Energy Statistics Outlier Improvements

I would really appreciate an improved workflow for fixing energy dashboard outliers.

It’s clear that my Zooz zwave switches are not reliably reporting their total_increasing energy usage resulting in the energy dashboard reporting that a single appliance has used 5 million kWh. My current workflow is to stumble on an outlier, open the history for the energy sensor to find the spike, identify the date and time the spike occurred, open the statistics page and select the date and time, adjust the outlier manually.

This hurts since I have quite a few energy sensors and wind up needing to fix multiple values per month. It hurts a LOT when I don’t notice the outlier before the history is wiped… then it becomes impossible to find the outlier in the statistics tab since I can only view 5 values at a time.

It seems silly to have to create my own template sensors or filter sensors for every energy monitor I have.

Would it be possible to come up with one of the following:

  1. Automatically filter anomalous readings from an energy sensor that generates results many orders of magnitude outside the realm of reality?
  2. Improve statistics browser to allow sorting by magnitude or any other way to quickly find huge outliers without needing to find it in the history of the sensor itself to narrow down the search?

Thanks for considering!