Running power is way off

peterdb said:

I have done a lactate test last year at a medical center, that showed my anaerobic threshold power is 201W.

Was that a running test or a cycling test?

peterdb said:

Why is the running power calculated by Garmin so wrong?

It's not wrong. It's made up. Contrary to cycling, there is no single definition of what running power is, so every provider devises its own formula.

And all that doesn't even take into account inaccuracies in the measurements that are used to compute running power. So running power is just a pretty arbitrary number in most cases.

Well, that just means that this whole distribution over five power zones is pretty useless. It is actually bad to call it "power" at all and to give the number unit called "Watt". It apparently is just an arbitrary abstract number that hopefully correlates a bit with the real power.
I just noticed I posted my question in the wrong forum by the way, it was meant for the Forerunner 965, not the 265.

peterdb said:

whole distribution over five power zones is pretty useless.

Couldn't agree more.

peterdb said:

It apparently is just an arbitrary abstract number that hopefully correlates a bit with the real power.

That depends on how you define "real power". Garmin has its own definition of running power, and I bet they claim it's real power. Stryd - the running power gurus - have a totally different definition of running power and they also claim their power is THE real power.

If you ask me, power for running makes very little sense at all. Currently there is no technical way to measure it in a sensible way, and any attempts to do it make very little sense. But that's just my personal opinion (supported by some personal - albeit limited - experience). I bet there are lots of people who think differently.

peterdb said:

it was meant for the Forerunner 965, not the 265.

Doesn't matter. Garmin's running power works the same on both watches.

peterdb said:

For my long easy runs it typically shows an average of about 250W. I know this is completely wrong since I have done a lactate test last year at a medical center, that showed my anaerobic threshold power is 201W.

For running power, as others said, you cannot compare 2 numbers if they don't come from the same apparatus. This is because different devices have different methods to estimate the total power needed to achieve running.

Stryd and Garmin say they include the lateral, vertical and horizontal components of the forces at play (wind, gravity), and to some extent there is some assumptions that might differ (in particular, horizontal power expended due to "braking" when the foot hits the ground) . Otherwise, notwitstanding accuracy of accelerometers and sensors (altitude, wind) and these assumptions, the main components are the same. The bigger difference is probably the assumption about the efficiency rate at which the physiological power become mechanical power along these components.

Stryd and Garmin powers differ by a steady factor of about 1.4 and that leads me to think the main difference is about the efficiency.

At any rate, I strongly disagree these are arbitrary numbers from an absolute perspective.

If you are so inclined, check out these articles that dive into the approaches and math behind the numbers.

http://www.georgeron.com/2017/12/stryd-running-power-model.html

http://www.georgeron.com/2017/11/the-govss-running-power-algorithm-and.html

https://geeksonfeet.com/blog/science-runningpower/

https://www.garmin.com/en-GB/garmin-technology/running-science/running-dynamics/running-power/

What is more important is that the power value correlates better to the effort than pace (which brakes down when not on flat). On flat terraing, power and pace follow the same changes. The key benefit is when altitude or air resistance changes. Power will reflect these additional challenges. It is not perfect though and doesn't work super well on technical terrains and unusual gait /form situations.

It happens that the Stryd Power tends to be closer to the biking power in absolute fashion. Maybe it is just chance, or it is by design so that plaforms dealing with FTP and lactate threshold can work normally with running power as they with cycling power.

I use both numbers because I bought a Stryd pod before Garmin released their running power. I ended up using Garmin's power because it is native and easier to deal with for workout designs, etc. But I still look at Stryd power for race planning because power-based planning is more convincing to me than pacing and I found the Stryd power-based predictions to be more often closer to my performance than the Garmin's predictions (although there is some switching around).