Octopus Agile vs Go

[dangermouse]

Since I switched to Octopus I've been playing around with their API that allows you to retrieve all sorts of product info from their server. I'm not currently on either the "Agile" or "Go" tariffs (waiting for a smart meter), but I find it interesting to play with this sort of stuff.

Here's the last 3 days of prices for Agile import (orange), Agile export (green) and Go (blue).



There was a distinct "blip" yesterday when export prices were higher than import for a few hours in the afternoon - I assume demand was very high but import prices are capped.

I was mainly wondering, is it better to be on Go or Agile for cheap battery charging overnight? Certainly from the last 3 days' data, it looks like Go is better.

If we look at the data from the last 30 days (it's a large image so click on the thumbnail to open the Postimage window, then click on the image to zoom, then right click and "Open image in new tab" to see the full size version) we can see the trend repeats on most days - the blue "Go" price is cheaper than Agile for those 4 overnight hours. For my personal use case (charging batteries and then hopefully not using any peak rate power during the day), it seems Go is definitely the better deal.



There was another blip on 11 Nov when the Agile price went negative overnight. If you were so inclined, it'd certainly be possible to use this data to buy electricity at cheap times then sell it back to the grid when the prices rise (and I've read about people on this forum doing just that, although I think the margins are lower than they used to be). You'd need a decent chunk of storage to be able to do that and also store enough for your own use, and of course some way of automatically commanding the inverter to either charge or export based on the day's data. A Raspberry Pi with suitable interface hardware would be the obvious choice. Not something I'll be pursuing, but I'm sure others have.

[Caesium]

https://www.energy-stats.uk/octopus-go-versus-agile/ has some interesting stats on this, I think they have a public Grafana dashboard somewhere, going back quite some way - basically charting the same info as you've done there.

Agile used to be a bit of a no-brainer, a few years ago - it was frequently cheaper than Go, and very often negative, sometimes substantially so. More recently as the energy market changed it tended to flatline up against its upper limit (about 35p/kWh at the time I think) and became extremely bad value if you imported. However, I think you need to be on Agile incoming to get Agile outgoing, and the Agile outgoing prices were the best around, so there were still reasons to be on it if you didn't need to import much (high self-sufficiency via panels etc). If you're on Go then you can only get the smart export guarantee as far as I know.

More recently still Agile was reformed and as you've observed, has become somewhat competitive again. How long will that last? Who knows, that's the catch isn't it

For me personally, Go gives me certainty and confidence that I can always charge my batteries up every night for the same price, for another 9 months before my fix expires. Agile could tailspin into a completely new ballgame at literally any point, so the potential for perhaps saving a few pennies isn't worth it for me; particularly as I don't export (no panels).

Agile is ultimately just a bit more of a gamble than the certainty of Go and it needs to be a heck of a lot lower of an average price than Go to be worth it for me personally, as pretty much 100% of my import can be done in the cheap hours of Go.

[Stinsy]

I was mainly wondering, is it better to be on Go or Agile for cheap battery charging overnight? Certainly from the last 3 days' data, it looks like Go is better.

This has been discussed lots on here and elsewhere over the years. The consistent conclusions have been:

• Long term, Go is significantly cheaper than Agile.

• The occasions where Agile is cheaper than Go are very rare and short-lived.

• With Go you can set-and-forget all your timers, whereas Agile is a full-time job.

[Oldgreybeard]

This has been discussed lots on here and elsewhere over the years. The consistent conclusions have been:

• Long term, Go is significantly cheaper than Agile.

• The occasions where Agile is cheaper than Go are very rare and short-lived.

• With Go you can set-and-forget all your timers, whereas Agile is a fill-time job.

I'd very much agree with this.

I tracked what our bill would have been with various tariffs for a year or so, using data from my home made energy monitor (records readings 10 times every hour). I scraped data for Agile half hourly prices, and used fixed time data for the Go, Go Faster and Economy 7 tariffs (the latter only from our supplier). For us, E7 was always very slightly cheaper than Go Faster, significantly cheaper than Go and a lot cheaper than Agile. The fixed times for Go, Go Faster and E7 are a significant benefit in terms of ease of use and reliability, too.

I've not bothered to look at the numbers at all this year, as the market seemed to be a bit of a mess, but the one thing I'm certain of is that Agile would have remained the worse for our pattern of use, by a long way.

[dangermouse]

This has been discussed lots on here and elsewhere over the years. The consistent conclusions have been:

• Long term, Go is significantly cheaper than Agile.

• The occasions where Agile is cheaper than Go are very rare and short-lived.

Certainly that seems to agree with what we see in the last 30 days of data I downloaded. I could grab more data but visualising and analysing it becomes harder as you get more!

• With Go you can set-and-forget all your timers, whereas Agile is a full-time job.

Although it would be reasonably easy to automate such a system - download the next day's prices every evening, and have a program to control the battery charging or export based on the price. There's a nice looking home made device for controlling Sofar inverters here. Although from my point of view, this is all academic, I don't think I'm eligible for an export tariff, and haven't got enough battery capacity to start market speculation anyway

[Oldgreybeard]

There's a nice looking home made device for controlling Sofar inverters here.

I'm using a slightly modified version of that to get data from our Sofar inverter into Home Assistant, works very well:

Sofar MQTT.jpg (158.51 KiB) Viewed 1351 times

[Stinsy]

• With Go you can set-and-forget all your timers, whereas Agile is a full-time job.

Although it would be reasonably easy to automate such a system - download the next day's prices every evening, and have a program to control the battery charging or export based on the price. There's a nice looking home made device for controlling Sofar inverters here. Although from my point of view, this is all academic, I don't think I'm eligible for an export tariff, and haven't got enough battery capacity to start market speculation anyway

The reality is more complex and less elegant.

Back when Agile was more popular people set their immersion heaters to come on any time the price was below 2p/kWh (or whatever they thought was reasonable). However there were long periods when the price never got that low so no hot water. Alternatively the DHW tank would fill up on 2p electric and have no room for any more just as the price went negative.

The other way to do it is to program your DHW immersion to operate on a fixed timer during the period when Agile is typically cheap. But this often missed the cheapest slots.

I have seen inverters programmed to charge the batteries during a fixed timeslot when Agile is typically cheap and reserve a proportion of that battery for the 4pm-8pm peak period.

[Swwils]

Exports are the MVP with agile...

48p a unit the other day. Exports are almost always well over 12p - so it doesn't make sense to divert solar to DHW at all.

I have Home Assistant and a automated "broker" that trades the energy as required when daily slots released.

Over the last 124 days i've averaged 17p kwh import and 31.06p kwh export. Summer is better, but forced export is real.

[Stinsy]

The real money is to be made by being on Go/IO (or whatever) with a big battery and have your nextdoor neighbour on outgoing.

I've been suspicious for some time that more than one member on here is doing that!

[Tinbum]

The real money is to be made by being on Go/IO (or whatever) with a big battery and have your nextdoor neighbour on outgoing.

I've been suspicious for some time that more than one member on here is doing that!

I should have done that as I have 2 meters, dont even have to go next door.