TR & Zwift side by side power weirdness

Have been using TR and Zwift together for a number of years now and have always noticed an issue with the power in the TR workout dropping off then increasing in line with gradient changes in Zwift, even though Zwift is not controlling the trainer. It’s almost like Zwift is getting its hook in somehow, causing the resistance of the trainer to change, resulting in the power for an interval dropping below target watts then overshooting, which I assume is power match trying to compensate.

My setup is Kickr18, Stages singles sided PM, using ERG mode and power match in TR. Stages and Kickr paired with TR, Stages set as power source in Zwift, controllable trainer option in Zwift not enabled.

I normally run TR and Zwift together on a laptop with x2 Ant+ USB dongles but have also tried various combinations with one program running on the laptop with Ant+ and the other on an iPhone/iPad using bluetooth, starting TR first, starting Zwift first, all with the same behaviour every time.

Normally the power in TR is pretty consistent to within a few watts of target when on flat terrain in Zwift, or on a climb with consistent gradient, but every time there’s a significant gradient change in Zwift, i.e., from flat to the start of a climb, or on rolling terrain, I can feel the resistance on the trainer drop off a few seconds before the gradient change happens in Zwift, then visually see it in TR as the watts drop off for a few seconds anywhere up to 20 watts, then overshoot by the same for a few seconds until it settles down. Definitely Zwift related as don’t have a problem if I run TR on its own. It’s almost like Zwift is trying to control the trainer and power match is sorting things out, but it only seems to happen for gradient changes.

Not really a big deal as have learnt to deal with it and always try to choose flat terrain or long climbs in Zwift to minimise the effect but just wondered if any other TR & Zwift side by side users had noticed anything similar?

I mix these two frequently like you. The key difference between our setups is that I am not using a power meter, just the trainer. I am using 2x ANT+ sticks (on extensions) to a Win PC, running both Z & TR with a Kickr V3 in ERG and 34t x 17t gearing.

Over time, I get the relatively frequent power data dropout on the TR side, about 1 per 1-hour workout or so.

  • Depending on precisely when that happens relative to the work interval (before / lead-in / during / lead-out / after), TR will hold the power target that was active when the dropout happens. But over some length of time, maybe 5 seconds of no data, TR and/or the trainer drops resistance to a very low value that is quite easy.

  • I end up riding all over in Zwift during these workouts and never have associated the dropout power feel with anything in the Z world gradient. I don’t think that is happening in my case, just a low default trainer value is my guess.

  • Ironically, I never see the power data drops on the Z side, just TR. So I assume that TR is having trouble for some reason. That happens even though I have done every variation between the two ANT+ sticks, launching either app first and such, with no difference or solution. TR is the only one to see power data drops. Additionally, both of my other devices (HR and cadence) are solid with no data drop. It’s just the power for some reason on the TR side.

So, I have seen data drops and ERG fluctuation, but I don’t think Z is where the trainer gets the resistance when TR loses it.


Yes, it’s not unheard of for athletes to experience some device and connection interference when running multiple applications, especially when you have some of your devices connected to TrainerRoad, and others to Zwift.
Interruptions like this are more likely to occur when both communication protocols (ANT+ and Bluetooth) are being used by your trainer and power meter.

Both Bluetooth and ANT+ are on the same wireless frequency, so it may be very likely that if you’re using both (or ANT+ for training with other Bluetooth devices like headphones or speakers), the Bluetooth is interfering with the ANT+ signal.
The reason why Bluetooth doesn’t commonly experience the same issue is because of a feature called “channel hopping”. When one wireless channel gets interrupted, the Bluetooth connection will automatically switch to another, which will still allow the connection to get around any interference (although dropouts can still happen).

Try connecting your trainer to TrainerRoad with Bluetooth instead next time to see if that helps, and you’ll have to go through your normal protocol in Zwift to make sure your trainer control isnt enabled.
Here are some ANT+ best practices as well that may help with future dropouts:

  • Turn off all other wireless devices in the room including speakers, headphones, keyboards, computer mice, etc.
  • Move the device running TrainerRoad closer to your training setup.
  • Make sure there is a direct line of sight between your devices and TrainerRoad.
  • Try plugging in the ANT+ Dongle before you start TrainerRoad, and then once TrainerRoad loads, try unplugging it and plugging it back in.
  • Make sure battery-saving settings are turned off on your computer (or iPad).

Battery Saving on computer:

  1. Open your start menu.
  2. Type Battery Saver Settings, and open the app that pops up.
  3. Uncheck “Turn battery saver on”

On iPad or iPhone:

  1. Go to Settings
  2. open ‘Battery’
  3. Uncheck ‘Low Power Mode’

Finally, here are some tips on how to pair when using TR and Zwift that may help:

How to connect to TR:

  • All sensors should be connected via BLE
  • A Pro tip to make sure you are pairing via BLE: Close Zwift completely when pairing to TR, and make sure the tile you pair to has the Bluetooth icon instead of ANT.

How to connect to Zwift:

  • Devices should be connected via ANT
  • Power meter should be paired as power source
  • Confirm the trainer isn’t paired as “controllable”

I experience something very like this running Rouvy alongside TR. However, I’ve come to the conclusion that it’s a psychological phenomenon, rather than a technological one. It only happens when I’m pretty fully absorbed in watching the route, and never when I’m taking some food, or distracted by something else. It seems to be the expectation generated by the visual cues on the screen momentarily affecting my pedalling as I prepare for a change in gradient.