Where did you get this?
Garage sale (upcoming feature).
Weight 2 stars 
It should be 0 stars with all the extra weight it is carrying. 
Zwift should overhaul the garage.
For one thing, ZwiftInsider tests should be rendered obsolete. Give each bike a transparent rating. Bikes with the same ratings perform the same. Higher rated bikes become unlocked at higher levels. Bikes would be rated on 3 levels: aero, weight, and tire capacity. They would also have an upgrade level.
The same with wheels: aero, weight, and tire radius, Higher radius tires have lower rolling resistance on rough surfaces.
Brand shouldn’t matter. All of the brands should have the same performance at the same rating. Maybe very brand doesn’t have bikes at every rating, but people sould mostly pick the brand they prefer without perusing mounds of data in Zwift Insider.
3 Likes
That wouldn’t be so good for his website traffic! I am guessing folks pretty well wait for his tests, and if the bike is slow, they won’t touch it. Even more so with mega inflation in Zwift.
Bikes should reflect their performance IRL. So a Cannondale R4000 halo should be an unstable slow mess (it was very difficult / unstable to ride). While a modern TT bike or a Cervelo S5 or any other normal road bike should naturally be faster. The S5 still is fast IRL.
If anything, virtually every normal bike in Zwift, mountain bikes, road bikes, gravel bikes should all be much, much faster than the R4000, which should be almost impossible to ride for even a few metres without crashing.
Sure – have performance reflect the real bike. Give it a rating for that performance, tell people what the rating is, and you’re done.
But bikes don’t exist independent of riders. The bike + rider has a performance (aerodynamics, weight) but with a different rider a different bike may give better results. So the idea that each bike has a performance determined to high precision is false.
And as to weight: the amount of paint on the frame has as much effect as which brand you choose. And you can’t always compare one brand to another since different brands offer different size options, and bigger bikes are heavier.
1 Like
And a huge part of real bike design is handling and vibration absorption, neither of which are reflected in zwift.
So I think it’s better to just have bikes performing to match their rating, and having lots of different options with the same performance so people can choose whatever one they like.
3 Likes
I assume it’s pay to play.. if you’re a manufacturer and want the best performing bike in the game, then you pay Zwift the most. I wonder if there’s actual annual bidding or whatnot.
Specialized was a major investor in a 2020 funding round.
I would love a Zwifterhacks kinda part on the Zwift Website where we could enter our own weight, watts and routee to see how fast the frames are actually for us and how much a frame differs from another frame.
Zwifterhacks is nice, but pretty much inacurate, and Zwift has all the data and stops the need for us to go to third party websites and find questionable to say the least info about things inside Zwift.
I doubt an SL9 is going to be slowed down or equal to something else.
I remember someone wanting an AI thing with a camera that could watch you riding on Zwift and give you worse or better aero depending on your position on your trainer bike.
I am sure it’s all just Specialized being very charitable. Lol!
All that’s on zwifterbikes already.
Thats what I said indeed, but that site is highly inacurate. And I kinda want that kind of info from Zwift and not some hobby project, no matter how enthousiastic, on a 3rd party website.
You said ‘zwifterhacker’ like zwifthacks instead of zwifterbikes. But, anyway, it is inaccurate because it uses average watts, not instantaneous watts. (Eric at ZI wrote an article on how average watts is inaccurate on efforts like Bologna time trial. Plus, any power-up will also throw off the results, especially on a very short segment.Drafting will skew the results all over the place. It is only intended to reflect a likely scenario. It isn’t something that Zwift needs to do. Ride and analyze it. That’s how it’s done.
I disagree on that. I believe all the info should come from Zwift and not 3rd party websites. No matter how much energy and effort they put into those sites it is still not official info. And since they are not official Zwift they can stop for whatever reason tomorrow.
And they are a bit confusing, well at least to me.
According to ZwiftInsider the new Cervelo S5 is the new Aero King and so faster than the Canyon and all others. But when I go to ZwifterBikes and just select the frames and go rank them flat those times are different and I see the Canyon and also the Venge being faster than the Cervelo …
And not even those, even the old Cervelo is faster according to Zwifterbikes and the Felt also …
That is one of the reason I would rather have this info from Zwift. They have the accurate and official data and, either I am looking at it wrong, it looks like ZI and ZB are using different meassurements for speed time.
4 Likes
Luckily these days we do not see bikes getting nerfed as they used to. Remember when the S5 2020 was the absolute king and everyone had them. Until it was nerfed and they all disappeared. Only the owners of them and devotees use them. I own one IRL and even i don’t bother with it in Zwift.
Why anyone puts any faith into interpolated data which is interpolated even further on Zwifterbikes is beyond me.
I guess because it’s quick and easy… and it’s a fun project for the person that maintains it??
I agree with Peter though, and my problem with the way all of this works, where it seems like Zwift is as I’ve said many times before, a marketing company, but seemingly not for their actual product but the idea of their product.
Eric S does a lot for the community, yes, but it’s crazy to me that people like him receive information that we can’t even get from the mouths of the company.
At this point should we just say Eric S is one of Zwift’s community managers??? He seems to know more than Zwift themselves wants to share with their own community via direct channels.
/rant
As for the bike data stuff; we used to have access to files that had all of the actual CdA data etc. from individual parts of bikes etc. But whenever the major overhaul took place a few years ago, that data is defunct and now we’re “reliant” on Eric from Insider to share his data from his testing bots.
When we formerly had literally the data that fed directly into the system…
I’ll once again argue Zwift needs to move to a 5 star system in the garage, and that’ll be good enough…
Final reminder:
Too many people look too closely at the data from Insider and ZB expecting it to be totally accurate, when they fail to realize Eric’s bot tests on tempus can differentiate by 100+ meters total distance, which is more than plenty to upset the top results of the fastest frames in the game; thus already recording less than perfect data.
That’s just the impact inside vs outside lines has on Tempus over 2 laps has.
I’m sure Chris here can tell us how much impact being outside or inside lane has on ADZ; as I imagine it’s significantly more than 100m.
1 Like
If he does the testing in a solo meetup where other riders aren’t visible it should all be the same lines taken, at least on a given pack dynamics/autosteer version.
But yes a bit silly that we can’t just get the numbers for the frames and wheels that drive the algorithms.