First a quick note as I write this, since the new ConnectStats service was introduced, the new service has now passed the 1 million activities threshold out of 14,000 users who used the new service rolled out in January! It is small compared to the big players, but nice nonetheless.
After working on improving the calculation of the best rolling curves, I have now implemented a first version of an improved summary analysis.
In the new version 6.0 of ConnectStats you can now explore your power curves or best rolling plot and study when you reached you best and compare in detail how you are improving over time.
Quick Recap of Existing Analysis
ConnectStats has been displaying in the summary page a summary of your power curve or best rolling curves for heart rate or speed/pace. It displayed the best curve for the current month, the previous month and the year. This is always one of my favorite graph as it shows well how you have been doing this month, versus the previous or versus the year.
You can navigate this best rolling graph by tapping once to go to the next month or swiping left to go to the next field.
ConnectStats version 6.0 introduces a new graph that you access by doing a long press on the existing best rolling plot. The new page looks like this
The first graph shows you the best achieved value for the current year, and the color tells you during which month you achieved that specific best.
For instance here, in 2020 my best running power over 30 seconds was a bit less than 400W and reached in February, while my best power over 5min was reach in may and was around 300W.
The second graph zooms into the selected month (here May 2020) and shows on which day of the month you reached that best. For example, we saw above than in the year my best power over 5min was around 300W reached in may, in the below graph you can see that that 300W was reached on may 2nd.
You can select the fields you are interested and the fields at the top of the graph as well as the month.
A few things to know
Note that these analysis can be a bit slow to calculate and ConnectStats tries to keep a cache of some of the analysis to display it quicker (summary for the years and the months).
Because it may happen that the cache ends up in a corrupted state or missing some activities, there is the ability to Rebuild the analysis for the current month. This is available on the same screen, but note that it can be a bit slow depending how many activities you have in that month, so use only if you feel the summary looks quite wrong.
As usual, I was only able to test this on my own activities, so if you find something looking quite odd in the analysis, please send me a bug report.
I am now working on another new analysis to show the evolution of a few best reached value over time. I am hoping that this will also be a good indication of your fitness evolution.
I will also update the log scale logic to anchor around better round value (1min, 5min, etc) instead of the current numbers.
Hope some of you will find this useful. Let me know suggestion and feedback.
Thank your for your work. Is kg possible to do a similar feature for walking. I Canon not jog anymore because my back. Instead I use a tredmill with a small upward slope. I don’t need an exact or correct wattage value. Any estimate is OK, because an evaluation of the progress is enough.
I suppose that there are lot of users who are middle aged or elderly and do not want on can run.
Regards Ari Takala, Finland
The app is currently not estimating the power, but analysing the power from a sensor. The sensor I use is stryd for running and it actually works for walking.
Sorry about the grammary. My iPhone was adjusted for Finnish vocabulary. Ari Takala
This is actually a great feature, very clear, quite rewarding to see today’s run contributed to a best along the PDC. Thanks for the continued innovation!
Thanks for the comment!
Brice – can you check the peak power chart – for an individual run it shows at time 00 a much lower peak power than the max power reported in the run. I run some strides most runs, today all of them peaked at 460W-ish (3s power seen on watch). But the peak power chart shows about 375W at time 00.
I think I know why, will fix/adjust
pushed a fix for this issue, will be in next version