Garmin and TR FTP's -- big difference

Setup is 5th gen Wahoo Kicker
Polar chest strap
Garmin Fenix 6 and TR on ipad both getting the same data.
Running “Indoor biking” free ride on the Garmin to capture ride for my Garmin metrics.

Did a TR ramp test followed by a 1hr Zone 2 Obelisk.

On completion of ramp test TR said my FTP was 207

When I completed the two rides which were back to back as far as the Garmin was concerned it said it wanted to update FTP to 231! Which I declined.

Am I under cooking TR? I suspect I would be crushed if I told TR 231.

How long do you think you can hold 231 for? What about 207?

@Chuck_Tatham this is just my personal opinion after watching the two metrics for a while…Garmin FTP estimation is typically higher than what I can actually hold for 45, 50, 60 minutes. I think it is more of a critical power number.

No problem there…critical power is a fine metric. But critical power is more of a 30 minute power number. TR strives for a lower intensity as the metric of merit (something you can hold for more like an hour). So if you use the Garmin FTP number to set your TR-derived workout schedule it is probably going to be too much to handle by week three or four.

I don’t think that’s the case based on my own FTP data and watching the forum

My Garmin FTP and Intervals ICU eFTP are very similar and usually noticeably lower than TR’s FTP. The TR number is currently still working fine but is not something I could hold for an hour. The eFTP and Garmin FTP are always very close to what I perceive my MLSS to be and a number that with a bit of psyching up I could ride for a reasonably long time.


I’m using an Edge 530, but only did 1 TR ramp test after buying the 530 about 3 years. Unfortunately I only wrote down the TR estimate and not the 530 estimate.

That said, my ftp estimates from the 530 are fairly accurate. For example the last field test I did 275W for 32 minutes and Garmin estimated 264W ftp. That field test wasn’t all-out or failure (my bad), but the Garmin didn’t know that. Anyways I believe the Garmin estimates are conservative and a little low, a few percent below what I could actually do on a long test.

1 Like

Im actually surprised how closely they track for me, to be honest. Garmin has had me fluctuating between 306 and 311 while my TR FTP is currently at 316.

I only use my Garmin for outdoor rides and as far as I understand, it’s FTP calculation is only for rides recorded on Garmin devices, and does not calculate for rides imported to Connect from other platforms.

I apologize in advance if I’m assuming or inferring too much, but it sounds like OP doesn’t have a large number of rides on one or the other platform. The only reason I say that is there was no reference to any progression of FTP on either. I expect things will converge over time. If you haven’t tried it already, Garmin head units have a ramp test workout available, maybe your watch does too?

Ultimately I am curious if others are struck by how close or divergent their TR, Garmin, Wahoo, Zwift FTPs are.

1 Like

It’s depends on what kind of rides/efforts you do for the Garmin to have something to estimate off.

Whatever, for your TR training, use the ramp test to go off and forget about the Garmin.

1 Like

Something is amiss, either not recording everything on Fenix 6, or not using a chest HRM, or something.

FWIW I’m also getting good estimates after a couple weeks endurance and some short tempo intervals :man_shrugging: but on a 530 with Garmin dual HRM.

@Bhoffman I have 155 rides on TR and 500 on Garmin. No shortage of data…however the 500 are non power meter MTB so not really inputs.

@WindWarrior always with chest HRM

1 Like

@schmidt I am leaning this way as the Garmin may be doing some thing with my non power meter outdoor rides? Although I have never seen Garmin put up an FTP message on anything but after having been connected to the trainer.

According to Garmin, auto FTP detection requires a steady 20min or more effort. My guess is it’s taking the 20min best NP from your ramp test and doing its calculation based on that, which is leading to an artificially high estimate.

1 Like

That can’t be true from my experience I’ve gotten ftp estimates from workouts like sleeping beauty which is 5 minute 30/30s. I’ve even had some from other vo2 and threshold workouts that aren’t 20 minute long.

Other news, Garmin, Strava, and TrainerRoad are all estimated within five watts of each other.

Strava: 323
AI ftp: 325

Interesting. I’m just guessing based on what Garmin says: Edge 530 - Automatically Calculating FTP

You sure you got estimates after rides that were under 20min?

I guess a ramp test doesn’t last 20min either, so yeah it’s not as simple as Garmin makes it sound. At least not the meat of the test that would produce a high NP.

Yes like I said I’ve gotten it from 5 minute 30/30s. I think it’s odd when it happens but it does happen for me. I use a garmin 530 too.

I guess it’s probably more like what @Brennus said, just plotting a CP curve and estimating based on that.

1 Like

”To calculate your FTP, the Firstbeat Analytics engine requires a stable VO2max estimate and enough quality heartbeat data recorded across a range of different intensities from low to high.”

The ramp test clearly has a range of intensities.

”Once this data is obtained, the Firstbeat Analytics engine identifies your FTP level by isolating the deflections in your heart rate variability that correlate to key indications of how your respiration patterns respond to the intensity of your activity.”

1 Like

The methodology is very different. ICU and TR use FTP estimates from MAP or extrapolation from 20 minute tests, Garmin tries to estimate where the ventilation threshold lies.

Garmin’s method allows it to get estimates even if you didn’t follow a specific protocol.

There’s probably cases where one can be more accurate than the other, and from person to person, season to season, fitness to fitness. Which is on average the most accurate? Simply because of the closeness to the thing being measured, I’d say a 20 minute test. I can see that ICU/TR would overestimate with high anaerobic capacity, and Garmin is probably off whenever your HR is a bit off for whatever reason. On the other hand, if you’re looking for the Z2 transition or MLSS, looking at the ventilation threshold is probably better than looking at the 1h power.


This has been all most interesting and educational so thanks for the insights. I do think the tight control over the process and devices by TR make it more likely to be accurate. If I am honest with myself, as much as I would like to think I am at 231, I am more likely to be able to hold the 207 for an hour.

Such conflict…I want the number to be higher but I want the suffering to be less :slight_smile: