Pace Partner Dynamic Pace Testing [March 2022]

I’ve reported to really weird behaviour when trying to join Coco using the Companion app BT bridge (Android) e.g. forced-exiting the Zwift app, different times allowed to settle into
the group’s pace.

I would go for this as well. Maybe name it C#?

Coco (on Watopia) is showing as a D cat at present.

Isn’t every other ride on Zwift varying w/kg?

Would it not be more logical to educate the customer on trainer difficulty?

Simply lower trainer difficulty to 10% or less and everybody can hold a steady w/kg. Terrain then becomes totally irrelevant. Group stays together without any complexity.

There is serious value in training sustained power.

1 Like

Not really true. You won’t feel the hill but you will still go slower if you keep the same power.

1 Like

Coco :heart_eyes: will show as D when going downhill as her w/kg will drop below 2.5.

The colours are output based.

Is that going to stay that way? I noticed it yesterday when riding - kind of threw me off as I thought Diesel had somehow found his way onto Titan’s Grove. :laughing:

On the main menu to choose from I meant. It looked like there was only one C Pace Partner to pick (on Makuri), and three Ds.

Yeah, because she was going downhill at the same time. In the same way that she would show as 2.8 when on the hills.

No - I believe we are looking to move to average power to denote the colour.


Huh, so you are shown Pace Partner options based on their real time w/kg at that exact moment you loaded the menu? That’s not very intuitive.

You are correct in how they are currently displayed.

It was fine before we brought in dynamic pacing.

Ah I see. That probably needs to be changed so as to not be misleading to anyone.


This morning I finally rode with Coco on Sand and Sequoias for the first time since dynamic pacing was implemented, and I’ve got opinions :joy:.

  1. Bruh.

  2. 0.3 w/kg increase on the hills seems like a bit too much. .3 w/kg moves me up or down a zone, so it changes the character of the ride from “steady Z2” to “Z2 with some excursions into Z3”. It’s not that bad on the short punchy climbs in Titan’s Grove, but it could be really punishing on longer climbs with faster bots. I think .1-.2 increase would keep the bots from getting dropped uphill and keep the character of the ride the same. More variation could be cool for some new bots, but I also enjoyed the steady pace from before.

  3. Dropping to 2.3 w/kg downhill is good, but dropping by .3 or even .4 would work well too. Especially for the heavier bots.

3a) Seeing Coco change colors when dropping to 2.3 is a bit jarring. Maybe they could get dimmer/brighter to reflect changes in pace.

  1. The gap from Diesel to Cadence is way too big now. People already wanted a 2.0 bot - the jump from 1.5 to 2.8 is pretty intimidating.

I sent a very similar screenshot to my colleagues earlier today.

I rode with Coco today and didn’t feel like it was too much. It felt like the normal “oh, look a hill”

We can and will look at tweaking the dynamics, but my gut was that going uphill was okay, but needed to slow down a bit more on the downs.

Yeah, I know. That was weird and we’re going to fix it.

The gap is the same. She still averages 2.5 (maybe fractionally higher now). However, we will definitely add a 2.0. and a 1.0. And a 3.0 and probably 3.7. Hopefully in the very near future.

I’m undecided on the 1.0, maybe a bit too slow?


Let the people decide, do a “test” for a few weeks and see if it gets a nice sized crowd. What is the minimum sized group to deem it a success? How many riders does the A bot usually have?

1 Like

Put the bots back they way they were and give us 2.0, 2.5, 3.0 etc. Like we have been asking. Not a big fan of the new bots. And PLEASE dear god stop putting them on Sand and Sequoias.


I actually enjoy the S+S route. It’s better (IMHO) than just cycling around on the flattest piece of road in game endlessly.

Also, there’s only been a change to one bot, so I’m not sure what your experience has been with other ones to make you say that?

Additionally, the only change we have made is for her to go slightly harder uphill and easier downhill. Because that’s what most people do.


It would be worthwhile running Coco or Brevet up the epic KOM one week to see how it works…
Probably need to run with and without increase in wkg feature.

1 Like

I wouldn’t ride with a 1.0 bot, but I think there’s an audience for it. Wasn’t one of the Halloween bots 1.0 or 1.3?

1 Like