Disable Lactate Threshold auto detect

Hi, what specifically is broken about it?

A couple months ago I submitted a ticket to Garmin because I thought the auto-detect wasn't working for lactate threshold. I thought this because I incorrectly assumed that auto-detection of a new FTP value would also cause a new lactate threshold to be auto-detected (based on the "manual" method that the lactate threshold is the heart rate during a 20-min max average power segment for FTP). However I later learned that the Garmin unit does not use this method at all. Instead it relies on yet another secret proprietary algorithm from Firstbeat that is based on heart rate variability.  On a later ride I received an auto-updated LTHR (without any change to my FTP).

Garmin Support also told me that I needed to have cycling-specific HR zones set in order for LTHR auto-detect to work. However I have kept only the basic HR zone set and not added a cycling-specific set of zones, and the auto-detect for LTHR still seems to work. I think the info from Garmin Support was not correct (it sounded completely backwards to me).

Lactate threshold auto detect is obviously broken because the detected value of 46 bpm is way too low. That's just not physiologically possible.

Mine works fine. Do you use a heart rate strap that is HRV accurate?  I am using Polar H10 since it is HRV accurate both in Ant+ and BLE. Wahoo Tickr ex is not. 

I would suggest there is an issue with your Edge device and/or your heartrate sensor. The auto lactate threshold detection has worked consistently on my Edge 530 with firmware version 5.10 and a Wahoo Tickr Fit heartrate sensor with firmware 1.0.11 connected via ANT+. It generally gives me an LTHR value of about 170 bpm.

I recommend contacting Garmin support again for a better answer.

I use a Garmin HRM-Tri. It gives sensible HRV values when connected to my Forerunner 935.

I have been using a wahoo tickr that must be why why LTHR on my garmin 830 isn’t working. 

Kyle J said:

On a later ride I received an auto-updated LTHR (without any change to my FTP).

Hmmm, pretty interesting. While I got  auto detected LT HR (and pace) for running from my Fenix 6X, I never got auto detected LT HR for cycling from my Edge 530 even if it is ticked. I assumed it was because I always used powermeter and if Edge 530 could see watts and could auto detect new FTPs then at the same time the auto detect of LT HR was suspended.

It seems I was completely wrong with my assumption.

Anyway, having seen Kyle’s post  it is very interesting why I have not got it ever, because my 530 is older than one year and it gave me new FTPs for 6-8 times at least..

I dont know what has changed and why, but I was on  a cycling journey in the second half of August when I made long climbs at relatively modest power and heart rate levels and my LT HR which I had manually set earlier (see my previos post) was changed and lowered three times in a row. 

So even if I was far from my LTHR of appr. 165 bpm during these rides, my 530 detected 162 bpm, then in a couple of days (and rides) later 158 bpm, finally 156 bpm.

 I have not the slightest clue why it happened after 18 months of non-detecting cycling LTHR by my 530. 

The same sort of interesting thing it was that parallelly my FTP was lowered and lowered even if I tried to keep my wattage under 75% of my FTP during these long rides. The decreases were much smaller, insignificant than for LTHR, namely they were 3, 8 and 4 watts, respectively.

When I came home I did an all-out 20 minutes ride, my FTP went back, more precisely 10 watts above the pre-journey level, so it was a jump of 25 watts.

So I must say that the algorithms to auto-detect both LTHR and FTP are imperfect, because sometimes they try to say something even if we ride at modest efforts.

And at the same time it is really a mistery that why  I had completely different experience during these two weeks than during the previous 18 months as regards LTHR auto-detection.

Yes unfortunately the FirstBeat algorithms used by the Garmin devices are secret-sauce proprietary and have not been peer reviewed, so it's difficult to know precisely what they are doing in any specific case. So my comments here are based on what I've read about them in their (very limited) publications, and my own personal observations with my own data.

I believe something that is common for all device brands is that they have no way to know if you are giving your max effort, or something less. They can only detect what they see, and will label it as your best. (Okay technically there appears to be some extrapolation to estimate our max even from sub-maximal efforts --- this is indeed FirstBeat's major claim --- however the reliability is obviously stretched the more we extrapolate beyond the actual data.) So to get more credible results, we need to regularly do strong aerobic efforts so that the device can see our "best" and trend it.

If we go an extended period of time without doing that, it seems the device will simply hold the last detected values. And it appears it can hold them for quite a long time. If however we then put in an effort that was just hard enough to trigger the detection level, but not our max, the device can detect that as a new lower value. Fortunately it gives us the option to accept/decline it, so we can ignore it if we're confident that it wasn't a max effort.

I speculate that this may be similar to your case. A fairly long period of generally similar effort mixes so it's holding the values for quite a long time. (Mine does this, sort of a "plateau" I guess, as I do a lotttt of "base" riding / "junk miles" haha.) Then the notable outing with the long climbs may have been high enough effort to trigger the detection algorithms, but they gave lower values presumably because it wasn't a max effort. Long climbs are a good recipe for new detections because they tend to give us steady elevated aerobic effort levels for an extended period of time. (Around here I can rarely do that on the flats without interruptions before running out of road.)

Of course there is always the possibility that your actual values could have lowered for real, depending on a wide variety of fitness and health factors. 18 months is a really long time and it's fairly certain to have some fluctuations during that time. FTP in particular can backslide fairly quickly from peak values in a matter of weeks, and if we don't do hard efforts regularly, the device won't trigger new detections to see the lower values (and our body won't adapt to maintain or improve our thresholds).

Your later hard effort that gave you the new higher FTP seems to confirm this is what happened, that the climbs were high enough to trigger new detections compared to your previous riding, but they weren't max effort. When it gave the new FTP, it should have also shown the LTHR in smaller size immediately below it. How did that LTHR compare?

Kyle J said:

believe something that is common for all device brands is that they have no way to know if you are giving your max effort, or something less. They can only detect what they see, and will label it as your best. (Okay technically there appears to be some extrapolation to estimate our max even from sub-maximal efforts --- this is indeed FirstBeat's major claim

Your way of thinking is logical, but the point is that FB claimed that involving HRV data helped them to define a lot of metrics like training effect, lactate threshold etc. so the algorithm knows pretty well whether you are below, at or above your (anaerobic) threshold at least, even if it  says nothing about your real short term max efforts.

And following this logic, both FTP and LTHR (the latter is evidently) linked to the threshold FB claims to be able to detect.   But if I did not reach not just my (short term) max, but did not go over the 75% of my FTP and/or over 85-90 of my LTHR, how the heck it could detect a new lower FTP and LTHR? Apart from the option that I became relatively untrained and/or ill, as you call it health factors.

Kyle J said:

Your later hard effort that gave you the new higher FTP seems to confirm this is what happened, that the climbs were high enough to trigger new detections compared to your previous riding, but they weren't max effort. When it gave the new FTP, it should have also shown the LTHR in smaller size immediately below it. How did that LTHR compare?

That is the point again. As it was between March 2021 and August 2022, the same happened since I am back from this journey: I got no new LTHR at alĺ not even after my all-out 20 min. ride.  It is still stuck at 156 bpm.

Obviously  I did not play with the settings of auto-detecting FTP or LTHR.

So even if my cycling pattern is really mixed, I have no training plans, just cycle at different speed, HR, watts and mood and following some short term desire, my Edge 530 thinked that this guy should get no LTHR in his home country, just in Italy and France while conquering the famous passes and cols. :-)

Kyle J said:

Fortunately it gives us the option to accept/decline it

I do remember that my 530 has an option to accept FTP or reject it, but I think that the same is NOT applied to the LTHR popped up at the same time or as a sole data at another occasions. At least it is how my Fenix 6x works. I mean I cannot reject my running LTHR, or my Vo2max,  but I could reject my FTP when I used my Fenix for cycling, too