VO2 Max - trail running for Garmin Fenix 6x

Trail running VO2max is already on Garmin Fenix 6X. I noticed it was reported on my last two trail runs.

The first one was on open terrain - more of a dirt road than trail - and my VO2max remained the same.

The second one was a trail run on proper technical hiking trails, and my VO2max tanked by one. The watch displayed a notification about 8 minutes into the run that my performance condition went down by massive 6 points. What is interesting is that the performance condition had recovered later in the run and ended up +2 by the end of the run, but VO2max still went down.

My concern is that VO2max estimate is supposedly based on pace measurements in relation to the HR and other factors, but in my experience pace measurements are pretty inaccurate on trails, especially towards beginning of a run. It seems the pace algorithm adjusts and gets a bit more accurate towards the end, but it seems VO2max changes are based on the first 10-20 minutes of a run.

I'll keep having VO2max enabled for trail running for now, but if it keeps going down I'll have to disable it.

1 minute of -6 and 1 hour of +1/2 and still VO2max declined?

That does not make sense to me.

I am wondering whether trail VO2max should be different that flat running VO2max. One needs special training to be as efficient on trails as on roads.

Yannis77 said:

I am wondering whether trail VO2max should be different that flat running VO2max. One needs special training to be as efficient on trails as on roads.

The whole point of the Trail Running VO2 max metric is that it takes into account that trail running is a lot harder than flat running, and adjusts for that automatically. In theory at least, the VO2 max numbers between Trail Running and flat running should be the same, which is why they combine them into one VO2 Max.

It's a brand new feature, so Garmin might still need to tweak it a bit before it's perfect - that's why, for now, they gave us the option to toggle VO2 on or off on the Trail Running activity.

That's really going to be an issue for me and my fave trail as the first 4K is a serious climb to the mountain top!

mcbadger said:

It's a bit more complicated than that, apparently - the VO2max calculation has been extended to include accelerometer data to detect terrain differences & allow for them (presumably, using collected anonymised road and trail data to support those changes). Trail and road use the same algorithm, but it is very different from what it was in the past. Lot of detail here - search for Herman, the name of the Firstbeat engineer who explained things to Ray.

1. OK, here is what Ray's explanation mentions:
   
   The new element in the mix here is how accelerometer data is being used.  We are basically looking for and identifying patterns in how you are running to recognize when more energy than normal is being used to keep pace. In other words, it’s about attributing the fact that your body is working harder due to the ruggedness and changes of the trail instead of simply assuming you’re working harder than normal because of inadequate recovery or a drop in fitness.
   
   The first few minutes of that run where my VO2 max tanked wasn't any different than the rest. Perhaps I was a bit stiff in the beginning, but it was an easy running at a slight downhill, nothing special that would indicate that I struggled fitness wise. Here is however some interesting data. I have zoomed at the first 8:30 of the run:
   
   
   
   Basically in the first minutes of my trail run Garmin Connect shows my pace as being in the range of 12-16 min/mile. That can't be right. I was definitely running 8-10 min/mile at the time. By 8:30 into the run the pace graph shows my pace as 8:00/mile. That was right after I got the popup about my performance condition being -6, and after that the performance condition had rapidly recovered to +2. In fact the cadence is stable the entire time and the terrain is flat to slight downhill, so there is zero reason for my pace to be so slow.

The point is that the negative hit on the performance condition and the VO2 max was likely based on Fenix faulty pace measurement in addition to other factors like accelerometer input. I've noticed that running on forested trails, the pace accuracy is especially terrible in the first 10 minutes or so. It is always way slower than the actual pace in the beginning - that is something I consistently notice. Then it adjusts and tend to be a bit better, although still very inaccurate. But if the faulty pace measurements are fed into VO2 max algorithm in any way, then there is no way it would be accurate.

Thanks for the analysis! So if I understood this correctly that could be the reason I´ve been consistently loosing ml of VO2 max since the update? 4 ml in as many runs. I´m currently back to the value I had a year and a half ago an I haven´t stopped running, at least 4 times a week with elevation gains of 300 to 500 mts in 6 to 12 kilometers. 

I just noticed that in todays run, my performance score is absent from the metrics for de first 13 min. Then it began with a -4 eventhough I was running up the steepest part of the trail. Then on the way down (with a faster pace but with less effort) the score stayed as 0. It's been really hard to make sense of this. 

Sorry if I am off-topic, I am wondering how you choose to start with trail run mode rather than « normal »? Do you use this mode for rather short distance like 5-20 km (or /1.6 miles)? « Trail run » is often associated with long distance. I typically run that kind of distances ni forests with uneven soils/rocks (but like 70% of the run), with 100-200 m elevation gain. Surely the pace is affected!

The Trail Run mode doesnt have any distance requirements, it's an Off Road mode. Distance doesnt really come into it.