The bigger question is probably which 37kg rider can hold 222w? If the numbers are nonsense there isn’t much need to look at weight and height.
2 Likes
The problem is that the mathematical model in zwifts code that determines CdA for a rider given height and weight is off. There is no need to do any speed testing if you have access to the code. Just look at the model in the code and compare it to the model proposed in for example this publication: Body mass scaling of projected frontal area in competitive cyclists - PubMed
1 Like
I can tell that the Zwift model comes from a scientific study 10 years younger (2011) than the one you linked. What makes this one better than any other?
Ok that is very interesting information. May I suggest you actually publish what model you use for transparancy? This would put an end to this whole discussion and all the frustration and increase the legiticamy of zwift racing.
I agree with your point, but I wouldn’t consider the riders in that race the “best Zwift racers in the world”.
Or alternatively, picking a random internet calculator out there such as below, describe if or how would Zwift’s model differ?
ht tp://www.kreuzotter.de/english/espeed.htm
The issue with a 37kg rider is absolute strength. How strong is any 81lb/37kg person you know? What 81lb person would be able to match just one Cat 1/2 attack without being dropped and done, if they could even hold onto last wheel if the pack got strung out simply pushing the pace with a couple big guys up front? Nevermind hold level with a pro rider. 81lbs can’t even ride in any significant cross/sidewind with aero wheels.
FWIW I think the person in question in the OP has either wrong weight, dodgy power readings, or it’s an adult using the kid’s account.
1 Like
But this has no bearing on the Zwift Physics, this can be anything from an adult using a kids account to a kid with a very generous trainer or anything in-between.
Yeah for sure, and I’d lean to that way too, but I don’t discount the physics being off as well.
1 Like
without passing judgement on the kid, no u23 under 40kg is capable of doing 220w for an hour with overs so the whole discussion is pointless
there was a 3x giro (or vuelta) stage winner and top 10 overall finisher who weighed 48kg, i forgot his name though. he was venezuelan. but also a fully grown adult
edit: José Rujano - Wikipedia found him. cool
3 Likes
That guy is tiny!
Unfortunately I can’t. It would probably get me fired 
1 Like
Agreed. Lot’s to address beforehand.
I don’t really like to question peoples performances, without direct contact. For obvious reasons. In this case, it’s a clear example of a flaw in user protection.
I’m happy that folks are aware of these boundary weight/height issues.
An idea I was pondering, why not run a very simple filter over all users performances, then contact the users that trigger an issue? This would obviously include sub 45kg riders, as there are numerous riders in this subset.
It can’t be thousands of users. Maybe hundreds, at best.
Just an email, polite etc. Simply to begin to verify the user. This alone would go a long way to cleaning up the user base. I believe the contact alone directly from Zwift would remove a lot of the causal cheating etc.
It would have to be worth the time of a single employee.
Just a thought.
I don’t.
I am massively faster in Zwift than in real life.
Which of us should the model be based on?
The 2011 publication year probably limits the options quite a bit, just sayin’…
1 Like
Agreed.
As I said, I do not know if the rider is legitimate. I wish someone could just confirm the data to help address these sort of issues. I have seen very young riders preforming like this on numerous occasions in Zwift. Enough to convince me that there is an issue.
Yes, it’s anecdotal.
However, from years of real life experience I have never ridden with such young children able to compete at the level that some are able to in Zwift. I simply think there is something breaking in the model.
I may be wrong, but I suspect I’m not.
Pierre Debraux 2011?
https://www.researchgate.net/publication/51660070_Aerodynamic_drag_in_cycling_Methods_of_assessment
I lack the interest to read and understand the whole thing.
3 Likes
My guess on the math error is that Zwift gets the allometric scaling wrong for very light riders.
Here’s a rather painful GCN video discussing it.
Basically, I suspect the model is far too generous for children and does not equate to real world performance.
I have no dog in the never ending light vs heavy Zwift rider debate. I’m a lightweight at 61kg. From my experience, I am vastly faster on flat ground in Zwift vs real life. I’m actually arguing to slow my side down 
In Zwift I’m the most aerodynamic small human ever created 
Not only am I incredibly aero in Zwift, I’m able to ride in this horribly uncomfortable position for ever second I am on Zwift.
Which in the real world is totally impossible. Further fueling the unrealistic speeds for the majority of normal shaped humans using Zwift.
Dude in garage sitting up having a drink… Zwift avatar, aerodynamic missile.
One of the gamification elements I actually hope Zwift adds soon is ‘pursuit mode.’
Basically, a rider gets 5mins of pursuit mode per race. They can use it at any time. All at once or just seconds. Essentially, it replicates the bent arm tuck position on a road bike and lowers your CdA. Much like a real life breakaway effort. However, it’s tiring and ‘most’ riders aren’t able to hold it for long durations.
I think this 5min aero powerup actually simulates real life quite well and could really add to the racing. It would certainly add an extra tactical element.
4 Likes
I was thinking the same but in the context of the PD4 thread and gamifying breakaways. A breakaway power-up. I put it aside since it can be less effective if many users have the power-up at the same time. It must be given very scarcely to have an effect.
Another idea is to be able to accumulate many (aero) power-ups and use them together.
1 Like
When you use the Aero power up does a rider drafting you at the time also get the benefit (or %)?