This is not a bug though - it’s a team at the top cheating in a pretty terrible and very clearly evidenced way, and Zwift not even recognising it’s an issue. Which probably just exposes the flaws of their base level support in recognising a major issue, but that clearly needs changing.
I am back home on Wednesday, so will detail everything then or Thursday.
We should maybe think about TrainerRoad as well. What on earth does Weightgate have to do with TR? If Zwift acquires TR (which is rumored but not confirmed, to be clear), they will be buying a complex bunch of software and intellectual property that requires upkeep, improvement, and integration with Zwift’s current code base.
So, above, we got confirmation that someone on Zwift’s tech support side was aware of the exploit as of about Jan 2021, and they claimed that Zwift was working on a fix. Eric Min himself then said that Zwift was “relatively” unaware of the exploit. I’m not sure if he meant exactly what he wrote, but there are several possibilities, and none are great. The issue could have been raised and then got lost amidst Zwift’s day to day, or else it was lost in communications from tech support to the dev team (but then we have to explain why the support person said they were working on a fix - maybe he was mistaken). Or the tech guy could have been totally mistaken, and the dev team were completely unaware of this for whatever reason. All these possibilities should raise concern about Zwift’s organizational capacity.
Zwift have already taken a long time to start integrating Zwiftpower into their own systems and procedures, and to start moving on race category enforcement and improvement. If they buy TR, their organizational capacity is going to be split in even more directions.
I don’t want to downplay Zwift’s achievements to date, but I don’t think they can handle a major acquisition, because we are already seeing indications that they already don’t have the capacity to keep on top of their current code base. If they buy TR, I would have to take that as a reason I should consider jumping ship. Also consider: Zwift have said many times that racers are a minority of their user base. I’m not contesting that as a fact. My point is that if we are a minority, they’ll probably put our needs lower down in the priority chain. The pros are probably adequately served by ZADA. They’ll likely put TrainerRoad issues, assuming they can handle those in the first place, above us.
Whoops! Edited my post correcting the link to the proper…James Eastwood for this case (sorry to the other one!)
Looking forward to reading your post on Thursday
BTW, I fiddled with this exploit after the Rapha Rising Stage 3 event this morning. HOLY TOLEDO! How come this hasn’t been fixed years ago? It was remarkably easy to do and you can do it on the fly. I am shocked I tell you! Shocked…
I sucked so bad going up Epic, I should’ve done it at the bottom instead of the top…
The GC of the Rapha Rising series is a complete joke now. A rider with 2.6wkg finishing on the podium (on Zwiftpower) at 2nd place in a cat C race between riders doing 3.5 and 3.1wkg.
He even has a screenshot where you can calculate his weight during the climb at 70kg, while his weight at the finish is 94kg.
I don’t see how on earth zwift could expect users to detect this particular cheat unless a rider was doing something particularly egregious. There weren’t even any comments made in the race when Jonathan Crain accidentally rode off the front at 2W/kg (and 25cm). I suppose the pack riders didn’t have a good look at his numbers due to the natural churn. No-one would have a chance of spotting a mere 10% change, which would easily be enough to radically change results in many cases.
Either zwift can check it, in which they should do so, or they cannot, which would be consistent with the fact that they haven’t actually detected any of those who have tested it out. You can be sure there have been many over the past week!
You’ve hung in there longer than I expected, James.
Having done The Exploit, I’m amazed at how easy it is to do. And I don’t know how it would be easily detected. A 10% change would be pretty effective (guessing here because I don’t intend on using it going UP a hill anytime zoon) and fairly difficult to detect.
Whilst I’ve said I don’t care about people doing weight doping and the like, this would be a different kettle of fish IMHO. If you’re in my category, currently C, the playing field should be level. Yeah, you might not be that fast IRL but all I care about is having a good competitive event. Altering weight in the event to climb faster is solidly cheating. This Exploit is just that.
Regarding the Rapha event… hasn’t it been said elsewhere that this is a group ride? Can you cheat on a group ride? Hasn’t it been said previously that weight doping can in fact be a feature to consider using for tuning a group ride’s difficulty?
Having not tried this exploit, I imagine that once you’ve started a big climb like reverse epic Kom in a race, you could subtly reduce your weight by say 2Kg every 30secs or so and riders around you who are able to view stats down the right while suffering won’t notice.
What possible legitimate reason would coders allow race critical, on the fly rider detail changes? This beggars belief.
It’s not even that, they’d have to fan view to look at your watts and w/kg and divide in their heads, while knowing what your weight was supposed to be. It’s ludicrous that people actually suggested this with a straight face. Especially if they can already check it themselves (yeah right).
You absolutely can cheat, because many people do group rides competitively.
If you aren’t able to keep up with the group, people may be willing to accept you weight doping to be able to stay with the group. It’s similar logic to how we might accept someone using an e-bike in a group ride. The problem is, if it’s too easy to do this, then there’s no limiting principle, and I would prefer that nobody used weight doping to keep up with a group. You may just have to accept being slower than everyone.
Also, the Rapha ride is a bit of a hybrid between a group ride and a sanctioned race. It is presented as a group ride, but the instructions also say there’s an optional GC, and to sign up using your race category if you want to be a part of the GC … but you are started with only your cat members, as if it were a traditional race rather than a full mass start group ride.
(sorry to the other one!)
