A Zwift racer who qualified for the 2023 UCI eSports World Championships has been sacked by his team and banned from racing for six months after the stunning ride that earned him his spot in February’s championships was found to be the result of a hacked data stream.
South African Eddy Hoole, a member of the Toyota CRYO RDT racing team, astounded commentators during a European and African continental qualifying event for eSport’s flagship event, breaking away from the bunch from the bottom of the final ramp to the finish before flying past the race’s erstwhile leaders to secure the win, and with it, an automatic place at the world championships.
Hoole’s barnstorming ride, however, raised a few eyebrows from the commentators and the almost 6,000 people tuned into the race’s livestream.
As reported by DC Rainmaker, mountain biker-turned-Zwift broadcaster Nathan Guerra even described the South African’s effort as “something we’ve almost never seen before”, and “what I thought was absolutely impossible”.
And, just like the days of turbo-charged ascents up Alpine passes in the 1990s, where if it seemed too good to be true, it probably was, Guerra’s incredulous reaction has been almost immediately vindicated.
Yesterday, Zwift published its ‘Performance Verification Board Decision’ concerning Hoole’s suspicious performance, concluding that the Toyota-backed rider purposefully cheated, and banned him from the sport for six months.
Zoning in on Hoole’s effort up the final climb, Zwift found that the South African had averaged 526 watts for the four minute 16 second effort.
Hoole’s suspicious figures
According to the report, “given the rider’s weight, this equates to a sustained average power output of approx. 8.5 W/kg, a performance that requires a VO2max of over 90 mL/min/kg.
“For comparison, these values are significantly greater than those that have been measured for Olympic Pursuit Champions and World Record Holders (average power output over 4min, approx. 7.5 W/kg) or Tour de France GC winners (VO2max, approx. 85 mL/min/kg).”
Pretty damning, then.
Crucially, and even more damning, Zwift noted that “there is no circumstantial evidence that might suggest that the rider is a globally significant World Class athlete. For example, the rider does not have any IRL [in real life] cycling (or other IRL sport) results, and their typical training load amounts to around three hours a week of low intensity cycling on Zwift.”
An independent test taken by Hoole to prove his credentials found that he could only manage an average of 400 watts during a similar effort.
Also, while the numbers from Hoole’s power meter – broadly in line with his race stats – showed that it was significantly miscalibrated, Zwift pointed out that the trainer the rider was using last month, and which recorded the data acquired by the platform, is self-calibrating (such as the Wahoo Kickr or the Tacx Neo), with a manufacturer claimedaccuracy of +/- one percent.
“Deliberate manipulation of data”
So what was going on with Hoole’s Bjarne Riis-esque numbers?
Analysing the rider’s data streams from his computer to Zwift’s servers, the platform found that one of Hoole’s data channels disconnected shortly after he joined the pen, and just minutes before the race started. Zwift also discovered that Hoole’s other races showed a similar pattern where one specific channel would disconnect just before the racing got underway – though no disconnections, interestingly enough, took place when he was simply training.
Zwift said: “It is notable that the disconnected channel normally carries analytics information about the riders system – in particular information such as the equipment that the rider is using.
“Zwift considers the absence of this analytics information to be equivalent to the presence of a masking-agent in anti-doping – for example, it would allow the rider to change their paired device from their trainer to a computer-controlled device that gave falsified power information, without such a change being recorded by Zwift’s servers.”
As DC Rainmaker noted, this could mean that the rider was inserting a device or some form of software into the middle of the data stream to change it, allowing him to artificially boost his figures.
> Zwift U-turns on ban for user who exposed weight-doping hack
When questioned about this evidence, Hoole provided no answers, but instead deleted 150 dual-recordings from Zwift Power (a site used to prove race data) and also deleted or made private all of his social media profiles.
Concluding that Hoole’s ride was the result of a “deliberate manipulation of data”, Zwift banned the South African from racing until 12 May 2023 for “bringing the sport into disrepute”.
Following Zwift’s report, the Esports Team Toyota CRYO RDT confirmed that they had “terminated their relationship with Hoole”.
The team said in a statement:
CRYO RDT / TOYOTA CRYO RDT has at its roots a goal of promoting fair and transparent sport. Since inception we have required all our riders to comply with all aspects of Zwift Esports Rules regardless of whether they race at Elite level or not. The rider in question had all elements in place and had previously been verified in elite level racing.
As can be seen from ZADA’s determination the nature of this case is such that the team would not have the means to suspect/identify/investigate circumstances such as these as they require access to Zwift Server log files and an in-depth knowledge of how to interpret these. However, it was clear to the Management Team that as a result of initial information received from ZADA without any plausible explanation from the rider there was only one decision open to us.
Esports requires a basis of trust on the part of all involved to ensure that the sport is fair, and we have and will continue to work with ZWIFT / ZADA in an effort to achieve this. We are saddened by this situation and will now study in detail ZADA’s report to establish any lessons which we can learn from it.
> World's first eSports pro cycling team sacks rider for cheating on Zwift
The recent boom of eSports and virtual racing has coincided, naturally, with a surge in the number of racers willing to cheat the system.
The most high-profile instance to date of a cyclist being sanctioned for cheating on Zwift relates to Cameron Jeffers, winner of the inaugural British eRacing national championships in 2019, the first time any national federation had staged such an event.
> Zwift national champion stripped of title because he didn’t earn the ‘Tron’ bike he rode within the game
Following the men’s race at the BT Studios at London’s Queen Elizabeth II Olympic Park, Jeffers was found to have manipulated data prior to the event to unlock a Zwift Concept Z1 bike – popularly known as a “Tron” bike – to give himself an advantage over his competitors.
He was stripped of his title, fined £250 and handed a six-month suspension from all racing, with the title awarded to James Phillips, who came second on the day.
Meanwhile, earlier this year Zwift was forced into a rapid u-turn after initially shadow banning a user that found, tested, and highlighted a known weight-doping hack.
Lucian Pollastri had received a 30-day partial ban from the platform as Zwift felt his actions were promoting the hack, but Zwift’s CEO later apologised to Luciano and vowed to fix the problem.
Add new comment
15 comments
eSports there must be some joke about Yorkshire in there.
It's an interactive training game....People take this far too serious. Get out and ride properly....
Not on those roads we have just now! BTW I never race on Zwift and certainly don't take it seriously..
This made me laugh: "the rider does not have any IRL cycling results, and their typical training load amounts to around three hours a week of low intensity cycling on Zwift.”
Three hours! So any old lazy slacker can win at Zwift. There's hope for me yet.
A few years ago, I thought about this, using a Raspberry Pi with ANT+ interface to take power, speed or HR and then adjust a fan accordingly. This led to the idea of using a second ANT+ interface to rebroadcast it to Zwift, with maybe a 10% uplift. The first idea became unnecessary as I got a Neo Bike with fans that do that, the second didn't follow because, well, it's cheating and I didn't fancy being kicked off Zwift.
Eddy A-Hoole.
I always think highlighting Cameron Jeffers in these cases is a bit harsh. His offence (not riding the hours required to earn the tron bike) is a breach of a very different rule than hacking your power output. The former was an artificial rule to make the user spend time on Zwift - nothing to do with his racing ability which would have been close to top of the field on any available Zwift bike.
Zwift Tron Bike is bogus anyway. I know zwift racing is fake but the Tron bike is an extra level of fakeness.
I tend to agree with you, and even more now that the Tron is no longer on a totally different level to other bikes (I sort of wish they'd do another Tron, or re-tweak Tron to be an Uber-bike, but that is a different problem).
Jeffers' actions didn't really, directly affect his race performance, which is a radically different thing to weight doping, or the hacking of this case.
Different rule but similar motive - he still went out of his way over a period of time to cheat the system.
How pathetic does someone have to be to make those kind of efforts just for a computer game? I dunno... And the guy is choosing to ride a static bike in his garage when it's sunny and dry outside. Needs to get a life.
Arguably it was an unfair system. Whilst I appreciate there is value for the casual user in "gamifying" training and unlocking "better" bikes, I don't understand why it's allowed to bleed into actual competitions.
Whilst I know in the real world equipment does make a difference, one of the advantages of eSports ought to be that you can eliminate that as a source of advantage, and have a "fairer" competition where the result is decided purely by the rider's performance in the race.
Also in the real world, allowing equipment differences drives innovation - team sponsors want their riders to win and so work to develop the fastest bike. But on Zwift, bikes are made faster or slower purely by a computer programmer somewhere changing some numbers by a completely arbitrary amount. So I can't see any justification for allowing it to influence competition results (certainly not from the UCI's perspective - arguably Zwift still benefit from convincing customers to ride more and spend more money to have any chance of competing at a higher level).
Zwift is no different to any mmo like World of Warcraft any one with programming knowledge can find a way to cheat. Also unless the races are done under the same roof you havnt a clue who and what is doing the riding, no doubt there is a fair few cheaters using ebikes just do a small power up when needed
You can't buy equipment on Zwift for cash, you can only use your "drops" credit built up through riding, so although it certainly might encourage riders to ride more everyone pays the same monthly sub and no more.
I meant simply maintaining their Zwift subscription in order to continue making progress, rather than switching to a cheaper alternative platform or pausing their subscription entirely over the summer months.
Ah I see - fair point.