Consideration for Agile's potential negative rates after switching to Go

[andrey-yantsen]

When using this script to switch from Octopus Agile to Go, there's currently no mechanism to account for the possibility of Agile rates turning negative in the following day. This creates a risk of missing out on savings, especially during periods of unusually low (or negative) Agile pricing.

The only idea I have on how to resolve this, is to add an extra run at the beginning of the day, to switch to Agile if the average rate there is lower than on the current plan.

[danmed]

The sscript runs at the end of each day and takes into account all pricing in the last period. If Agile has been cheaper then it switches you and it applies from the previous midnight.. There is no need to take into account tomorrows pricing as that will be looked at tomorrow.

[andrey-yantsen]

@danmed thanks for getting back to me quickly and for sharing your thoughts, but it looks like I didn't phrase what I had in my mind properly. Let me try to elaborate on what I mean :)

Knowing, when the prices are negative (or even just zero) on any given day allows automatically switching the heavy load to these times. E.g. if the price is negative between 1pm and 3pm, it might be beneficial to turn on the electric water heater during that period, or charge the batteries from the grid.

When the current plan is Go, that info is not easily accessible from Home Assistant, which will lead to just running the water heater and charging the car only between 1am and 5am (or whenever's the off-peak time there).

[danmed]

I see.. personally i use an app on my phone for that (Octopus Watch).. so i always know what tomorrows agile rates will be regardless of what tariff im on.

…

On Fri, May 23, 2025 at 9:12 AM Andrew Y. ***@***.***> wrote: *andrey-yantsen* left a comment (eelmafia/octopus-minmax#91) <#91 (comment)> @danmed <https://github.com/danmed> thanks for the input, but it looks like I didn't phrase what I had in my mind properly. Let me try to elaborate on what I mean :) Knowing, when the prices are negative (or even just zero) on any given day allows automatically switching the heavy load to these times. E.g. if the price is negative between 1pm and 3pm, it might be beneficial to turn on the electric water heater during that period, or charge the batteries from the grid. When the current plan is Go, that info is not easily accessible from Home Assistant, which will lead to just running the water heater and charging the car only between 1am and 5am (or whenever's the off-peak time there). — Reply to this email directly, view it on GitHub <#91 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AA5S4WWKATUQIVNQFM5QH4D273J6ZAVCNFSM6AAAAAB5YAWIA6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMBTGYZTQNRZGQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[andrey-yantsen]

@danmed yeah, it is an option, but I don't think it works well when you want to automate things as much as possible. In my current setup, at least when I'm constantly on the Agile, the only time I need to check the prices is when I need to understand when to run the washing machine. And even then, I don't really need to know the prices — just the time of the cheapest interval. And most of the other power-hungry regular stuff is run by Home Assistant automatically 🤷

[joeShuff]

Definitely feels outside of the scope of this project. The problem this project solves is to switch your tariff based on your usage and those prices AFTER the usage.

The data for agile prices is publically available so you may need to see if there is a project to always pull in agile prices to HA regardless of your tariff, or write some rest sensors yourself to pull the data in.

It's still within the realm of possiblity that your house turns on applicances during Agiles "dips" but then a different tariff is actually cheaper at the end of the day, this tool switches tariff, then you've turned on applicances which haven't benefited from agile dip, but it would still be cheaper over the whole day.