Following my recent “WindSportTracker and Android’s “Force Full GNSS Measurements”” post about speed & accuracy, I have reviewed some of my recent sessions, and I noticed on 2 occasions in the past year the Best 2s being lower than the Best 5s, with 1 of the 2 oddities a non-insignificant difference (-1.3mph, -0.3mph), and it led to wonder if there a +/- precision to those numbers, or if it the result of some blip data points.
Full disclaimer:
Both of these sessions were bad, low speed sessions, not necessarily worth your time to tropubleshoot.
I have run both under the v15 Analyzer.
I windsurf with my Google Pixel 7a in a water proof pouch tucked between my impact vest and my wetsuit with obviously “no clear view of the sky”, except the Bonaire session which requires no wetsuit, just a sunshirt .
This was before I was aware of the Android phone’s {} Developer Options “Force full GNSS measurements”, so it was OFF.
I am assuming that:
with duty cycling, the Best 2 is based on the best 2 & only 2consecutivedata points,
with duty cycling, the Best 5 is based on the best 5 first & last data points, 5 seconds apart.
Thanks for the screenshots. If I read them correctly, your Avg planing speed is higher than your Best 2s, which shouldn’t be possible. Is that what you mean?
Just to clarify: Avg planing speed is a separate metric from Best 5s (which you’ll find under Speed Details), so those two can legitimately differ.
If it’s the Avg planing vs. Best 2s issue, then it’s a bug in how Avg planing speed is calculated, not a problem with your recorded data. Could you send me the original file (Download → Original) for one or both sessions? support@sunbits.de
That lets me reproduce it and fix the calculation.
Thanks for sending the files, that made it easy to track down.
Your planing phases in these sessions are very short, mostly just a few seconds at a time. That affects the numbers in two ways:
Best 10s and best 2s look at any window of that length, whether you were planing or not. With very short planing bursts, those windows include slower sections.
Average planing only counts the time you were actually planing, so it isn’t pulled down by slower sections.
The metrics are also calculated slightly differently. With planing this short, both effects together can push best 10s and even best 2s below your average planing speed. Nothing is wrong with your recording. It’s an edge case of how the numbers are defined.
Since it’s confusing to see best 2s below the planing average, I’ll adjust the calculation so that can’t happen anymore.
.
