forth, but I think what’s interesting especially from a physiology standpoint is how they start to come back in.
So one of the projects that I led at Zwift was basically the formation bringing back as an official arm what was called ZADA which was the Zwift Analysis and Data Accuracy team. I was the internal lead for that team and it was basically how you detect people cheating in Zwift and there was a lot of work around sort of pattern recognition and sort of physiological buckets. Are you depleting your W Prime and then sprinting at a level that should be impossible based on what your W Prime value should be at like critical power, but it’s interesting to see that because in video games the data is the performance and vice versa, but this is more true now in regular sports as well because of instrumentation. The UCI really really wants pro teams power data and the pro teams really don’t want to give that data to the UCI because you think you could be so much more informed in your targeting of who you think is cheating because you know the performances that are outliers even within a field of outliers. So I think that’s interesting.
Dr. Stephen Seiler 1:04:38
I’m a guy who has used some of that a lot on Twitter and there’s only a few athletes left in the peloton that are still able. One of the greatest cyclists around is a guy named Mathieu van der Poel from the Netherlands and when he does well, he’ll put out a full FIt file of his data with heart rate, power for classic races and I’ve gone in and done the analysis, but a lot of the riders they’ve even told me that their team will not let them they used to put stuff out, but their team has banned them.
Jordan Rapp 1:05:14
Although, Mathieu van der Poel is interesting. I think you see though the limitation of the data, I think it was he when he won the last Amstel Gold and they did a power analysis and it was so clearly obvious from that Amstel Gold file versus his previous winning Amstel Gold that the power meter was miscalibrated. He ended up doing something like 380 watts for normalized power for six hours and you were like no. A rider who is that good doesn’t become even in a span of several years doesn’t become 10 better. His peak two-hour power was something like 15 better than the previous time that he won Amstel Gold and I remember reading an analysis of his power file and it was all this is incredible and I’m like it’s not actually. It’s just clearly bad data. It’s just he didn’t zero offset it and so I think you see some of the flaws in that as well as that you have to be able to recognize bad data as bad data and not as evidence of cheating because if that Fowler file of Van der Poel’s was real. It was essentially inhuman because you’re like oh yeah this guy who is already one of the best cyclists in the world in a span of two years basically improved an absurd amount and you can’t look at that and think, if that’s good data he’s obviously cheating and I think, but then you have aerodynamic data where you could say well you can tell the data is garbage and that you can basically feel good that he’s maybe not cheating because if he’s putting out that much power to go that speed his aerodynamics are awful and yet we know that that’s also not the case.
Dr. Stephen Seiler 1:06:57
Strava does these weird things if they don’t have the actual power data it’ll make estimates that are just ridiculously stupid and then so because it doesn’t take into account they’re in a peloton.
Jordan Rapp 1:07:09
Oh, yeah. No I mean there’s a lot of that. It’s funny with e-bikes to see I’ve lost some KOMs to some guy with a Strava estimated oh yeah that guy did 900 watts for five minutes. Did you really think that he did that? I think it’s interesting sort of on that type of cheating and then you talk about motors which is sort of very like electronics coming in motors and bikes, but then you have sort of traditional doping in video games as well. ADHD and stimulant medication, a lot of that is clear and so it’s yeah you have digital doping, digital anti-cheating and all of this. I mean people we for sure had people that we suspected of cheating on Zwift and the way that we thought they were cheating was that they were riding an e-bike on the trainer because it would be almost impossible to detect and that’s where you start to get into things like patterned recognition and all of that. I mean, yeah it’s interesting you talk about hiring people that make cheats, well yes.
Dr. Stephen Seiler 1:08:16
We got to jump on this. We’re currently writing a paper on stimulants in esports. That must be for Monica nutrition, but that reminds me of some data from years ago where they gave military soldiers who did these obstacle courses, they gave them amphetamine in an experimental study and what you then what might expect is the stimulant helps them perform better, but what they ended up doing was they performed worse, but they thought they performed better. So they were hyped up, but their actual physical performance was worse because they were making mistakes and so forth. So it’s not always that a stimulant has the desired effect.
Jordan Rapp 1:09:08
No certainly not, but I think you know cheating is there’s a lot of it is mental. You believe it and in some cases, does it help because you believe that it helps? I mean I think military has certainly I think it came has come out right that doping stimulant, amphetamine doping is massively rampant among U.S. Special Forces in the Navy SEALs because it’s so grueling and especially sleep deprivation. They take these things and does it actually help and then it’s well it helps the sort of the placebo effect.
Dr. Stephen Seiler 1:09:41
At least keeps them awake and that’s probably a good way to stay alive. So I mean and yeah we could enter into that conversation in some businesses or situations like space flight or firefighting or soldiers maybe I’m going to say hey I don’t mind if you dope if that keeps you alive and keeps me safe.
Jordan Rapp 1:10:04
Yeah, no I mean, but I think this would be interesting to see how cheating is a huge problem in any sport, but I think it’d be interesting to see the way that sort of some of the practices from video games and esports do they make their way into sort of more traditional sports where you start to use data as a way of sort of identifying cheaters in a different data that’s different from sort of the oh you failed the drug test.
Dr. Stephen Seiler 1:10:34
Well it is, there have been some papers on where they’ve tried to look at performances using statistical approaches like you’re talking about saying what is the probability that improvement would happen with this athlete given their trajectory of performance in the past. So there is some of that kind of work being done, but actually being able to take it to a court and say yeah we know they cheated. They can’t do it and so anyway, but I think we’ve had a wild and good discussion of these intersections between esport and sport, what I was thinking is if there are questions, very specific questions from the listeners, from the students why don’t we do a quick Q&A and it would be kind of nice if you’d actually use your own voice and just turn on your mic and ask your question to Jordan.
Jordan Rapp 1:11:33
Or the chat. I mean whatever is whatever.
Dr. Stephen Seiler 1:11:36
Yeah if you don’t want to, but you’re welcome to actually speak. Got to think of the million-dollar question, so give yourself, but it can be a fifty thousand-dollar question. It doesn’t have to be a million-dollar question. So we are waiting with bated breath for a question. Otar has raised his hand go for Otar.
Otar (viewer) 1:12:04
Yeah so we’ve seen a race in and it’s becoming more and more valuable in gaming. Such as in Valorant where they have the attitudes placed in the innermost kernel on the PC. Effectively being able to mine every single piece of information on your computer, turning it into a safety risk if abused. What’s your thoughts on how far they have to go morally and legally to combat the cheaters?
Jordan Rapp 1:12:42
Yeah I mean, I think that I forget the name of the Valorant one, but it does.
Jordan Rapp 1:12:49
Vanguard. So the thing with Vanguard is that it’s not actually as effective as I think it should be given the level of sort of intrusiveness that corresponds to it and I think that sort of similar with kind of traditional anti-doping and I think to me I don’t think you necessarily need kernel access, but you are right. That sort of the value I think, Roblox just acquired this company, Bifron that makes a software called Hyperion which basically encrypts the binary itself so you end up doing a lot of work to sort of change where it registers and I think certainly as cheats themselves become more valuable, like anti-cheat software becomes more and more IP and yet I think you also see that there is sort of a shared community here.
The core of apex’s entity cheat is easy anti-cheat which is made by Epic, which is Fortnite and for them it’s sort of it is a business of course, but I think it’s also they could choose not to share it. They could choose to say that we’re going to use easy anti-cheat and only make it available in Fortnite, so that Fortnite itself is a safer game because that will be good, but I think epic made what I think which is the correct decision which is that it’s better that the gaming community has access to good anti-cheat software then that we alone have it and I think Roblox has sort of taken the opposite approach with acquiring Bifron and saying that Hyperion is no longer going to be supported for customers. It’s only for Roblox and I think the more that you can sort of take a collective approach to cheating in video games is bad for the industry the better off you’ll be, but I think it’s hard because at the same time it’s the most effective anti-cheat is going to be specific to your game because you will know the vectors for cheating and then you’ll know how players can abuse them or not.
And so I think in some cases it’s obvious, but I think in other cases it’s sort of less, but I think for sure the biggest area of growth is sort of going more in the way of physiology of saying the bio passport, where saying statistical probability, machine learning of saying it is literally impossible that this person would hit 10 headshots in a row.
Dr. Stephen Seiler 1:15:29
Or reaction time things like that. There’s limitations to how quickly you can get a signal from you brain to your muscle and things like that.
Jordan Rapp 1:15:37
So I think it’s sort of it’ll go in opposite directions. I think for esports, I think anti-cheat detection will become more grounded in the fact that the games are physical as well as sort of mental whatever, that there are very clear physical limiters on esports and I think in traditional sports it will go more towards the esports style of saying there are very clear statistical and data indicators that a performance should not be trusted that are separate from oh you failed a blood test. I think it’s too bad that the bio passport got weakened because I think the bio passport was the right step in that direction of saying we didn’t catch you for this, but over historical trends we can see that your data is bad.
Dr. Stephen Seiler 1:16:28
For those who don’t know bio passport, we’re talking blood. You can get some data on blood characteristics and even though you don’t detect the doping, you can detect the effects of the doping on the blood.
Jordan Rapp 1:16:41
Right, which is how most video game. It’s we don’t detect the cheating, we detect the sort of the byproduct of the cheating. So I think hopefully esports grabs more of the wata model of physiological and hopefully wata starts to grab more of the esports model of looking at the data.
Dr. Stephen Seiler 1:16:58
You’re predicting a kind of a mine meld here between these worlds that have each they’re bringing something to the table and Zwift which was the original topic of this particular little discussion, which I happen to have 1,500 hours in game, so that makes me fairly decent as a gamer in Zwift. I guess, but Zwift is kind of at that intersection.
Jordan Rapp 1:17:23
Yeah I mean this was really, I think the longest legacy that I have at Zwift past, was really basically bringing and sort of leading at the time that I was there the anti-cheat effort, which was entirely based around this intersection of data and physiology and so there were people that we banned for performances that were physiologically not sort of trustworthy, but then also people that we banned because the data was not trustworthy. Tthey would sprint and their power would flatline and generate the same power and it’s impossible.
Dr. Stephen Seiler 1:18:05
Simon has a question, so I’m gonna, kick in Simon.
Simon (viewer) 1:18:09
Thanks I asked the question earlier in the chat, but I’m not sure if you noticed it so I’m gonna ask again: Have you thought about bringing in people who actually design and make cheats themselves to help you fight cheaters?
Jordan Rapp 1:18:27
I mean yes. Yes and also no. I think there is a little bit of the story that if you’ve seen the movie Catch Me If You Can where the FBI ends up hiring the check forger because he knows all this stuff, but I think the difference and the reason that we maybe don’t do it more is that it is not that we don’t know how players are cheating. It’s not that we don’t know the weaknesses and so from that standpoint, I think it’s not an absence of knowledge, it’s more that the games are millions and millions of lines of code and the average I think if you look at sort of historical computer science, it’s typically like well-written software is one defect per sort of about a couple thousand lines of code. So you just think the number of opportunities for exploit and I think there’s also then the question, getting into what players will tolerate in terms of invasiveness. I think when Riot introduced Vanguard they said basically in order to prevent cheating in Valorant, you must give us basically the lowest level of system access to your computer where we could basically break your whole computer so that you could play the game and players were understandably reticent about that and so I think there’s not more of an effort I think to reach out to hackers because it’s not that there is an absence of skill. It’s not that they are remarkably gifted programmers who just see these things that other people don’t see. It’s they’re in many ways quite unsophisticated and so it’s sort of it would be how often do banks hire someone who basically just robs a bank and it’s not that these people are necessarily super clever, it’s more that they’re just they’re clearly quite immoral like robbing a bank.
Dr. Stephen Seiler 1:20:29
High-risk takers, yeah.
Jordan Rapp 1:20:32
Yeah. Right and so I think you imagine even just the value of the source code as well. If someone is willing to sort of profit off undermining the integrity of the game how confident are we really that they’re not going to just steal Apex’s source code and just put it on the internet or basically just use their job to figure out how to make even more cheats. I mean this isn’t to say that we have we have engaged, it’s very common to hire players who have made mods and sometimes you could look at those mods as being cheats, but I think the difference between what sort of the modern community, which is I’m going to change the game because I want the game to be more fun or different or better is very different than I want to change the game or break the game so that basically people can cheat.
There certainly is a strong appetite to hire people that are interested in basically tweaking the game even if it’s in ways that maybe we don’t like ‘want’, but less of an appetite to hire people who are really focused around sort of exploits.
Dr. Stephen Seiler 1:21:36
You want innovation, but with a certain degree of a moral framework.
Jordan Rapp 1:21:42
Yeah. I mean you think of somebody like a Graham O’Bree, who in cycling world, he certainly pushed the envelope on what was allowed under the rules and I think he did a lot of things that the UCI didn’t like, but he never and so you could say he was flexing the rules, but not sort of clearly breaking them.
Dr. Stephen Seiler 1:22:05
The rules were very unclear back then. The framework was loose.
Jordan Rapp 1:22:10
I think this is one of the things that’s may be most disappointing about the Team Sky thing, where they sort of talked about all these marginal gains and then at the end of it the parliamentary inquiry was oh but you guys are actually all on corticosteroids. So all of this talk of sort of special beds and everyone has a blender because they do these recovery shakes and it’s oh, but you’re also on drugs that have been abused for the past like 30 plus years in the program.
Dr. Stephen Seiler 1:22:33
Well I think I would almost give them credit for doing a very a smart kind of hand trick where they’re saying look over here on my right hand of all these marginal gains, well on the left hand actually I’m doing some fairly straightforward things that are illegal or at least on the edge. Plus I’m paying millions of dollars, which is that’s not a marginal gain. That’s just stab in the back. Yeah.
Jordan Rapp 1:23:02
No, but I mean I think there’s always an opportunity I think to learn from the way that people are exploiting and I think you saw with Balco or some of these labs, should WADA have engaged more with the way somebody like that. I don’t know and I mean even you could say that we at Apex we’re wrong for not engaging more with the cheaters because they might give us access, we might have flaws that we don’t sort of realize and I think that’s a tough, I think there’s no clear, it’s a moral calculus at that point of do you want to hire somebody who you don’t necessarily agree with their morals for the greater good and yeah.
Dr. Stephen Seiler 1:23:56
Jordan there’s a good question here. Katie asked or Katrina asks: do you have any numbers on boys versus girl players? how many boys and how many girl gamers? Is it a big difference?
Jordan Rapp 1:24:08
So I mean I we have numbers, rough numbers for Apex and then there’s the gaming industry. I would say interesting, the biggest growth over the pandemic was in female gamers. Amazon has more data. People like joining Twitch and people EA has more, but for sure I mean I could tell you in the pro league there were it was 120 players and it was 120 men, men, boys you know whatever you wanna. So which I found quite I think that’s disappointing. Now certainly first person shooters as a genre seem to appeal more to boys, men, but like Apex as a game diversity is super important to us. I think we’re basically 50/50 in terms of Legends male to female. The narrative arc is a strong part of the Apex game and there’s transgender characters, there’s non-binary characters. There are LGBT, that is a big part. Gibraltar is a gay man. Bangalore is a gay woman. We want to have characters that are real and yet it is certainly disappointing that the diversity of the characters in the game is so much higher than the diversity of our player base, but I mean part of the reason that we push on diversity in the game is to hopefully sort of change that because. I mean video games as an industry is basically has traditionally been young men and we’re trying very hard to make it sort of more inclusive and I think certainly over the pandemic the rise among in the gaming community of female gamers was dramatically more. It was sort of like boys and men already knew that playing video games was a reasonable way to connect with people without sort of social.
Dr. Stephen Seiler 1:26:20
Without being able to go outside.
Jordan Rapp 1:26:23
Dr. Stephen Seiler 1:26:24
Are there games or gaming, game-like, esport-ish type activities where there is an overweight of females? Are there any genres, any specific kinds of these esports activities where the pendulum has swung the other direction?
Jordan Rapp 1:26:45
I there may be. I could not give you an informed answer there. My intuition says probably not, but I think but I think Zwift certainly has done a very good job. From the outset there was basically equal coverage, equal prize money for men and women.
Dr. Stephen Seiler 1:27:08
But there is not equal participation on Zwift.
Jordan Rapp 1:27:11
No, no, no. I mean it is certainly very skewed, heavily male, but I think again Zwift is a big believer in, I mean Zwift is the title sponsor of the Tour de France, the women’s Tour de France. Yeah brought that back. So I think video games, the accessibility barrier isn’t there, but there still is certainly a cultural barrier. We have a lot of active chat monitoring and that type of thing that there’s a lot of toxicity. People behind keyboards are just not very nice and so I think that’s not a super welcoming environment.
I mean same with Twitch’s own numbers. I think there’s basically among the top 100 most subscribed channels of individual Twitch broadcasters, there is one female because there were two, but one of them decided to take a break. So now there’s one. I mean which is not great and I think even you see there is a lot of sort of quasi exploitation. The hot tub streams. That the female streamers are popular because they are sort of traditionally attractive and they leverage that more than because they are very good at games.
Dr. Stephen Seiler 1:28:40
My guilty pleasure is I watch a lot of mixed martial arts and UFC in particular, the United, Unlimited Fighting. Whatever it’s called Championship and even there are quite good and female commentators. So you’re even behind the most kind of traditionally male thing there which is beating the crap out of somebody.
Jordan Rapp 1:29:04
Well, I mean the Apex, the broadcast team for Apex is basically 50/50. It’s just the player base. I mean from that standpoint I would say, not behind the UFC in terms of who’s paying to watch Ronda Rousey versus Conor McGregor. I think almost every sport is still a laggard in terms of participation and engagement. Ronda Rousey never has, there has not been equivalent of the Conor McGregor sort of Floyd Mayweather. There’s nobody getting top top earners. I mean the Williams sisters were remarkable, but I think even as much as Serena Williams made I think in her career they said I think twice as much as second most, which was her sister Venus and then they’re that much higher than everyone else. Even they pale in comparison with a Cristiano Ronaldo type of. There’s no billion dollar female athlete yet. Yeah. Rose Namajunas, she’s quite popular. Although she’s kind of had a layoff right now. Well guys this has been good. Let’s say one more question and then we’ll call it a night, it’s been fun. Any last question? Yeah, Katrina. Katie go.
Katie (viewer) 1:30:26
Thank you and I wanted to ask about the report system in Apex because as a female player you hear a lot of shit and that can kill like your motivation to play the games and you have the report button, but you never know how seriously the company takes the reports so how many times do you have to report a guy for telling you to show your through the whole game before he gets banned?
Jordan Rapp 1:31:02
I mean so I will say that every report get goes through, I mean it’s not oh we just throw them away. I mean certainly the frequency of reports there’s all kinds of algorithms to detect, but again there’s certain, there’s keywords and all of that stuff where you report someone and then the system looks at what they said and then there’s certain things that trigger a more aggressive response than others. I mean I think it’s very hard because you imagine the volume is just massive. There’s literally no way that you could have an actual person read every report, and so any automated system is going to have flaws.
And I think it’s interesting we had a discussion about this where there were some players, they were a group of players from Mexico and they received a ban because they were talking on, they basically were pointing to a place on this basically saying we’re gonna go here and then they used the word negro or negro, black in Spanish. But like the game detected that they were using the word negro as a slur and then so they said like, “We’re just saying the color black.” But of course we have tons of, we have players where they will then give the report of that word is then being used in a very derogatory fashion, and how does the system intelligently differentiate between those two? And you basically, you go too far one way and then you sort of soften it and then you go too far the other way and I mean it’s awful.
And I mean I would say that we try very hard to basically make it a good environment for players, but we could certainly do more, do better. And I think moderation is a massive challenge because it gets into sort of the frontiers of language analysis and sort of actual AI of this is someone being derogatory because it’s obvious when a person sees that, oh yeah, this person is clearly being derogatory and they should be banned, but how do you make sure that a human sees that when the volume of sort of material being generated for review is impossible? I think this is a reckoning that the industry as a sort of the entire tech industry, like Twitter or Facebook, all of these things yeah everybody has this problem.
Dr. Stephen Seiler 1:33:43
Well, we have AI research at our university they’re particularly good in language and some of the work I do with professional cyclists, we’re trying to look at text analysis, but the problem is that any given team, there’s multiple languages that are being mixed into each other while they’re communicating because you’ve got a Norwegian cyclist that blends English and Norwegian in the same sentences, its very normal.
Jordan Rapp 1:34:14
Yeah that’s what happened with the players in Mexico. They were they mixed English and Spanish so that it was not obvious when they used the word negro that they were using it in Spanish.
Dr. Stephen Seiler 1:34:24
So it’s a super tough challenge.
Jordan Rapp 1:34:26
Yeah, but I think to Siemen’s point. There are actual people at the end of this and the depression is real. Facebook moderators, there has been,there is actual study data out there that the mental toll on Facebook moderators is massive. That you can only do it for so long and that it’s probably, it is almost certainly going to permanently damage your psyche to some extent because you just see and hear things. I mean especially on Facebook for pictures and videos and things like that need to be moderated, I mean you will see things that are unimaginable.
Dr. Stephen Seiler 1:35:06
Katrina do you have your hand up again or was it you just haven’t taken your hand down? You have a new question?
Katrina (viewer) 1:35:13
Dr. Stephen Seiler 1:35:14
Oh, okay no worries. Well, guys it’s been fun. Jordan thank you so much. I have learned tremendously about gaming from you and from the class, but this has been great and we’ve recorded this so people will I’m sure go back and listen again some or those who weren’t able to do it. So thank you so much for joining us Jordan.
Jordan Rapp 1:35:40
Oh, it was my pleasure.
Dr. Stephen Seiler 1:35:42
Great discussion of this intersection between sport and esport and I think there was a lot of technology that was being discussed even though it was kind of infused into the discussion in a subtle way. So yeah fantastic job. Great. You are our last guest speaker for the semester and I think it was a highlight. So great job thanks again.
Jordan Rapp 1:36:05
Yeah. Well, thank you so much for having me and when I was still racing I actively followed your research and I was always a fan. I still, the sort of my favorite and I’ve shared it countless times. This sort of the Maslow’s hierarchy of needs, but for sport I think is one of the best sort of visual representations of data and good training.
Dr. Stephen Seiler 1:36:34
I still remember making that and it was just in a moment of frustration.
Jordan Rapp 1:36:38
Yeah, I mean I think well sometimes brilliance comes out of that, but yeah so I think it was for me to sort of come back into because sports is such a traditional sports is such a big part of my life and to sort of now be in this other side and to see these intersections and to see or I make video games and yet what I did for most of my adult life is to me, I think it’s still incredibly relevant. Even though on the surface it might not seem like it.
Dr. Stephen Seiler 1:37:07
No, I’m the same. I was super skeptical to teaching an esport course because I thought what the heck. I do not know a thing about it how can I possibly do anything, but it’s actually there is there are connections so I’ve enjoyed it. Anyway, all right guys. You have an assignment. Remember December 1st you’ve got the third assignment, so don’t forget that and we’ll let Jordan get on with his day.
Jordan Rapp 1:37:32
Yep and then you have a request for that your graph the Maslow’s hierarchy of needs, but for sports.
Dr. Stephen Seiler 1:37:39
Oh yeah. I’ll show that yeah. It’s actually for endurance training. Yeah I’ll do that.
Jordan Rapp 1:37:45
Okay. Well, thanks again. Have a great rest of your day and yeah thanks for so much for having me.
Dr. Stephen Seiler 1:37:51
I’m gonna try to get you back here next semester or next year at least.
Jordan Rapp 1:37:55
Yeah. Any time.
Dr. Stephen Seiler 1:37:56
All right, take care.
Jordan Rapp 1:37:58
Dr. Stephen Seiler 1:37:59
Bye. All right guys, I’ll put the link. I’ll put something out on the hierarchy. The only thing I ever learned from Maslow, but don’t forget you’ve got an assignment on the first and then it’s due and I know you’ll do a good job and then we you’ve got your last exam or your exam on the 15th as I recall and so then you’ll be done. I hope you enjoyed this, good job take care.