Training Effect

Are your HR and power zones set appropriately?  What sensors are you using for HR and power numbers?  Have you done enough rides with good quality data captured to feed into the calculations?

TSS and IF are explained here: https://help.trainingpeaks.com/hc/en-us/articles/204071944-Training-Stress-Scores-TSS-Explained

Hr sensor Garmin HRM-Tri and for power using smart trainer Elite Direto XR. And with that combo in past 30d 900km. Last calibration on smart trainer was week ago.

And all data seems to be pretty accurate (ofc Zwift distance is different, but TSS showed in game at end of activity is same as Garmin. But Garmin own training load/effect (which is calculated in device) is showing pretty strange numbers. Sometimes feels numbers are way too high and sometimes like on this ride which equals 1.5km walking training effect.

Also yesterday's activity did add Training load correctly (atleast recovery time raised 1+day), but on training status (7-day load on 830) is same as day before.


This picture is from same activity. So it seems it collected data normally without any sensor issues.

Check that your HR and power zones are set correctly.  If they are off then training effect will be off.

I can check if there is something out of order.

Last week I did almost identical activity on same route, sensors, calibration, just with bit more effort and shorter distance (no proper warmup)... but all data seems to be good like on y-day activity, besides that huge difference on training effect (which is calculated in device).

So for me its obvious that issue is in 830 software and how it handles data. And cause Garmin patented data is calculated in device
and cannot be modified/corrected, so doesn't it mean that device "broken". Atleast when its one of biggest selling point for device.

And again. Training effect bugged and with it goes training load and load focus. Still calculating correctly recovery. 
This is happening around once of 6-7 rides.

Still any fix for this?

Check your HR zones.  I've had this sort of problem when I've had an HR spike, which then resets the HR zones so a low intensity ride like this ends up all being in Z1 and giving little TE.  Garmin/FirstBeat claim that TE is based on power and HR but it seems heavily based on HR and HR relative to HRMax as far as I can see

How did you record the exercise? Did you start "Indoor Cycling" on your Garmin 830 or is this the data recorded via Zwift? The data from zwift can't calculate the training effect due it is a calculation from a 3rd party software.

Recording data on both (All screenshots here are from garmin .fit files). Recording activities on 830 to get Training Effect, Load Focus and Training load data to Garmin ecosystem and Zwift for accurate elevation/distance/speed data.
That's why have all activities duplicated in Garmin connect, cause still haven't found any solution how to merge .fit files to include all real data (Garmin should done some workaround in this issue already 2 years ago if u ask me).

Here is how sensors recorded data on garmin file..

And here same data from Zwift .fit file.

So no connection losts and actually nothing special in data... Once had to stop when first training ended (after 50min) and then continued with another training.