Metrics - No Early Access :(

A while back I enabled early access to a very rough version of the new metrics. I still hadn't figured everything completely out, but the concept was there.

Since then I've made a lot of improvements. There's nothing I want more than to share them with you all, but I had to make some changes that are incompatible with the live version of the site. What that means is I can't trivially give early access to any of the new features.

I'm pretty bummed about this, so really the only thing I can do is provide progress updates. Here's the current state of things. It's not too different from other screenshots I've shared; just a little cleaner.

Setpoint - My average setpoint vs everyone else.

Setback - This one is still running so the data isn't very good, but this will show setback comparisons. A setback is the temperature difference in your home and away profiles.

Balance Point - Where each of the temperature profile lines cross the x-axis. Most useful for heat pumps, but still interesting for everyone else.

Runtime Per Degree Day - My attempt at taking total runtime and normalizing it against degree days to provide a way to compare runtime.

One thing you might notice is that there are very few "others" in most of these metrics. This is because I am doing a better job of picking your comparison group. Number of stages is now included and it's also excluding thermostats that have less than a year of runtime data.

The other thing I've been working on is attempting to do is make these relevant and readable. In short, when someone looks at these I want them to see two things: What their value is and how that compares to other users. This is the closest I've come in my mockups:

I think (and I hope you agree) that it's apparent my cool setpoint is 68°F which is higher than 62% of other users in my thermostat group.

I've designed these metrics a dozen times and I finally feel like I've found something I'm happy with. It makes good use of the space, is easy to understand, and exposes a lot of data without going too overboard. I would love your feedback here.