Hi, i saw that Strava imposed some changes to 3rd party apps and that you made some changes to the syncing process. I guess this is why I saw that i was missing a couple of months worth of activities.
To fix that, I used the “Download old data” button and the activities all appeared just fine.However, my fitness number did not change from before when the activities were not there. Is there a way to force recalculate fitness / fatigue etc… to fix the issue?
Tx. It will take a while to “sync” all the activities after they first show up on the calendar. Your fitness should be correct when that process is complete.
thanks for that David. I has been a couple of days since i force imported missing activities but it has had no impact - the only thing that has moved my fitness has been the activities I have done since.
Is there a way to force recalculation of imports? my fitness number in TP has adjusted , but not in intervals…
i think it was around the end of April - i managed to force reimport all activities from Strava since then (so I don’t have a gap in intervals anymore) but my fitness curve only started to move back up with activities i have done since the import (i.e. the imported ones don’t seem to have had an effect on my fitness curve). And although I know that TP fitness is not necessarily calculated the same way, it does still look v similar to the Intervals one, except here, the TP fitness curve has moved between late April and late July, whereas the Intervals one has not.
but, what actually happened is that my intervals fitness (16) did not really move up when i force uploaded all missing activities and it currently remains less than 1/2 the level of late April.
whereas my TP fitness, which like the intervals calculated fitness, also fell sharply after i slowed down my training in late april , has returned to levels of late april
so yes, they are the same, but it looks like the little training i did late April to july did maintain fitness levels in TP, but did not have any effect in Intervals
i noted the fitness lines to show you visually the difference of now vs 4 months ago on TP and Intervals
The CTL (fitness) and ATL (fatigue) calculation in Intervals should match TP very closely. What might be different is the TSS (training load) for individual activities. This depends on FTP (for power), resting HR, LTHR and max HR for heart rate and so on. So if you could look at some of the activities during that time and compare Intervals.icu load vs TP TSS that might help resolve this issue.
ok, so i have downloaded TP and intervals activities since end April and have charted CTL, ATL and TSS and there are some notable differences.
On CTL, the spread between TP and intervals grows over time (TP increases faster as activities add up).
And on TSS, there is also a spread with some volatility but still quite rangebound, with the exception of mid August (which is when I think I force reuploaded the missing Strava data in Intervals), where the TSS are just wildly different (by more than 5x).
charts below show the phenomenon, and happy to send you the data if you want
(the gap in the TSS chart is an activity i did without HR)
and the massive differences in TSS between TP and Intervals in the middle of the bottom chart comes from the fact that Load is 2 in Intervals vs 100-150 in TP for two activities. Not sure why that is though. Did the resync of the activity, but it made no difference.
i noticed that my TP account is fed by Garmin Connect and my Intervals account by Strava (which is itself fed by GC…). Could this play a role in these v weird TSS numbers?
The data in intervals comes from Strava, whereas the data in Training Peaks comes from Garmin. And the data in Strava comes from garmin…
More importantly, I don’t know what data actually gets imported during syncs - I.e. I don’t know what is imported and what is calculated locally.
So i get that resting HR is wrong in one activity, but TSS is wrong in both activities, and I don’t know if TSS is imported from Strava, or calculated in intervals by using data imported imported from Strava.
But in the end, I do know that the raw data is location, speed, cadence, etc and HR. Everything else is calculated from that somewhere, on some platform, which is why I don’t understand this huge discrepancy.
