Matched Clean Power Index (matched.energy)

36 points by bensg ↗ HN
UK suppliers claim “100% renewable” even when selling fossil power at night.

Newly launched not-for-profit Matched Clean Power Index [1] shows, using open data, each supplier’s true renewable share hour by hour.

Built by a small team of engineers and energy analysts — including a former Tesla engineer — it combines half-hourly data from Elexon (demand), National Grid ESO (generation), and Ofgem (REGOs) to calculate the real renewable fraction for every major UK supplier. It's the first open dataset of its kind [2].

The data exposes a £1 billion-a-year distortion: consumers pay for “green” certificates that don’t align clean supply with demand. Redirecting that could fund storage and flexibility instead.

The best suppliers match 69–88% of their demand with real-time renewables — far better than today’s “100%” claims.

We’d love your thoughts on:

- Next features/datasets: storage, nuclear, or CO₂ intensity?

- API design: what endpoints or update cadence would be useful?

- Visualisation: how would you show renewable matching over time?

[1] https://matched.energy/clean-power-index

[2] https://matched.energy/methodology/v1

9 comments

[ 5.6 ms ] story [ 33.5 ms ] thread
As far as I know, this results from allowing a producer to produce 100% fossil fuel energy and buy certificates to "convert" that into 100% renewable.

It's an artifact of stuff like carbon credits. Companies like apple, google, and amazon declare that they are carbon neutral, or will be by some date, and what they actually do is burn a lot of fossil fuel but "offset" it by paying for carbon credits.

I’ve been told that here in Australia our “Green Power” system where well meaning consumers can pay extra for greener power is similarly broken.

A similar index for Australian power would be pretty useful.

“This is not fraud” - that’s a big claim. There’s no accountability for accuracy, so of course there’s some fraud, waste or abuse.
The fact they have Drax energy at 88% tells me they are counting the giant coal factory converted to burn the worlds forest as renewable, it definitely is not just because its biomass. They also have EDF energy, the nuclear power producer, very low tells us they are counting Nuclear as a none renewable. They aren't looking at temporal moves of energy either, which is a big part of Octopus energies customers who are solar and battery owners and do a lot of energy shifting for the grid. These three components mean the wrong companies are at the top and bottom. This index is useless as it stands its not about who is good for the environment and not.

The fundamental idea we need, but the implementation here is just terrible they aren't promoting the right things and it makes me wonder who is really behind this given the bad producers who have got themselves to near the top.

Thanks for the feedback. You raise three important points that I'll address in turn:

1. Biomass: We report renewables as defined by Ofgem (the UK energy regulator), which for better or worse includes biomass. However, we recognise this is contentious. To address this, we show the breakdown of generating technologies for each supplier so users can see the composition. But regardless, your point is valid and has been raised several times this week, so we have a PR open to add a filter that lets viewers exclude biomass if they disagree with the UK's legal definition.

2. Nuclear: Our initial focus was on renewable power (where intermittency is highest), but we'll be adding nuclear to the index in the next few weeks. This will increase the scores for EDF, Octopus Energy, and British Gas in particular, who purchase the majority of nuclear power. We've called this out in various places on the index (e.g., on the EDF page: https://matched.energy/clean-power-index/edf-energy).

3. Load shifting: Because we measure generation and demand at a half-hourly level, we do account for load shifting that suppliers like Octopus incentivise through agile tariffs. This is actually one of the key advantages of our approach over historic annual accounting; the half hourly temporal granularity means we reward load shifting behaviour.

> burn the worlds forest as renewable, it definitely is not just because its biomass

The world's forest is renewable. Maybe the rate of harvest is unsustainable. Maybe the forest is not being renewed for other reasons like the land is being used for grazing or development. But it is renewable.

These voluntary carbon offset things are only ever going to be a step in roughly the right direction.

It feels wrong somehow to be trying to turn them into a perfect way for committed individuals to opt into paying the full price for their individual contribution to only one aspect of the problem.

Instead we should perhaps take a step back and try to assign the costs of the whole system to everyone, regardless of whether they want to volunteer to pay more to make things happen faster.

It interacts poorly with EV or heat pump adoption where volunteering to pay for the last 5% of green electricity could tip the balance against those purchases. But buying them would lower your carbon footprint at even much lower percentage of clean electricity.

Are green conscious consumers ready for the complexity of paying for 95% clean energy?

Hey, thanks a lot for sharing, a couple comments/questions:

Would it be possible add a little more writeup in tooltips etc to help interpret graphs?

The time series data isn't possible to interrogate on a phone at all and in general I suspect it's too complex for most users, unless you're downloading a dataset and analysing. So I wonder if there's another way of displaying the data that gives some interpretation. Presumably the main figure we care about here is something like "hourly total renewable supply" and "hourly demand" so the energy mix might be better shown as a stacked chart not individual lines?

Perhaps the "non renewable" hourly component could be represented as a grey fill, to help communicate the total mix?

Would it be possible to also show the daily/monthly/annual totals in a sort of balance sheet, so we can see how/where tradeoffs between non-renewable and renewable certificates were made?

Perhaps I'm not explaining myself too well but while the point is to illustrate that at a real time level many suppliers are not matching demand with renewable supply, it becomes a bit confusing because it almost looks like they just aren't meeting demand at all.

Thanks for the work, it's a great project and will no doubt be valuable for policymakers