CDA Measurement on a long Course. What was happening?

[quantus]

Hi,

today I made a real world test run in Profile 4 to get a better understanding.
2 times 20 Minutes with same setting at threshold power on the same out and back course, Lap 2 and Lap 4.

The rides were pretty consistent, some wind and light traffic.

The analyzed CDA are different to garmin data. What was happening?

Screenshot 2025-08-22 212657.png (27.75 KiB) Viewed 2824 times

Screenshot 2025-08-22 212809.png (27.75 KiB) Viewed 2824 times

Ride file is attached.
Sorry for my english. I am from Germany
Thx

[Velocomp]

It looks like you bumped (rotated) your AP around 51 minutes. CdA and watts went crazy afterwards.

Lap 2 and lap 4 were out-and-back laps, good for CdA testing.

VA analysis can overcome the error in the "bumped" lap.

Analyzed lap 2 CdA was 0.204

1.png (40.18 KiB) Viewed 2774 times

Analyzed lap 4 CdA was 0.201 (analysis corrects for errors in rotated AP).

2.png (62.59 KiB) Viewed 2774 times

So, you got good results despite the bumped AP.