Not really sure what happened on this ride, but surely this doesn’t look right:
https://www.xertonline.com/activities/cn1xnjahphpdubcf
The power meter was calibrated prior to the ride (as always) and the power numbers are in line with what I would expect, so why is the MPA tanking like that?
For reference I climbed the same hill the following week, with similar power, but very different MPA profile: https://www.xertonline.com/activities/enpp8qnwtodl2lms
I/We can’t see your previous signature (Advanced MPA tab), but from what I can see, the average power on the first attempt was ~20 Watts higher AND your TP increased by about 20 Watts on your first BT.
You pulled down MPA to TP in about 6 minutes the first time, which isn’t weird. Due to increased TP and the lower average power, that took you 15 minutes on the second.
I’m not sure where you read that the first attempt was harder: this is where being able to define segments and laps would be extremely useful @xertedbrain…
The power from the two attempts was within 2W from TrainingPeaks.
Yes the fitness signature was stale on the first attempt, but I would not expect the model to predict brakthru that early into the climb…
I just zoomed in on the climb for both rides and then compared the data that shows for that particular time frame, but I might have misread it.
To analyze the BT with your old signature, you should go to Advanced MPA and then click Previous, for both efforts…
Like I said, from your first ride, it took you ~6 minutes, which is not weird with a stale signature.
Thanks for the (ahem) useful data. 
To analyze using MPA, having predefined intervals or laps isn’t entirely necessary but you can simply look at the ride and see what’s happened. In the first image you attached, you can see that something isn’t quite right. Usually a lot of yellow dots is a sign that something is wrong.
You can click and drag with your mouse/finger on the chart to choose a section from the ride and examine the average power data. This also allows you to zoom in to see things closer. You can click and drag on the bar too, to move the selected region. This is similar to using a lap / interval feature.
However, once you become more familiar with what Xert is doing, reviewing your average power for a section of your ride is no longer necessary (unless you’re just curious) since the system infers what your power duration curve would have been at the time of your activity. Click on the Power Duration tab and you can see how the MMP (highest average) data from the ride compares with what Xert sees as your Fitness Signature derived values. Only when an effort in your data is perfectly paced, starts from an unfatigued state, ends in complete exhaustion and never goes below TP, will the MMP data match what Xert sees as your best. In any of these cases are not true, you had more to give. Hence, reviewing sections for average power is rarely meaningful unless all these conditions are met. Comparing one effort to another can be done by simply reviewing your Fitness Signature and the MPA data. Comparing averages isn’t really providing a useful comparison.
Let us know if you need any help sorting out the yellow dots. Send a note into support for help and we’ll be happy to look into it.
