Octopus Saving Session

Trapdoor

My result for 30/11 has just been posted to my Saving Session dashboard! 9984 points this time!!

tlcgrantham

I’ve got mine for session 3. Saved £5 which was what I expected using calculations above.

Petriix

ColinLHB said:

Has anybody managed to use the published algorithm to predict accurately what they should get? I have written a script that I believe should work, but it is about 20% off (in different directions) for the first two sessions. So either my script is wrong or Octopus is not using the published formula.

I found similar. I have all my data in an SQL database so just wrote a query to pull the relevant data, but can't make it add up with what Octopus say. I think there's a degree of interpretation about the calculation as it's certainly not clearly worded. 

tlcgrantham

Petriix said:

ColinLHB said:

Has anybody managed to use the published algorithm to predict accurately what they should get? I have written a script that I believe should work, but it is about 20% off (in different directions) for the first two sessions. So either my script is wrong or Octopus is not using the published formula.

I found similar. I have all my data in an SQL database so just wrote a query to pull the relevant data, but can't make it add up with what Octopus say. I think there's a degree of interpretation about the calculation as it's certainly not clearly worded. 

Perhaps Octopus make their own adjustments rather like they do with the Wholesale Gas price for Marex.

[Deleted User]

tlcgrantham said:

Petriix said:

ColinLHB said:

Has anybody managed to use the published algorithm to predict accurately what they should get? I have written a script that I believe should work, but it is about 20% off (in different directions) for the first two sessions. So either my script is wrong or Octopus is not using the published formula.

I found similar. I have all my data in an SQL database so just wrote a query to pull the relevant data, but can't make it add up with what Octopus say. I think there's a degree of interpretation about the calculation as it's certainly not clearly worded. 

Perhaps Octopus make their own adjustments rather like they do with the Wholesale Gas price for Marex.

We know they give 'bonus points' and things like that - but were kind of hoping that the base calculation was understandable.

razord

I've got my 30th November results.

My calculator is exactly 50% of what Octopus says I've saved... but I can't figure out where I've gone wrong. Have Octopus forgotten to halve the 30 minute periods when converting from watts to kWh saved? That's the only thing I can think of that could cause this.

For the first 2 sessions nothing adds up (but, it is much closer to the real number), but third my calculation is exactly 50% lower than Octopus say.

Scratch all that... turns out, I was using kWh 30m measurements as average watts, so I didn't need to divide by 2 at the end!

peter3hg

For the second session my calculation was within about 2% of their result which could well be rounding. I haven't done it for the subsequent sessions.
There are a few pitfalls to watch out for such as excluding previous days that had a saving session for the in day adjustment baseline.

peter3hg

I've calculated the third session using the usage down to three decimal places and I am exactly 1 Octopoint off their result (12,871 vs 12,872) so they are definitely calculating it according to the published method.
For the fourth session I calculate it to 7,433 Octopoints so hopefully see tomorrow what the true result is.

Edit: I've tried recalculating the second session with the accurate data and weirdly I'm still about 2% off their result for that session.

razord

peter3hg said:

I've calculated the third session using the usage down to three decimal places and I am exactly 1 Octopoint off their result (12,871 vs 12,872) so they are definitely calculating it according to the published method.
For the fourth session I calculate it to 7,433 Octopoints so hopefully see tomorrow what the true result is.

Edit: I've tried recalculating the second session with the accurate data and weirdly I'm still about 2% off their result for that session.

I recalculated mine and I'm exactly 1 octopoint under on my calculation too. Maybe they +1 to avoid any rounding errors.

My calculation for 1st Dec is 3943 Octopoints too. Excited to validate now.

If it does work, I'll throw up a version using the Octopus API so folks can check theirs and play with the inputs to see how much they could save.

bristolleedsfan

Extremely low user, first 2 sessions I got 112 and 96 points

3rd session I put oil filled radiator on during same day observation period, used approx extra 2 kWh, got 1416 points.