Since you’re under 60kg here, the Zwift CdA engine will assist your speed, but what is fact is 200/55 = 3.63wpkg. Your lower rolling resistance will help net you more velocity
Zwift theoretically assumes a CdA for your biometrics, which raises questions for the validity of their model. I know several 80+kg guys that are greatly more penalized on Zwift than they are IRL
BestBikeSplit is a more applicable model to estimate air resistance effects for biometric inputs
Are we still discussing PD4 or comparing Zwift to IRL.
Lets keep this on topic.
We’re using data from IRL to prove Zwift physics are wrong.
A goal of PD 4.0 is “replicating realism,” so I’m offering my advice to correct their physics engine such that breakaways and attacks actually work as intended. There still remains the missing element of a proper lead-out train by formidable TEAMS!
This absolutely is on topic. We are pointing out pd4 is way too fast compared with IRL and using IRL data to prove it.
Whatever has occurred recently this week with PD4, the robo pacers are absolutely zooming along even without the help of drafting.
When you have Marek, Cat Allen and various others who ride huge distances in Zwift all reporting the same, it’s not just imaginary.
That’s perhaps some other problem. Remember that the changes of PD4 do not change anything when someone is riding alone.
I understand what you’re saying. But when I ride, I try to moderate my speed so I don’t hit the front and lose the benefit of the draft. It is very rare that I find myself in the front unintentionally—only when everyone has backed off and I lost focus and didn’t notice quickly enough. And I don’t surge in those cases—I figure speeding up for a moment probably wastes as many watts as an auto brake!
Is it just me, or does there seem to be a consensus that everyone thinks that the way they ride should put them right behind the RP? IOW, nobody should be at the back or on the sides.
100000000000%
It’s a massive problem.
Pack speeds are wildly too fast. It needs to be addressed once and for all. It’s a significant portion of why Zwift racing is so flawed.
I and others having been harping on about it for some time. Nothing has been done to remedy the issue.
Please ignore Cyrus, his intentions are likely very positive, but his data contradicts every power to velocity cycling equation for regular sized humans. Maybe he’s the most aerodynamic human on Earth…
It’s very simple math. A regular sized male human (let’s say 67-77kg for easy math), with a normal aerodynamic profile, on a modern road bike, cannot maintain an average speed of 43kph on flat ground at 3w/kg. It’s essentially impossible. They’d need the CDA of a World Tour rider on a time trial bike.
Yet, this is the speed a large Zwift bunch travels at around 3w/kg.
Not a single rider in the group is doing the w/kg required to hold this speed.
When riders are on the front at 43kph, it should feel like they are on the front. They should have to do dramatically more power at that velocity. Something in the 4.6 to 5.5w/kg range, depending on overall rider size. Allometric scaling allows larger riders to maintain higher speeds at lower w/kg, so it’s not a singular metric.
However, this is not what’s happening in a Zwift blob. Everybody is doing 2.7 to 3.2w/kg, drifting forward, drifting back, slowly churning, nobody really staying on the front, nobody doing the actual power or w/kg required to hold the high speeds.
PD4 has improved aspects of the pack dynamics, but has done almost zero to solve the fundamental issue…
Pack speeds are dramatically too fast.
I’m sorry if this sounds blunt, but It’s a mess.
I know about the Pace Partner Surges from a month ago…i don’t think it’s related to 1.34 release. It has been happening for a while.
Maybe it is not the robo-pacers who are zooming along, but rather the other riders in the pack who end riding jerky because of PD4? (some accelleration at the front => pack stretches out => riders try to catch up => more auto braking when they slow down => pack slows down relative to the RP who rides at a constant W/kg / is not auto-braked).
I really don’t think pack speeds are too fast : my data (you could check my strava, i don’t ride frequently with packs but these activities are easy to find they coutait « poly » in their names) show that I do ride easily in such packs. And these packs contain people taller and a little bit fatter than me.
But all of them are aerodynamic, that’s right. You can’t ride on front at these speeds without wearing tight cycling clothes and tucking a little bit. But when you do rotate, the front rider is really working but inside the pack, the efforts are really lower, except when going out of a corner… in front I am at 270-300w, going out of a corner at 500-600w, but in the pack I am at 130w, and sometimes less.
When at full speed, you don’t need much power irl if you are totally protected from wind as inside a pack.
You can also check pro speeds… pro packs ride a lot faster, but they are not putting down very high power, when there is no attack in front.
Anyway, my point was to say that pack speeds are not anormal in Zwift, but also to say that the most important thing is that now breakaways can stick.
I changed my way of riding with pd4 in Zwift and when I see the pack (I am in B) rides at 3wkg, I know that with 5wkg a little time and then 4 wkg I can get away.
If the pack still rides at 3 wkg the breakaway will get 10s easily before any reaction of the pack, and after the chase will start.
Before pd4 it would never work as a normal weight solo rider would need to put 6wkg initially and then 5 wkg to build an advantage (and if you do that you are not B but A)… only heavyweights (90-100kg b riders) could make a breakaway, which was not realistic because irl they suck…
There is another thread here with accounts of strange behaviour from the Pacers: Robo Pacer Draft Question - #44 by Arend_teRaa
At the moment it doesn’t appear to be anything related with PD4. I’ll try to investigate further what is going on.
Maybe, but it’s not just pack speeds. It’s Zwift everywhere. If I’m riding solo on Zwift, I can ride much faster than I can solo in the real world.
(i.e. it’s a feature, not a bug, and we should just embrace it rather than trying to simulate real world physics too closely)
Remember that there is no wind in Zwift yet.
Well, there might be (one parameter looks like it can be related) some CdA penalty for the group/peleton in PD4, that will act as some head wind. Single riders or smaller groups will not get the CdA penalty and should be able to ride faster than IRL with equal conditions (no wind).
The more i ride with pd4, and the more i read of other peoples experience with it the less im sure it actually solves any of the problems with pack speeds and breakaways.
PD4 are trying to optimize things under some constraints thats always going to limit any affect it will have. Its more or less trying to stop a moving train, because of 3 factors.
The CDA and CRR of Zwift seems to be the best case scenario of the most aero rider in perfect road and weather conditions. If the pack is doing 40 and the breakaway is doing 45 they are 12,5 % faster. However if the pack now is doing 50 and breakaway 55 they are only 10 % faster. There are of course nuances to this, with not knowing if a lower overall speed would be directly proportionate with effort.
Churning in front of group because of lacking collision mechanics. This is what autobreaking tries to at least mitigate, but its still a problem. As is i can still roll through to the front carrying my speed without a lot of effort pushing up the speed, and then another does it and another and… Most likely requires steering to fix it like RGT, all though their system and especially the Wahoo Steer is not the best solution.
No braking in corners (except for one corner in London). Cornerbreaking helps with two things, it takes the speed out of the peloton as it actually should be going slower than a smaller break in corners. It also tires out riders having to do accelerations after each corner, meaning that the speed on the straights also falls.
Without at least solving 2 and 3, i don’t see a way of PD4 working as intended without sacrificing other things instead
Edit: The removal of the supertuck is probably the thing that helps breaks the most, but considring that all descents in Zwifts are effectively straight line descents i shouldn’t have to pedal to follow a group unless someone is really pushing the front. So its a fix for a problem that in an ideal world shouldn’t be there at all if factor 2 and 3 was there
Yeah, but that’s part of what I’m saying. It’s not real, it doesn’t need to be modelled like reality. So we’d expect pack speeds to be faster, and can be OK with that. =)
Yes, I agree with that as well. The main reason for it being that CdA of riders are too “optimised”. Perhaps making them a bit higher so everything would slowdown by 1-2km/h would make a positive impact. However I expect a huge push back from that change both internally and externally, because everything would be slower and every single record from the past would be “obsolete”… 
We never claimed that with PD4 everything would be just like IRL. It’s one of those things, when we inserted the word “realism” into the event description it got a bit out of hand with people assuming everything would be a simulation. What we tried to do, is to bring a bit more realism to the dynamics of breakaways. And so far, from everything I saw I think we were successful with that. We also don’t claim that in every race a breakaway will be successful, but I think the chances improved quite substantially.
Yes, it does not have to be the same, but unrealistic speeds have some undesirable effects like requiring huge efforts to brake away (speed squared), big effort to just stay in the a group (with the current drafting algorithm), etc., which results in less dynamic racing.
Also the claim that pack speeds will be slower, should have an asterisk. They will be slower if the pack behaviour is more “relaxed”. If not, they could be faster.
So if everyone would just ride like before, and forget that PD4 exists, they will be slower 
Personally, I’d be OK with that, but yes I can anticipate the backlash. IIRC there was one related to changes to the Jungle some years ago (although I don’t recall what they were).
As for records, well it’s not the end of the world. When the UCI outlawed the superman position, the standing Hour records became obsolete, but people cracked on with setting new records under the new rules.