Bug: VO2 max in training status different from VO2 max detail page

Garmin uses different rounding of values in different places and my guess is that this is the case here. Check the VO2 max value on the watch to see the real value. In Garmin Connect it is always rounded to an integer value.

Wouldn't random rounding methods still be considered a bug?

It is probably not random. Code does what we tell it to do. Sometimes it can be unexpected behavior that feels random. Here I guess one page is using truncation and the other page uses rounding. 

Just report it to https://support.garmin.com

e7andy said:

Check the VO2 max value on the watch to see the real value. In Garmin Connect it is always rounded to an integer value.

This was true, but it's a little out of date - things have changed at some point.

Also, you didn't clarify that you mean the points on the respective graphs are based on unrounded data for the watch, but rounded data for Connect. This is the part that has changed - now Connect (both the app and the website) display points on the graph based on unrounded values, just like the watch.

For the watch, the Connect website, and the Connect app, the actual displayed numbers are always rounded. (Unless I am missing something that has changed on the newer watches)

Now if you look at the Connect app (not the website), you will see that the curve of the 4 week VO2 Max chart is actually exactly the same as the curve on the watch (in Training Status > VO2 Max), except obviously the app chart is bigger.

Iow, the chart on the watch is just a tiny version of the chart on the app now.

So actually, the advice to look at the chart on the watch is obsolete (this is advice I've given in the past btw, so I'm not criticizing).

BTW, the 4-week chart on the app has distinct points as frequent as once per day (less if your VO2 Max didn't change that often ofc). But the 4-week chart on the website only has 1 point per week.

So it's def best to look at the chart in the app, as far as VO2 Max charts go.

Either way, looking at the chart won't help you get the real (unrounded) value except that you can kinda eyeball it, since all displayed numbers are still rounded (including when you hover points on the graph and the app/site shows the value for that point.)

However, there are at least 2 ways to see VO2 Max numerical values with more than 0 decimal places:

1) Export from the Connect website, to see infrequent values with 1 decimal place

Open the Connect website, navigate to VO2 Max, and export CSV for the various time periods: most recent, 4 weeks, 6 months, and 12 months. Unfortunately the resolution of this data is bad:

- most recent = 1 value  (well, this is ok)

- 4 weeks = 4 values (1 per week)

- 6 / 12 months = 6 / 12 values (1 per month)

This data has values to 1 decimal place. If you only export the most recent value, you may not see 1 decimal place, as Garmin drops the trailing ".0" if your VO2 Max rounds to x.0. e.g. Let's say your VO2 Max is 53.95. Garmin will round that to 54.0, but it will drop the ".0" for export.

But if you export for other time periods, you will most likely see values with 1 decimal place, unless your VO2 Max isn't changing at all

2) It's also possible to see your VO2 Max to 2 decimal places, with greater resolution than 1 value per week / month.

- In the Connect website, navigate to an activity that was recorded on the day you're interested in

- Export the activity (gear icon > Export Original). This will download a ZIP containing the original FIT file

- Open https://www.fitfileviewer.com/, select Open FIT File and select the ZIP you just exported

- Scroll to the User Metrics table. This table has a VO2 Max cell which shows your VO2 Max before the activity, to 2 decimal places

Using the same site, you can also see the "activity VO2 Max" (in the Activity Metrics table) for any activity which generates a VO2 Max value. This isn't exactly the same as the user VO2 Max (which is what your watch and Connect are reporting as VO2 Max), but it's close. I'm not sure what the exact relationship is, though. I would guess that weighted / rolling averages are involved somehow.

2337912 said:

Wouldn't random rounding methods still be considered a bug?

Garmin has so many bugs and inconsistencies, there's no way they'll ever be all be fixed. I guess the same could be said for any tech company, but I think there are varying degrees.

I just noticed that the "1 year" view for steps only shows 364 days.... (The 4 week view shows 28 days, as expected)

e7andy said:

It is probably not random. Code does what we tell it to do.

Probably taking "random" too literally here.

From the user's POV it is "random" in the sense that:

- it's inconsistent

- it doesn't make any sense

- there's seemingly no good reason for there to be different rounding methods

When people say "random", they usually don't literally mean "determined by chance", they often just mean "senseless".

e.g. Your friend does or says something out of left field. "Wow so random!" That doesn't mean their actions were non-deterministic. It just means they were unexpected.

As another example, look at the gear selection dropdown in the Connect app (for adding/removing gear). When I open this dropdown, it is usually in a different order than the last time I opened it. Loosely speaking, I would say the order is "random".

That doesn't mean it isn't actually governed by some deterministic process [*], it just means that on the surface there's no good reason the order should be different when I try to change the gear for 2 different activities in the space of 20 seconds (for example). It doesn't seem to be based on "last selected" or anything like that, either.

[*] Yes I am a software dev and I know how software works. I even know that most supposedly "random" processes in software are completely deterministic.