They do, and that makes it even worse. If the RP would hit a particular maximum speed and then just not draft those folk, it would prevent this.
Anyhow, I’m tired trying to explain this.
1 Like
Not recently but way back in the James Bailey era he said they tested it and chose not to implement it. IIRC that was around the time when they introduced dynamic pacing and more pace partners, and before PD4, so my view is it would be a good idea to do some new test events now.
Yeah, I think some folk are conflating dynamic pacing and drafting. We could have the former even if the RPs couldn’t draft.
sorry - was just trying to get my head around it - not suggesting anyone is/isn’t wrong or correct.
1 Like
Ride with any PP through the desert flats in Watopia and you’ll understand the draft effect.
Pace always ramps up significantly through there as its the flattest and fastest part anyway but because its always so busy with other non-PP riders, there is a huge draft affect.
Constance on tempus fugit…yowzers!!
1 Like
Any time Constance is in that section the speed escalates massively. You can be at 50km/h at times.
Whenever you see another RP group ahead you just smash the power ahead of that because otherwise you’ll end up dropped and /or risk getting slowed down to their speed.
I would just rip the bandage off and put all the RPs all on the quickest TT bike for 3-4 weeks and see the difference.
I think the problem with TT bikes might be trying to explain to people what the pace is. For instance, if you post a 3.2 w/kg pace partner, and they are pushing 3.2w/kg with no draft while you yourself are not on a TT bike you personally would be doing so much less power than them if you’re in the draft of a bunch of other riders. So you’d be pretty far from the advertised pace of each of them, and that would be again very variable based on how many folks you’re riding with and thus drafting from.
I mean, I think it has its advantages too, but my guess is there’d also be a lot of confusion.
They can leave Watopia as is and use Makuri Islands as a sandbox. 
The speed tests done by zwiftinsider show that a TT bike is actually quite close to the speed of a rolling pack at the same power with the current pack dynamics. When they first tried it the group was way too fast for the bot to keep up.
But this really just needs to be tested for user experience. Run this setup on the France routes for a month and see how it goes.
1 Like
That won’t work properly - because you need the amount of people who are in the huge watopia tempus fugit groups to test the effects properly.
You need a group with 200-300 riders in it moving at very fast speeds to see if has the effect of the majority slowing down to stay with the robopacer and the faster folk jumping up to the next robopacer or riding away in their breakaway group.
The Makuri groups are always much smaller (because that world is hilly) and you never will see the differences.
1 Like
If they put the pace partners in an event-only world, and put 250xp powerups at every banner, they would get huge numbers for any test.
1 Like
LOL, that’s certainly true.
I think if they just cap the maximum speeds so the robopacer can’t speed up too much in the draft would go a lot of the way to preventing these excessive speed increases.
The test I saw ZI do was a TTT test, which is different. In the new dynamics there’s still a big difference between being in the draft (normal non-TT bike) and not getting any draft (TT) bike. So if a pace partner is not getting draft, but everyone else is, then everyone else will be doing a lot less power to keep up with the pacer. Unless of course they change the physics just for the pace partner group.
But you can see the difference yourself by joining a large pacer partner group with a TT bike, and seeing how much more power you need to keep up with that group. It’s a pretty big difference - draft is very helpful at 40km/h.
My pace is a lot smoother with the TT bike around the robopacer groups, I’m not having to surge all the time to beat the damn sticky draft.
The effort with yumi group yesterday was nearly as high as riding with Constance group because of all the surging needed because Zwift keeps stealing 3-4km/h from me to park me behind at 2.5w/kg rider.
It was annoying how bad that effect was with the huge yumi group (Tempus) versus 20 riders in Constance group (big ring course) where things were far more realistic.
I’ve barely used Zwift for the last 12 months but revisited the RoboSurgers quite a bit this week. I really can’t believe that they are still so bad. Even largely eliminating the effects of dynamic surging by riding with a number of bots on a flat route, Tempus Fugit, it is impossible to know what pace is going to be required with any particular bot due to the effects of group numbers, and the relative position of some riders to the bot. For example today on Tempus Fugit I rode with a Coco group of about 70 riders and averaged 197w. I then rode the same section with Yumi with about 40 riders and averaged 194w. With Coco 20-30 riders were in front of the bot at any one time. With Yumi most riders stayed behind the bot, hence the lower watts required to stay with the faster bot. It really is time Zwift revisited this problem. It would be great to hear back from Zwift about this issue.
2 Likes
Yesterday night Genie group was rolling at Constance pace with people having to do above 5w/kg to not get dropped.
1 Like
You still hear the argument that your power output should be about the same as the pace partner, because you’re in the same draft. But the problem is that as soon as a gap opens up the draft isn’t there, and it’s the power spike required to keep up without the draft that is the issue.
I’d love it if the pace partners were as stubborn as some of the shop ride leaders around here, those groups have perfectly predictable pace.
1 Like
I think it’s more that you should have some idea of what power you are going to need to maintain with any particular pace partner, but you really don’t on Zwift. When I use an equivalent pacer on Indievelo I can roughly predict what I will need to do, so it is possible. If only the scenery on Indievelo were a little more engaging.
1 Like
Some of this excessive fast pace from the draft effects would be solved by putting the all the robopacers on mountainous courses more frequently.
The folks pushing at the front would get a nice workout on some steep hills then can recover on the flatter parts. Everyone gets variety of road types instead of the endlessly flat courses.
That will never happen though… majority wants flat flat flat.
3 Likes
I agree. The only current way I know of to undertake a predictable and steadily paced ride on Zwift is to create a workout with a set wattage and ride it in erg mode. This is what I do now for zone 2 etc but it is not very engaging. I find it ironic that Zwift was created to make training more fun and engaging but they have so far chosen not to do so for this type of ride.