More (potential) ways of cheating with Zwift

Triathlon Forum
1

Recently people way smarter than I (Ray Maker (dcrainmaker) and Shane Miller (gplama) among others) have been pontificating on the problems within Zwift as a serious e-gaming platform, especially where it concerns cheating. And during my casual YouTube scrolling tonight it occurred to me that cheating with your weight is only the start of the problems.

Cycling Hub appears to have developed some sort of test harness where they can have a rider lock into any power or cadence. At first I thought they used a Kickr or somesuch and that I’m just weak because I can’t do a loop of Watopia at 400W.

But then they had a descent down L’Alpe at 2000(!!)W:

But then they had a descent down L’Alpe at 2000(!!)W.

(The embedded player doesn’t seem to like the start time on the link. Go to 5’52")

Needless to say, that can’t be done by a human. So what I think they did is the following: They developed a little program that basically pretends to be an ANT+ power meter. This isn’t that hard; the ANT+ protocol is open so you can exactly look up what the flow of data should look like. You probably need to plug in the proper vendor IDs but that’s also not hard. You use the same libraries Zwift and TrainerRoad use to receive data on an ANT+ dongle to send data.They can probably set a power and cadence value and just start broadcasting. Zwift thinks it’s a powermeter and hey presto.

And that got me thinking - what if you refine that, and make the power controllable, maybe using a joystick or the arrow keys? You can superimpose some random noise over the user selected power to make it look more like a real cyclist, but I think there’s nothing stopping you from entering into a race with such a “powermeter”.

And then I got even more clever: What if you read in the power data of a real powermeter over bluetooth, and then send it out over ANT+ (like those protocol converter things you can buy), but allowing the user to dynamically inflate the power, basically emulating unlimited powerups? You could have various types to make it look real - you can have for example a “5 second attack” button giving you 5 seconds of 1000W, a “drop the hammer” button giving you 10 minutes of 120% of what you’re really doing, an “I suck a sprinting” button giving you 30 seconds of, I don’t know, 1300W, etc, etc.

Again, this is not hard. All you need it is a computer, an ANT+ dongle, and coding skills. If I wasn’t pathologically lazy and uninterested in cheating I could probably develop a proof of concept in a week or so.

2

Because what does someone have to gain by doing this?

Cheating in “real” video games usually brings in some sort of real world money or increased chance at getting the best items and becoming a better player. Or it is used to reduce a grind.

Of all those things, I see one being useful. Itd be nice to have a tron bike but I dont ride enough inside to get it quickly.

The other facts… ultimately no one is winning a contract or money off of this that wouldn’t be verified against real world results (which is what zada does).

3

There’s no coding skills required. There are tools out there that do this, point and click. This is exactly what they’ve used.

Doing this in isolation (disconnected from the Zwift services) is borderline ok… if it’s to see what’s fastest, what’s slowest, what gains more points / something… for “science”. Doing it while connected to them is against the signed/agreed Terms and Conditions of the service and can have you booted off.

imo, impacting anyone else’s experience in a negative way should have you booted off the service for good.

4

Because what does someone have to gain by doing this?

Cheating in “real” video games usually brings in some sort of real world money or increased chance at getting the best items and becoming a better player. Or it is used to reduce a grind.

Of all those things, I see one being useful. Itd be nice to have a tron bike but I dont ride enough inside to get it quickly.

The other facts… ultimately no one is winning a contract or money off of this that wouldn’t be verified against real world results (which is what zada does).

I don’t know, man. We’re living in a world where dentists will spend serious money on drugs which, aside from theoretically making them faster, actually physically endanger them. And they do this because they hope to get between 5th and 10th in M40-44 in the local 70.3 instead of the 27th place they got last year.

I’m pretty sure if a smart college math- or CS student sat down for two weeks and started “marketing” whatever he came up with, he could buy himself some nice Zipps for the local spring crit series within a month. Because you, the cheater, don’t have to code. You just procure a piece of software.

5

imo, impacting anyone else’s experience in a negative way should have you booted off the service for good.

Just like using real-life doping gets you kicked off the service provided by the UCI and ITU. But that doesn’t stop the dentists I mention in my post above.

Meaning: obviously you’re right. I just think it’s (another) gaping hole in the idea of using Zwift as an online e-sports platform.

6

Because what does someone have to gain by doing this?

Strava segments and ego stroking.

7

Because what does someone have to gain by doing this?

Strava segments and ego stroking.

I’d think it’s more trolling. Just messing with people for amusement.

But I only think that because I’m just unable to fathom how someone’s ego could be boosted by cheating in the manner we’re talking about. (or any manner, but at least in say drug doping there’s still some actual athletic input required so that dopers can rationalize things…I don’t see what rationalization could be used by running a script.)

8

Yep, these tools exist for a reason. Developers need tools like this to test the software as they are developing the game. Can you imagine having to hop on a bike just to test some small change you made to the Zwift code base? You’d be on the bike all day for work.

And it’s not just Zwift. Every Ant+ device needs testing which means software developers need tools like this to ‘fake’ some streaming power data.

Cheating would be easy for any mediocre developer. The way to ‘catch’ these types would be to compare it to real outside rides and preferably, road races with power data over a known course/distance/elevation/time.

Cool topic. Thanks for posting!

9

A dedicated sophisticated motivated dickhead can always figure out ways to cheat at Zwift, but there are ways to put safeguards in the game to identify not quite so sophisticated dickheads. It turns out that faking a power file isn’t as simple as many people think – it’s still possible, it’s just that it’s not simple.

10

I just typed out many words and deleted… At the end of the day, people are going to lie. I know people that lie about PRs and weekly volume (USAT winter challenge) - stupid because you can look it up and/or the 20+ hours a week they are allegedly logging only produces silly results.

Dont lie
don’t cheat

I get the frustration. I moved to Trainer Road. Not because people are dumb on Zwift but because I can just focus on the task at hand.

11

Quite frankly, I don’t give a shit what others do. They are only cheating themselves. I’ll keep working hard, raising my FTP and one day I’ll be winning races without cheating. They can cheat, feel ‘good’ about themselves for a day, but they are only cheating themselves long term. Who gives a shit about Zwift races anyways? People have too much time on their hands to complain about everything.

12

In computer security we refer to this as the trusted client problem. Relying on the client not to cheat does not work and has never worked. Video game developers have tried to make that work for over 20 years and have so far been unsuccessful. The user can do anything they want with full access to their local client.

13

Quite frankly, I don’t give a shit what others do. They are only cheating themselves. I’ll keep working hard, raising my FTP and one day I’ll be winning races without cheating. They can cheat, feel ‘good’ about themselves for a day, but they are only cheating themselves long term. Who gives a shit about Zwift races anyways? People have too much time on their hands to complain about everything.

^ what he said.

14

This is why we can’t have nice things.

Humans always find a way to cheat a system. It is nature.

15

Hey easy on the dentists! I dont think age groupers dopers doping has anything to do with what profession they are in.

16

Because what does someone have to gain by doing this?

Strava segments and ego stroking.

I’d think it’s more trolling. Just messing with people for amusement.

But I only think that because I’m just unable to fathom how someone’s ego could be boosted by cheating in the manner we’re talking about. (or any manner, but at least in say drug doping there’s still some actual athletic input required so that dopers can rationalize things…I don’t see what rationalization could be used by running a script.)

ego stroking as stated. There are messed up people in this world where it’s more important that they can say their place is some event is so important that they will cheat no matter what.

17

I agree that as the tech evolves a lot of cheating is possible. However, for me (and probably most) a Zwift race is just another workout. I honestly don’t care if someone is doing something like that - they’re only cheating themselves. Yes - it can be frustrating to see someone drop some absurd power in a Zwift race, but that’s all it is - an online adult video games that falsely boosts egos for those trying to cheat. Let’s see how they’d do in a ‘regular’ outdoor environment.

18

Hey easy on the dentists! I dont think age groupers dopers doping has anything to do with what profession they are in.

see! always some dentist trying to avoid taking the blame for their actions

just kidding just kidding

19

Because what does someone have to gain by doing this?

Cheating in “real” video games usually brings in some sort of real world money or increased chance at getting the best items and becoming a better player. Or it is used to reduce a grind.

Of all those things, I see one being useful. Itd be nice to have a tron bike but I dont ride enough inside to get it quickly.

The other facts… ultimately no one is winning a contract or money off of this that wouldn’t be verified against real world results (which is what zada does).

I thought there was prize money in the zwift KISS superleague. Anyway, if there is not, prize money is coming. For sure. And teams with salaried riders and sponsorships.

And anyway haven’t we proven that ego is more than enough of a motivator in triathlon and cycling?

20

Clearly what is needed is an ADAMS system for Zwifters that allows testers to enter your home unannounced to inspect your Zwift setup. Zwift is serious biznuss.