Since last week I"ve noticed that all my workouts are getting differents results between the plataforms.
I use Strava as a source to have all my workouts from different source, but when I analise the workout I use the Intervals, but my last workout shows me different WATTS number if I compare to my Strava.
A have a series of 30 seg intervals, which I did around >300w, I saw this at the moment when I was riding, I see this in the Strava file, but here it shows something like 20 or 30% less watts.
Intervals does not actually calculate the average power for each specific interval, rather takes a 30s rolling average (I think, and Iām open to correction here).
Iāve experienced your issue many times and it is a total pain for short 30s intervals like you did. The power for the actual interval is just never correct.
You can try adjusting the interval on the tab circled in the screenshot below.
Yeah, i know that intervals does not calculate, it just take the information from another source, right?
Thats what I found strange, because that other source shows the right amount of information data ā¦ so why is diferente here?
Last workout I tried to adjust like you did, changing the time, but it seems a lot of work for each workout I do
Intervals does calculate, just differently.
It doesnāt take an interval in isolation and give you the average power for that. It gives a 30s rolling average power. Iām open to correction on this if anyone else knows for sure
I think @nasatt is correct on the 30s rolling. Not entirely sure but I think Iāve read it before too.
on the other hand, since you;re looking at intervals. Iām presuming youāre using laps, can you turn on āuse lapsā instead of having intervals.icu calculate the averages for you?
I think by doing āuse lapsā you can get it to take the values directly from the FIT file / source instead os intervals calculating it.
Guys, what I cant understand is, it always worked fine with 30 seconds intervals, I have a whole library of worlout that I done, all of them seems perfect, I use to analise and compare, but since January this problem begin to happen ā¦ If Intervals use the number that took from Strava why there are this difference in numbers?
Iām always using LAPS (use laps), thats the only way to have here the workout that I done ā¦
Iād like to still use intervals, itās the perfect plataform, but i need the right numbers for it =/
The rolling 30s avg is only a problem with Normalized Power NP. Intervals calculates NP for the entire activity first and then uses that stream to display NP for an interval.
Avg Power for Intervals should always be correct and is calculated by Intervals. It is simply the average of every single data point within the interval. It is not taken from the FIT-file or any other source. Donāt know where you get that?? @Ritter : Are you sure Strava displays Avg Power and not Normalized Power? If it is normalized power, short spiky efforts will yield something much higher then Avg Power. Secondly, verify that the intervals are correctly synced with your Power Trace. A few seconds off also causes differences.
The source was Zwift, and was manually upload to Imtervals. Strava is purely a social app for me, and I donāt consider the analytics at all.
WKO5 is closer to Intervals, but only when I select the interval duration on each lap to correspond with the actual workout plan. In Intervals itās the same process.
This is the explanation on the WKO FAQ page:
Why donāt my lap numbers in WKO5 match TrainingPeaks or other apps?
Many recording devices, analysis apps, and websites summarize recorded data, such as power and speed. These summaries and metadata are what are used in TrainingPeaks to display workout data, while WKO simply reports actual data recorded in the file. For this reason, you may see variances between numbers stated in TrainingPeaks and those stated in WKO. If the device summarizes incorrectly, TrainingPeaks will summarize incorrectly, too, as it simply reads the metadata from the device.
Laps are a good example of this, usually because a device records the wrong start or end time of an interval. If a device records the wrong start and/or end time of an interval and then summarizes the data and communicates that summary to TrainingPeaks, TrainingPeaks will use the summary data and match the device.
However, when WKO downloads the workout, it pays no attention to the device summary, but instead reads the actual data in the file.
Hereās how to test this:
In TrainingPeaks, do the following:
Open the workout and click the Analyze button.
Zoom in around the first interval.
Now manually drag that interval on the chart for the exact same time.
Review the power for the dragged selection; youāll likely see that it changed, even though you dragged the exact time section reported as the interval.
If a device saves the interval data summary and hands it off to TrainingPeaks, TrainingPeaks will display that summary for the laps until you actually drag the selection on the chart.
Unfortunately WKO has no way of knowing that whatever marked the lap marked it in the wrong place. If youād like, you can create custom ranges in any workout in WKO5 following these directions.
Whatās the original source for the power measurement? Is this a Zwift ride or from say a garmin device? Just wondering if you have smart recording turned on and itās affecting the results. I know itās a pain to deal with for very short laps if turned on.
Do you have Strava privacy circles turned on? That can cause Strava to give Intervals.icu bad lap data. If you look at the chart that 30s interval (number 7) misses the 30s power jump. That is why the power is so different.
Just to confirm what @Gerald said. Intervals.icu computes all the interval stats and so on from the power, heart rate and so on traces and does not use the summary information from the device (which may be different). This is so you can edit the power, fix anomalies and so on in Intervals.icu.
