Hello,
I know I’ve posted this before and I sent a session GPX file for inspection.
After the app improvements unfortunately the stats still are wrong, and not only few turns, most are wrong 
.
The wind direction is correct, but even if I move it some degrees up or down nothing substantial changes.
Is there anything I need to do in the app that I’m not doing?
Thank you
Are you on version 2.0.24? From this version onwards, wind direction is edited via the Weather Editor – not like before.
Have you adjusted the wind direction there? At minimum start and end point of the session need to be edited for the interpolation to work correctly.
Which device did you use to record, and was the GPS set to 1-second intervals? Longer intervals significantly affect maneuver detection.
Are you windsurfing with slow tacks/gybes?
Feel free to send me the GPX file of the session and I can take a closer look.
Thanks for the reply.
I’m on the latest version.
I use Garmin Fenix Pro Solar. Indeed configured for 1 second.
The stats just started to get worse with the latest version.
Generally I may fail 20/30% of the turns, be jibes or tacks. But app normally flags most of my turns as failed.
No matter if I slightly tune the wind direction, nothing significant changes.
The stats show turns got really worse in 2026, which is not correct.
I sent track before but nothing changed.
I appreciate it must be complex to nail all the stats, but my stats actually show disastrous, I’m not a great windsurfer 
but I’m not that bad either 
I have the same problem. How can I reduce the flight speed? 10 km/h is too much for wingfoil. My wing still flies at 8 km/h.
Thanks for the detailed feedback and the screenshots.
I took another look at the session file you sent in May — this time with the latest analyzer. I checked every single maneuver manually, and the analysis matches what the GPS data shows: 31% of jibes and 62% of tacks failed according to the recorded data. I can only work with what the GPS tracks tell me, and that’s what they say for that session.
A few things worth clarifying:
The May file vs. current sessions. That file is now a few months old. If you’re seeing worse stats specifically in 2026, I’d need the actual session files from the sessions you think are wrong — not older ones. Can you send me a few recent ones?
The 2025 vs. 2026 comparison. There’s a good chance your older sessions were calculated with a previous version of the analyzer. If so, the comparison isn’t apples-to-apples — the old version may have missed some maneuvers or scored them differently. That could explain why 2025 looks so much better than 2026 in the stats, without your actual sailing having changed.
To move forward I’d need current session files from sessions where you feel the detection is clearly wrong. Without those I can’t tell whether there’s a bug or whether the data is accurate.
The flight speed threshold isn’t configurable—and that’s intentional. If every user sets their own threshold, the statistics become personal and lose comparability. The whole point is for comparisons to carry real weight: if your friend says they spent 98% of their time flying, while you spent 99%, that really means you were slightly better.
The threshold isn’t an arbitrary 10 km/h; it’s based on measurements. Some foils lift off a little earlier, while others do so later. But the key point is that nobody flies right at the minimum takeoff speed. As soon as you’re in the air, you’re already going a few km/h faster. The brief moment when the foil lifts off and accelerates is so short that it can be disregarded—and since it affects everyone equally, the statistics remain comparable and reliable.
Nevertheless, if something looks wrong in your sessions, it’s best to investigate using the actual data. Could you send the session file to support@sunbits.de?
Thanks for the follow up.
With regards yo the May session, the app categorized as failed gybes what clearly were tacks, with a tack shape and directional lines.
It’s fine, no need to analyse any current session.
Best regards
