Vector 3 and TR Virtual Power Discrepancy

Does anyone have any experience with the large wattage differences between Vector 3 pedals and TR? I have replaced my pedals several times because of it and now I am at a loss. RPE and TR match up nicely. Garmin is usually 50w off at least. I have the correct crank length and wheel circumference and usually calibrate the pedals before every ride. Any ideas? Thanks.

Not sure I understand the question. Where is TR getting power readings from if not from your Vectors?

Sorry about that, I have the pedals selected for cadence only, and use TR data for power. My Garmin watch records the Garmin power data and when I compare those two there is the BIG discrepancy.

TR doesn’t invent power data so I assume you’re using a smart trainer in which case the discrepancy is between that and your PM.

My first choice would be to use power match so that TR uses your PM to control the trainer. Second option is to make sure you are doing a spin down on every ride, there will likely still be a difference but hopefully less than 50w.

What is The TR power source?

I have noticed something similar this past couple weeks. For now, I have been measuring power and comparing in lap intervals, but need to do the data analysis in more detail.

My setup:

Kickr core for power and erg mode -> TR
Vector 3s -> Garmin Edge 1030
Precision crank -> Garmin Edge 520.

Each device is calibrated each ride. I do a 10 min warmup and spin-down calibrate the trainer. I calibrate the power meters before I get on the bike and sometimes again at the spin-down before my main efforts.

Bottom line, precision tracks pretty closely with kickr. Vector 3 are way off at low power, say 170w and below and typically 20w. But once I get up to 250 etc things seem to converge.

There is daily variability.

I am adding a set of sworks dual sided power cranks to the test setup for a new build to get a direct comparison set with the vectors. But for now it seems like I need to change the gains on the vectors as GPLama suggests on his deep power dive video.

A shame as I was using vectors as my sole power source for a while. Definitely highlights the need for multiple power sources to get serious about the numbers.

1 Like

Or…highlights the need to just use one power source for all your training (or at least one power source for doing all your FTP tests and structured intervals in erg mode) and spend the time saved on setting up, charging, calibrating and comparing multiple power sources to train more!

Yup, it’s a pain, but like I say, only if you want to be serious about the numbers. My reason for continuing to do this is how the data from different sources converges across the range. This suggests non linearities or at least improperly calibrated linearities, ie offsets. This will undermine training with power even if you don’t care about the absolute values.

Also, I am a scientist and this is how I roll :slight_smile:


I have my bike on a Cycle Ops Fluid 2 trainer. I was using TR for a long time before I decided to get the V3s and start using power data on the road. I noticed the discrepancy when I first tried the pedals and was using their power data connected to TR in a workout. I was killing myself to put out 125W with an FTP of 248. I decided to troubleshoot the problem by comparing multiple data sources during the same ride.

My experience was very similar with a 2017 Kickr. When I started with TR, I was using the Kickr as it arrived, and very shortly into traditional Base, my FTP climbed to 300. Having raced 10 years ago, and at the time using an SRM, my FTP was about 230, sl over 3 watts/kg. I thought gee, retirement and regular sleep must be really agreeing with me. Got some vector pedals, and could not get them to agree with the trainer. Returned the pedals and got some Assiomas. Same discrepancy as the Vector pedals. Turns out my Kickr was reading 60 % high, and after a factor calibration, was much closer to the pedals. No, my FTP is nowhere near 300. I would suspect your trainer power readings are inaccurate.

Just get TR to read the power data from your Vector pedals and all your problems will be solved.


Main Virtual Power info.

Info about why you can’t compare Virtual against Real power.

Basically one is fake, and the other is real. Sometimes they align well, most often they do not.

Use your pedals for all data (especially in TR), perform a new FTP test anytime now (if you haven’t already done so) because you must retest anytime you change your power measuring device.

Also, create a new season in TR from this point forward. It will allow better tracking of your power PRs with the nee pedals.


In a way, yes, my readings would be consistently inaccurate.

Why do you say your readings would be inaccurate using the Vector 3s? As chad says above, Virtual Power isn’t real power, your Vector power is.

1 Like

I think one reason for erroneous power meter pedal output could be that the pedals are not installed correctly to the cranks. How did you tighten the pedals? Also, do you calibrate the pedals before every use?

It is also possible to check the pedal output by performing the static torque test. Never done that, but you can find the instructions with google.

Guys, there’s likely nothing wrong with his pedals. OP has a misunderstanding about virtual power.

Best to stop offering suggestions here that are likely to have him spiraling down a rabbit hole he doesn’t need to.


This is not true for Vector 3. They are no longer dependent on a precise torque to give accurate data. That was true of earlier model Vectors. Further FYI, Vector3s auto-zero offset prior to each ride unless selected otherwise.

1 Like

My “layman’s” understanding of Power and Virtual Power are Power Meters use sensors and Virtual Power uses math. I would think that math was more accurate. That could be completely wrong.

My other issue with my V3s is the Power correlation to speed while on the trainer. My V3s will say I’m putting out 50w but according to my speed sensor I’m going 14 mph. That’s not possible.

And another issue is when I ride outdoors my V3s seem to match RPE and speed is close to target. Go figure. :man_shrugging:

I will do this. Thank you. It will suck but thanks.

Indoor trainer speed is also inaccurate. You’re trying to compare apples and oranges. Trust the actual power meter and speed outdoors. The other two are relatively meaningless estimates that shouldn’t be used for anything once you have real data.

What you’re doing is looking at two measurements that are just rough estimates (Virtual power and trainer speed) and comparing them to actual measured power, and you’re questioning the measured power. You should be disregarding virtual power and trainer speed. Your Vector 3s are likely fine, and you’ve probably wasted your time trying to fix something that isn’t broken. I hate to be blunt, but that’s what’s happening.

Trust the data you’re getting from your Vectors. Disregard virtual power and indoor trainer speed.