[Greg S]

I’m wondering how to determine my Running Functional Threshold Power from the data gathered during an AnT test conducted on a 30-minute uphill run. Is it simply the average power over the course of the 30-minute test, or should a correction factor be applied?

As background, I am using Dr. Markus Holler’s RunPowerModel data field on my Garmin Forerunner 955 to determine power. In his book, Power to Trail and Ultra Runners, he describes dividing the average power during his test protocol (a 20 minute run) by 1.02 to correct for the fact that it’s possible for the effort to exceed true threshold power. I don’t know, though, whether that correction factor has to do with the nature of the test or the nature of determining FT Power in general.

For what it’s worth, my primary interest in determining my Running FT Power is to use it for TSS calculation in TrainingPeaks. I’m also interested in exploring power as a pacing signal during sustained uphills as a secondary concern. It could be that it doesn’t really matter whether I use a correction factor or not for these use cases as long as I’m consistent, but I thought I’d seek the collective wisdom of the forum.

Any insights or advice would be appreciated.

• This topic was modified 1 week, 3 days ago by Greg S. Reason: Correcting name of data field

[Andy Reed]

Hi Greg

Good question. My gut feeling here would be to use your average power over the full 30 mins. If you divide that value by 1.02 then I think the difference that it makes is largely negligible as to make any meaningful difference. For example if your average power is 300W, then using Markus’ math would only change it to 294W, and as you will no doubt have experienced, power bounces all over the place if you follow it real time, when off-road. I think that 6W difference is really not going to make a big difference to how you train, especially on trails. And remember that watches are not even directly measuring power, it’s all math and algorithms, and there is no real consensus as to the most accurate algorithm, and different brands use their own proprietary methods. Given the inherent problems in determining power, I really don’t think that the correction factor is going to affect much in reality. As with anything though, the best thing is to maintain a consistent approach. Use the same approach whenever you repeat a test – make it as repeatable as possible, and if you do apply a correction factor, be consistent. Hope that helps.

[Greg S]

Thanks, Andy. I appreciate the added perspective and example. I’ll go with the average over the full 30 minutes. That’s what I’d been leaning towards as well.

[Author]

Posts