I’m not expecting much change to the more organised series as they already DQ for ZP/HRM after so a lot of riders avoid them.
Good to enforce though pre-race now but "We want to provide you all with the mental clarity that will let you push for better results, not have to spend time thinking “is this person worth chasing down?”
If this the intention the first thing we need is a single results system so we are not guessing who is on ZP or not.
well, it’s rare you see anyone actually using ZP now, that’s true. i am hoping they will be able to detect when someone is using an elite muin or apparently a powerpod as @Douglas_Cook said above since sometimes zwift seems to detect them as power meters and sometimes they are (correctly imo) interpreted by zwift as ZP. there were guys using muins in zrl A1 in the last season i did doing 7wkg for 5mins, it was ■■■■■■■ infuriating.
being unable to switch power source mid race is a nice addition
We’re not expecting large turnout, just to see that the feature works as expected (obviously we have done internal testing already but you guys have a unique point of view).
I disagree actually. There’s often some moaning that ZPower riders shoot off the front and spoil the racing dynamic. I’d expect reasonable takeup once made available.
Might just be the choice of races I do that it seems to be less of an issue now, likes of tfc mad monday, closed events. Plus those that always DQ for ZPower on ZP tend to have less zp users so it seems a bit less of an issue than it was…and i’m rarely near the front these days so that could explain it
Agree though I think the take up will be more than reasonable by race organisers leaving zp riders limited to the Zwift owned events.
Thinking about it what are the chances zwift ever look at more than 5 pens?
If you could setup a race with say 10 pens to account for different types of riders. (or different w/kg aka split cats) this would be allow greater flexibility.
and on that is hardware requirement going to be an option per category?
We’re working on ways to automatically detect egregious power curves for instances like this.
For example, if you did as you mentioned and adjusted your pedals to be higher for that one specific race but had never done that in other Zwift activities, that would be invalidated and while you might reap the benefits for that one race, it would have no long term benefit for you.
Pairing and hardware identification for this is based off the device’s unique number, regardless of BLE or ANT. So long as the core number of the device doesn’t change, you should be able to swap between signal types without a worry.
However, if you find that to not be the case then let us know. This is why we test.
Agreed. This is a step in that direction, just as category enforcement was. I believe that a lot of this second-guessing of results comes from the fact we still use power-based categories and that we don’t have unified results and scores, as you’re saying.
There’s a reason I put “Scoring” in the roadmap update for last month
I guess I’m wondering at the end-game, after testing then. What’s the plan? A race would have to be split to 4 separate registrations to cover the PM vs ZP, HR vs NoHR, each with 4 cats? A race field then gets split to 16 groups then instead of current 4?
I think most race organisers will stick to one configuration. Mostly likely PM and HRM.
That was my suggestion if it was easy to add more cats within an event maybe they would add further configs but if means more events to cover all scenarios they won’t bother. Overall just means less races for ZP users but potentially better races for those on PM
That happened in a race I did a couple of days ago. ZPower riding doing 5w/kg on the front. After about 20 minutes they got the warning flag (I presume they had been doing a steady 400w for the whole race so far) and were booted/retired from the race.
Still, cheaters are gonna cheat. I was watching this race last night and B. Hall had some interesting moves. Overall, they won the race at 5.2w/kg and 182w which would make their weight 35kg but there also must have been some severe height doping.
42:22 B. Hall is 20 seconds behind the lead group doing nearly 6w/kg. 21 seconds later they absolutely fly past the leaders at the start of the leg snapper. The w/kg alone can’t explain the speed they were traveling at.
Towards the end of the race B (45:42) you can see in the nearby riders list that Hall is in second place doing around 2w/kg (70w, remember their weight is 35kg) and gaining on the leader who is doing 5-6w/kg.
Configuring your power meter to report incorrect data is nothing new and this test does nothing to fix or make that worse.
It would be nice to have dual power recording being automatic when two sources (trainer and power meter) are available.
Or in the alternative, even if not saved to the FIT file, just upload both power streams to zwiftpower so that it is automatically available for review.
I would love to see Zwift introduce a community race series (Future Works?) where their main emphasis is placed on fair racing i.e. implement measures to address weight & height manipulation, limit the power source to a list of verifiable smart trainers (e.g. no third party power meters) and enforce CE.
If there was some kind of reward for placing well (e.g.a uniquely liveried Zwift bike), I’m sure it would attract enough participants.
ZRL really is a bit of a joke as far as sandbagging is concerned.
Hi James. Yes, I’d pay for it if I could be convinced that the system was as bullet-proof as possible (with regards eliminating cheating/sandbagging) and that it would be actively policed by Zwift to keep it fair going forward.
I guess you could call it the Premium Race League available only to premium-priced subscribers