You forget, Zwift are a business. Alienating any user due to inadequate customer support can have huge consequences with the prevalence of social media.
To enact any arbitrary enforcement requires some level of moderation that so far seems unforthcoming.
3 Likes
Malfunctioning power source .: riders need to learn if they over perform they should stop and not race. LOL
Algorithms could highlight unrealistic spikes in performance, but I fear if Zwift pushes for things this complex right now, it might slow down other more pressing issues.
Inaccurate power source. : I would say Zwift can track trainer numbers and if you introduce a new trainer it will notify you that you are now a new racer and have to do a FTP test and new racer events.
Sometimes people will go up when they get a new trainer, sometimes they’ll go down. It’s only a temporary advantage and I don’t see new trainers negatively impacting race results in significant ways. Even if a trainer is 10% out, it only ruins someone else’s race if that person was at the very top of a category and goes at 110% of the category limit.
Realistically, accuracy will always be something you can improve but never perfect - you can help better educate people and reduce inaccuracy but never remove it. I look at it a bit like this: On a club ride or time trial, some will have bikes which are brand new and cost thousands and give them a mechanical advantage, while others have poorly maintained older machines. On Zwift some trainers will naturally give advantages or disadvantages. So long as there’s consistency in the readings, people with a mechanical advantage will naturally advance to race with people the same “speed” as them.
Shared account weight adjustment. If you share an account sorry for you no racing.
There are other answers to this - limit people from changing their weight up and down too regularly (an incentive to stop people sharing accounts), or instead of using your avatar’s “instant” weight, make it so a user’s race weight is their average weight over the past x weeks, so there’s no incentive to make a sudden drop in your weight.
3 Likes
A relatively quick procedure for removing user reported faulty data seems reasonable. On a limited basis. Perhaps two events per month can be removed - and they are removed ENTIRELY.
But continuing without category enforcement, to save alienating a few riders with power meter issues at the expense of thousands of C and D category riders whose races are being ruined, doesn’t make much business sense.
5 Likes
I don’t think it’s as clear cut as this.
Yes I’ve seen people who are clearly cheating and don’t care. I’ve seen someone be accused of cheating and say “I’m too old to be in that category”.
I’ve also seen newbies get it wrong. In the Giant Crit Crushers, Zwift tells newbies to start in D, and I’ve seen newbies say “think I entered too low a category there.”
When I joined Zwift as a newbie, I was part of the problem - I just dived in and started in D, my results were up and down, and I decided I’d probably move up when I got some regular top 10 finishes. It was only when I started reading ZwiftInsider that I learned what w/kg meant and that I should actually be racing in C. Zwift could have nudged me in the right direction earlier by telling me my racing category when it detected an FTP change.
3 Likes
How many subscribers quit Zwift after losing a race due to inadequate categorisation? Very few if any.
How many would quit Zwift after being booted from every race they entered due to inadequate categorisation? A large enough number that Zwift would care.
Despite the complaints, I’d wager few if any leave Zwift due to its current issues - riders work around it by cheating or giving up racing altogether.
Introduce mandatory categorisation based on wkg and you’re sure to alienate many riders.
Just look at the out of category statistics on Zwiftpower - these are higher in many races where the 50% not opted in are mostly out of category.
1 Like
If categories were enforced, you might have entered your first race in the wrong category. Then Zwift would have put you in the right one. If your performances are somewhat erratic to begin with, no big deal. Perhaps you get moved up or down a category while you’re learning to race.
Again, new riders are not the BULK of the problem. If you don’t know what w/kg means, you’re probably not an experienced cyclist and likely belong in one of the lower categories anyway. It’s the experienced higher category racers who - for whatever reason - habitually enter lower category races.
2 Likes
In conclusion.
Results based categorisation tied to wkg power curve as an initial seed with regular “seeding” races to determine your ability and moderators to respond to any issues within 48 hours.
Job done.
6 Likes
You suggest “regular seeding races” and this could work for regular racers but how do you seed those who only race for fun every now and again?
I’m not quite sure what you’re trying to say here, but in any case I don’t think that anybody has data to answer these questions.
That may indeed be true. But only because there isn’t a viable competitor (yet). And that’s not a very good reason not to fix the problem.
A few, perhaps. The vast majority of Zwift racers, as evidenced by this thread and many others, appear to be in favor of category enforcement.
4 Likes
The problem with your assertion is that this forum does not represent the majority of Zwift racers. Only a tiny percentage ever post here.
Again let’s not overly complicate things… New rider gets given a category after 3 races - you get 3 free races to race in any cat…
The vast majority of users will be limited to their Cats so these riders will be minimal and also filtered out in zwiftpower if in the wrong cat…
Add a marker next to their name that denotes new racer for the 3 races if you must like the zwift academy sticker users get…
It needs fixing ASAP, assigning cats is the quick win with the biggest reward whilst the long term solution is implemented.
Low hanging fruit…
12 Likes
I have done a search for “anti sandbagging” on the forums and found these points. I guess there are many more:
- Are anti-sandbagging measures actually turned on in the anti-sandbagging races?
– It seems like the 1 and 5-minute power is not enough to detect sandbaggers that avoid these kinds of efforts. They can sit in for 20 minutes at a high steady pace and sprint for 30 seconds and not being flagged. As I see it, more metrics should be measured. E.g. 8 min power, 20 minutes power in addition to 1 and 5 minutes. Riders with very good 15 seconds power will be able to perform very well in many Zwift races, even if their 20-minutes power is on average. These are the riders that will end up with good ZP points (low), at the top of their category and should be bumped to up a category. - Anti-sandbagging is unfair for lightweight riders or young
– Only looking at W/kg and not minimum wattages at FTP (as ZP categories do) falsely flag lighter riders as sandbaggers. - The fact that riders can be flagged at any time during the race, means that they can sandbag for 99% of the race and get flagged in the sprint. This means that other riders will have to ride with this sandbagger almost the whole race which can potentially affect the race outcome.
- Sand bagging
– People don’t understand why racers performing out of the category limits are not flagged. This is back to the point about what power metrics are used to detect sandbaggers. The sandbagging mechanism must follow the announced categorization limits for people to find it fair.
1 Like
I think @Gru is right in his assumption, but to be absolutely sure Zwift could create a poll/survey to get more feedback.
Polls and surveys (or questions on a forum) struggle to pick up the many nuances which I think are important with this. Focus groups could give stronger feedback.
1 Like
It depends on the audience for the poll/survey and how it is created and also who and how many are in the focus group.
1 Like
@Lee_H has nailed it. All of this discussion is great and will hopefully shape the medium to long term shape of racing on Zwift.
But we need something in the short term, and I’d say 95% of racers are united in what they want that to be - auto categorisation to current categories based on current best w/kg.
11 Likes
Maybe it is time for you @Mark_Cote to comment again and tell if @Lee_H is on to something or if you want more input on the questions you have asked us.
1 Like
That is a good wrap-up. But, damn (Zwift and Zwift users) why make that so complicated?
First, I wanna say (damn again), some Zwifters are so stupid. Why go racing in the D cat. when you are clearly a B cat.? Who is proud of this? You should not. Zwift should go scrapping some data and find users that clearly cheated from numerous races and put a cone of shame over their head until they ask publicly some apologies. OK, that last sentence is mostly a joke. But they would merit that…
There is a basic solution: you win a race with a power avg. over the category range, your are force to do the next one in a higher category. You win a race with a power avg. way over the category range, you are just disqualified and forced to a higher cat… With time this solution would normalize the categories. Now, there is still a way to cheat that: stay in the peloton to control your avg. power, but sprint very hard in the last meters. That could be controlled by hearth monitoring or by FTP. And there is well known data charts that can be used to spot cheaters: for example, that is not possible that a 2 w/kg racer can sprint at the end of a race at 12 w/kg.
At the other end, if you finish last with a power avg. lower than the category range, you should be force to downgrade. Why do that? You do not want to deplete of racers the lower categories. And you do not want to downgrade a racer finishing last of a tough race while having the correct power.
It is a simple effective solution. So why, oh why, Zwift, why are you complicating things? It is OK to rollback a bad feature for a better one. Do not be so proud!
3 Likes
Alternatively allow uploading of real-life ride data, such as for example is available in TrainerRoad (or get it from third party coaching sites such as Strava, TrainingPeaks if the user allows connection).