It’s not the w/kg, heavies vs lightweights, cat boundaries etc etc
It’s the game physics. They are beyond terrible.
At this point, I think rankings are not the next logical step. Fixing the game physics should be the absolute priority. That alone, will promote vastly more diverse racing results.
As it stands, all races will end virtually the same. It’s so massively flawed that a simple mathematic equation can very accurately predict Zwift racing results.
Unfortunately, it seems many Zwift users are not regular bike racers IRL and do not understand how flawed the game physics are.
James’s four points are the whole ball game.
I’d add that the pack churning for no reason is a massive part of the problem. Without this fundamental fix, Zwift racing will always be incredibly predictable.
No amount of categories or ranking system will fix it. They are all just band-aids.
I’m wondering if part of this problem is just that riders can go through other riders. IRL if I’m in the middle of a pack and want to get to the front I can’t simply go a bit faster and go through other riders (or have others just nicely move out of my way) and get the benefit of the draft all the while accelerating and just pop out the front into the wind with much greater speed. I wonder if Zwift made the riders solid so you can’t go through each other and made you go out to the side into the wind to move up if that would fix or at least improve this situation.
I think the main problem is the server-side rider positioning. When a rider gets to the front of the pack, they should hit the wind and therefore have to produce more power to maintain or accelerate the pack speed.
The problem is, who is at the front is different for you, the person next to you, and the server - even if only by a small amount. So someone pushing a bit of extra power in the middle of the pack could actually be accelerating it at the front, and you get this churn affect.
My gut feel, and this is not proven, is that the problem is similar to the micro-bursts exploit - deceleration as power is reduced is not pronounced enough. If this was increased, the pack would slow more as riders drop power, it would take more of a deliberate effort by a few riders to increase the speed.
I would not implement this unless some sort of steering is available to all riders (even if it’s just simple “move left/move right” buttons).
RGT does not allow collisions and it can be disastrous. A couple of times I’ve gotten stuck behind riders who were dropping back. The pack dynamics/steering algorithm couldn’t figure out how to steer me around them and the anti-collision wouldn’t let me ride through them, so regardless of how much power I put down I was pushed back out of the pack and out of the race. I think they are working on this and I haven’t done a race on RGT in quite a while, but they’ve been trying to get it right for years.
I can’t stop you from doing a series of recovery rides in races for a period in order to downgrade so you can smash the cat below for a period until you get upgraded again. If I’m in that lower cat myself, then at least I don’t have to put up with you half the time.
Right now I have to put up with you ALL THE TIME because as a malicious cruiser (let’s assume you are) you will never get promoted unless you #$%& up and take your eyes off your trailing average in the Zwift Activity Monitor app.
I say that’s an improvement.
If you’re a heavyweight guy who keeps winning not because of intentional cruising but just because you have that advantage, then you are getting upgraded soon enough too, and I won’t have to deal with you anymore.
That’s a definite improvement too.
Welcome to the wrong thread (you know it, you already have a physics thread). I don’t agree at all. They are two different things.
Sticky Watts and these micro intervals exploit. The awful rubberband group dynamics that is not at all like outdoors. Etc. Etc. Yes, I know. It sucks. It should be addressed. But Zwift can never be a sport at all with the current ethics in racing (implicitly saying things like “Underperformance is to be rewarded with podiums”). I/we have been urging Zwift to change the ethics/race rules/cat system for years. You have been asking for physics changes for, what, 2 weeks? So please take a number tag and get in line.
You’re obviously like Flint. You don’t understand yet. Do I smell lucky cat A here perchance? In that case you are definitely in the wrong thread.
you glossed over the bit where i said i would sit up before the line and let everyone pass me to maintain my rank, which would be very simple to do in the absence of a power ceiling. the important part for me is that my training history is recorded on zwiftpower, which requires completion of an event. it doesn’t require a gold trophy or a rank increase
how are you going to solve that without some sort of static power ceiling? are you presuming that everyone who enters a zwift race, including those who habitually win, are doing it because they think of it as a competitive sport?
You maintain your rank while others use yours to improve theirs, that’s what will happen. You’re an artifact along the route. I don’t really see the problem. You will not be winning when you do that. What is unacceptable is when the same racers can win eternally because of a badly designed cat system.
Also, I’m not really a ranking lover. There could be a hidden rank or a rank that isn’t used that actively other than as an e-peen. Unlike some others, I don’t see a problem with fixed categories. They just need to be results based, like in the real world, in real sports, whether physical or e-sports.
People are just a bit stuck in their minds on ranking because they’ve seen it in CS:GO or whatever. But the real glory is never being high rank in any of those. The real glory is always winning events. You may need to reach a certain rank to be admitted into a league in an e-sport, but winning the league is what matters and you do that by winning games, not by climbing in rank which is just the daily grind, the practice games.
As long as Zwift keeps the performance categories it can never become a sport (I don’t care about the unreliability of electronics in this discussion, this topic in itself is enough to kill the would-be sport in Zwift). UCI e.g. would never accept performance based categories, they’re not stupid. But Zwift gets away with it since they front cat A in both high profile races and in the Zwift Academy. They could never run junior leagues in different categories, though, and have the children of the early adopters race up to cat A. They’d have to cheese on with their W/kg BS until some of the kids hit cat A, and then all the spotlight is on those. That’s not how you build a sport. It is all so wrong.
Having never been part of an outside road race, it would be nice if you could add a graph comparing zwift dynamics to the real world to get a better idea how awful or maybe not so awful zwift’s model actually is.
The so-called null hypothesis, i.e. our initial assumptionis, what we would expect of a cat system if it worked well, is that e.g. heavy weight does not give an advantage in itself in races. Rather, it should be completely different factors that decide your placing in a race. I have shown with statistical testning that the null hypothesis does not hold up. We have to refute it. Weight does matter. But it makes no difference because the average reader in here (and that includes Zwift staff) is teflon coated and highly resistant to facts.
How does this weight advantage relate to outdoor races? This is where people get stuck. But if I could test for equality between Zwift and outdoors (I don’t have suitable outdoor data and I think it would be hard to find) it still wouldn’t matter because of this teflon coating.
As for you, since you’re asking, I think what would be most productive is to fully understand why this weight advantage appears at all. Once you do, you will realize that it appears out of necessity. It will appear. It must appear. It will always be there, as long as we have a cat system with performance ceilings relying heavily on W/kg. The tests are just intended to show that what theory predicts actually manifests itself in real Zwift races… like predicted.
Sometimes you do explorative research, collect data and see what you come up with. And then formulate theories based on those findings, to try make sense of it all. Make it into a model or something. In this case it was the other way around. I already knew beforehand before doing my first test (the one I linked above) what I would find. I was dead certain I would. It was pure logics. Couldn’t be any other way. And it has nothing to do with Zwift’s physics model which, with few exceptions like pack dynamics/aerodynamics, is the same physics (modelled) as in real life. The cause is the race rules/cat system.
So read this link (same as above, my earlier test) if you havent. There is also an interesting article on Zwift Insider on why light riders are disadvantaged in Zwift even uphill, something I didn’t cover myself. The key to understanding is to understand the implications of a performance ceiling and to understand that in any cat there will always be riders who can operate right under that ceiling, the top racers in the cat. It is there that the advantage shows, between a light and a heavy rider both able to go close to the max allowed W/kg in the cat. And this is a quirk that does not exist outdoors, not because Zwift failed to model the physics of cycling somehow but because outdoor races use results based categorization with no performance ceilings and Zwift doesn’t.
@Flanders , for some in-game flavor of how the weight advantage has come into play, here is one of several earlier race reports from racing against heavies where I can’t actually win even though I’m actively and consciously cheating.
Yes, the new model changes things a bit in that you don’t necessarily get an auto-DQ for taking up the fight against the heavies, but the performance ceiling is still there, although obscured, in cat B-D so in reality very little has changed. And I just proved it at the start of this thread.
You picked a flat route to race against guys with more raw power and you are surprised that they won. If that race went over one of the bigger climbs you would have gotten a podium.
What is even more interesting is those that beat you have moved up the C category. One of them is in the top ten in most of his races now.
size and strength are only tenuously related … the reason i believe why light people, of which i am one, generally don’t win often in zwift racing is because most people of all sizes don’t do any strength work, you can fine tune it on a bike but at some point you really need to go to a gym for that.
if you put two untrained people together and make them do strength efforts, the bigger guy is probably stronger, yeah. and zwift racing is fundamentally a strength competition, it’s not really about endurance or anything
This post is gold! Had to skim it because I’m in a rush. What I’m seeing is people who are sandbagging in Cat C taking it easy and then stepping on it at a much earlier stage compared to what is possible for most C cat’s (if you have that high of a power output for that long you aren’t cat C). I wonder if these people are warming up for another race or what…
Back when I was Cat D under the old system, I got that green triangle… sadly when I was simply trying to keep up with the peloton. Whatever the system was, it penalized power output rather than speed. Things seem better now… kind of (I don’t race as often, see my other post). My last race however had such bad sand bagging that I had to come here and vent however…
I doubled my weight on Zwift and went slower… What gives? Surely being heavier gives me extra advantage. It has nothing to do with the fact that I’m stronger!
I think the issue is with the ranking system is still mainly based on w/kg. For 2 people with the same w/kg the heavier rider will have more power and will in most scenarios in Zwift have an advantage over the lighter of the 2 riders in Zwift. The larger weight also means that the heavier rider can have a larger power increase before being bumped up a category.
For your test to make sense you need to ride around for 60 days with your weight doubled to allow Zwift to put you into the correct category and then try racing again. Yes you’ll go slower, but will be racing down several categories.
I also put a weight jacket on when I went out in real life and the same thing happened, I was much slower. But people keep I siting, the heavier one is, the more advantage they have!