Energy data

Wind and Prices This Week

Contents

Wind and prices, day by day

Wind is the largest single thing that moves the price of electricity in Great Britain from one day to the next. When it blows hard, wind displaces the gas plant that would otherwise be setting the price, and the wholesale market falls. When it drops, gas comes back and the price goes with it.

This page puts the week’s wind forecast next to the half-hourly prices that have actually been published — not a modelled price, and not a forecast drawn further than the publisher issued it.

The week ahead

Three forecasts of three different lengths. Transmission wind runs 3 days ahead, the embedded wind and solar outlook covers the whole week, and half-hourly prices are published only to Mon 23:00. Empty cells below are the end of a forecast, not a still, dark week.

Day	Wind	Mean	Embedded	Solar peak	Agile price
Monday 21 Septpart day only		5.2 GW	0.6 GW	7.2 GW	40.56p27.54p–56.00p
Tuesday 22 Sept		3.7 GW	0.6 GW	8.0 GW	not published
Wednesday 23 Septpart day only		6.0 GW	1.1 GW	6.7 GW	not published
Thursday 24 Sept	beyond the forecast	—	1.6 GW	10.3 GW	not published
Friday 25 Sept	beyond the forecast	—	1.8 GW	7.8 GW	not published
Saturday 26 Sept	beyond the forecast	—	1.4 GW	7.4 GW	not published
Sunday 27 Sept	beyond the forecast	—	1.9 GW	5.5 GW	not published
Monday 28 Sept	beyond the forecast	—	1.3 GW	5.8 GW	not published

“Mean” is transmission-connected wind, the fleet the system operator dispatches, running 1.5 to 9.1 GW across the half hours this forecast covers. “Embedded” is the wind on distribution networks, which the operator sees only as demand it does not have to meet — they are separate measurements and are never added here unless both are published for the same half hour. The wind forecast is published hourly and each figure covers its whole hour.

Does windy actually mean cheap?

Across the 16 half hours that have both, the windiest third averaged 30.10p/kWh and the calmest third 48.69p/kWh — 18.59p a unit cheaper when the wind was blowing.

Two groups compared, not a trend. Over a single day of overlap a correlation would be an anecdote with a decimal point.

Contains BMRS data © Elexon Limited copyright and database right 2026. Supported by National Energy SO Open Data Tariff data from the Octopus Energy public API. Energy Review is not affiliated with, or endorsed by, Octopus Energy.

Solar outturn

What solar actually generated

Every figure here is a model, not a meter reading. Almost all British solar sits on rooftops and small sites behind the distribution network, where nothing meters it centrally: PV_Live estimates the total by scaling a metered sample of sites up to the known installed capacity. It is the best national figure that exists and it is not a measurement.

Latest complete day: 20 Sept 2026

Generated across the day: 86.2 GWh
Peak output: 13.1 GW
Installed capacity: 24 GW

At its best moment the fleet was running at about 54.8% of what is installed. Solar never reaches its nameplate figure: that number is measured in laboratory conditions, and a British afternoon is not one.

By network region, same day

Estimated solar generation by distribution network region
Region	Generated	Peak
A	14.7 GWh	2.4 GW
H	14 GWh	2.2 GW
B	13.6 GWh	2.3 GW
L	9.9 GWh	1.5 GW
E	7.5 GWh	1.2 GW
K	5.4 GWh	853 MW
M	4.4 GWh	677 MW
J	3.9 GWh	687 MW
F	3.4 GWh	507 MW
D	3.4 GWh	520 MW
G	2.9 GWh	458 MW
N	2.4 GWh	383 MW
P	1.5 GWh	229 MW
C	774 MWh	129 MW

The national figure is modelled separately from the regional ones and is not their total. Across the stored days the fourteen regions add to between 98.5% and 103.2% of it, so the regional figures may never be summed and presented as the national one.

Source: Sheffield Solar, University of Sheffield — PV_Live estimated solar PV generation (national and DNO licence area). Great Britain, and the 14 GB DNO licence areas (PES regions) within it. PV_Live by Sheffield Solar (University of Sheffield), licensed under CC BY 4.0.

Two kinds of wind, counted separately

Great Britain has two wind fleets and only one of them is visible to the system operator as generation. Transmission-connected wind — the large offshore farms and the big Scottish onshore sites — plugs into the high-voltage network, is dispatched directly, and is forecast by Elexon as WINDFOR. Across the 3 days this forecast covers, it runs 1.5 to 9.1 GW.

Embedded wind sits on local distribution networks, and to the transmission operator it is not generation at all: it appears as demand that never materialised. NESO estimates it separately, alongside embedded solar, in the same forecast that gives this page the rest of the week. Over the 8 days it covers, that runs 0.3 to 2.9 GW.

Adding them is legitimate; substituting one for the other is not. Past the point where the transmission forecast ends, the embedded figure is all that is left — and where both are published it is 14 per cent of the two together, so this page leaves the total blank there rather than drawing a cliff that is an artefact of two horizons.

Why the table runs out at different points

Three forecasts, three lengths, and no honest way to make them one. Transmission wind runs 3 days of the week. Embedded wind and solar run the full week. Half-hourly prices are set in the day-ahead wholesale market and published in the late afternoon for the following day, so for most of any given day the price column has nothing to say beyond tonight.

A single chart drawn to the shortest of those would throw away 5 days of the wind outlook. Drawn to the longest, it would invent 7 days of prices nobody has set. The table shows each series for exactly as far as it exists, and labels the end of each.

Why windy does not always mean cheap

The wholesale price is set by the most expensive generator needed to meet demand at that moment. Wind pushes gas plant down the stack, which is why a windy night is usually the cheapest time of the week. But demand matters as much as supply, and so does whether the network can actually move the power to where it is needed.

A very windy day in Scotland with the north-to-south boundary already full is the case where the two come apart: the wind is there, the turbines are paid to stop, and the cost lands in balancing charges rather than in a lower unit rate. That constraint is the largest single line in what it costs to run the system.

If you want the same question at half-hourly resolution and for your own region rather than for the country, that is on when to use electricity.

Frequently asked questions

Why does the wind column stop partway through the week?

Elexon publishes its transmission wind forecast a few days ahead and no further; this reading covers 3 days of the 8 shown in the table. The embedded wind and solar outlook from NESO runs the full week, so the later rows still have those figures. An empty cell is the end of a forecast, not a still day.

What is the difference between transmission and embedded wind?

Transmission-connected wind plugs into the high-voltage network, and the system operator dispatches and forecasts it directly — across the 3 days this forecast covers it runs 1.5 to 9.1 GW. Embedded wind sits on local distribution networks, and over the 8 days its own forecast covers it runs 0.3 to 2.9 GW — the operator does not see it as generation at all, only as demand that never appears. They are measured by different people in different ways, so this page keeps them in separate columns and only adds them where both are published for the same half hour, where embedded wind is 14 per cent of the two together.

How accurate is a wind forecast a few days out?

Good enough to plan around and not good enough to bank on. Elexon reissues this forecast several times a day and successive issues disagree about the same hour, by more the further ahead you look. This page cannot show you that disagreement, because only the newest issue of each hour is kept — what it can do is say which issue it is reading and when Elexon published it, which is printed above the table, so the figure can be judged rather than just read.

Why are there no prices for most of the week?

Half-hourly prices come from the day-ahead wholesale market, and the following day’s rates are set in the late afternoon. Nobody knows next Tuesday’s half-hourly price, including the supplier — so the column is blank rather than modelled.

Does more wind always mean a lower price?

Usually, not always. Wind displaces the gas plant that would otherwise set the price, which is why windy nights are cheap. But the price is set by whatever is at the margin, and a windy hour with high demand, an interconnector outage or a network constraint can still be expensive. Where enough half hours carry both, the page compares them and reports whichever way it came out.