On Sunday, 14 July 2024, the Wimbledon final and the final of the European Football Championship took place. Both events were broadcast on television. The Wimbledon men's final, a 5 set thriller between Carlos Alcaraz and Novak Djokovic, was viewed by 11.3 million people (BBC). Up to 22.3 million people watched the football final between England and Spain (ITV).

The question is: which event had the greater impact on household electricity demand?

Below is the load profile for 230 random British households on that day (red line), compared to other Sundays in June and July for 2023 and 2024. The data was collected by smart meters as part of an EDOL pilot study.

The Wimbledon men's final took place from 14:15 to 16:38 (marked in purple) and the football final from 20:00 to 22:00 (green).

What is going on? We don't know. A reasonable expectation would have been for the Euro final to have a profound effect on electricity use, but it didn't.

Of those 22.3 million viewers, some will have been at home, and many in pups. One could expect an increase from TV usage, or a decrease from reduced home occupancy. For these effects to be cancelling each other out is unexpected, but that appears to be what has happened. Perhaps people watch things on Sunday evenings anyway and whether it is Football, Strictly or the Bake Off, it doesn't make much difference. Or TVs are not what they used to be. Old cathode ray tube TVs used to be power hungry, but modern flat screens are less so. And those watching on portable devices have no impact on the metered electricity use at the time at all.

The main difference between 14 July and comparable Sundays is at noon, when demand is noticeably higher, and between 2pm and 4pm, when it is significantly lower. These deviations are significant, because all other Sundays follow a repeatable pattern.

We don't yet know the exact mechanism at play. People may have been glued to the Wimbledon final and therefore suspended many other---more energy intensive---activities. Or it was an indirect effect of the football. In anticipation of the big match, people got out in the afternoon. The weather was good. Similarly, the afternoon and evening events could have triggered the early lunch we see in the 12:00 spike.

Note that the famous 'TV pickup', which results from synchronised use of kettles and toilets (e.g. water pumps), would not show up here. These occur on short timescales of seconds or minutes, not hours as in the above image.

What can we do for a better understanding of such changes?

Three things:

1. increase the sample size
2. increase the resolution
3. collect contextual data

A sample of 230 is relatively small for activities with high societal heterogeneity. Many people do not care about football or tennis and will have watched the Tour de France or done something completely different instead. Any skew in the representativeness of the sample could affect the results. We know that it is representative in terms of income, age and other national statistics. But we don't know how many are football, tennis or cycling fans.

The larger sample size is already in hand: SERL collect smart meter data for over 10,000 homes. For better resolution, we are about to issue 2,000 of them with additional devices that improve the temporal resolution from 30 minutes to less than one minute.

As this example shows, the electricity profile alone doesn't really explain very much. We don't know if people were at home or away, how that impacts their appliance use and what appliances were in use. This is why we are developing new and innovative instruments to collect contextual data. These include surveys, interviews, and sensors that can detect occupancy and appliance use.

Why is this important?

Understanding, predicting and managing (i.e. deliberately changing) the patterns of electricity use will play an increasingly important role in energy systems that rely on large shares of renewables. When supply becomes less controllable, such as changes in wind or sunlight, demand must become more flexible. In some cases that is easy. Charging a car or running a washing machine a few hours later can be achieved with little effort and 'smart technology'. Even hot meals have been shown to be flexible with the right incentive. But when special events in sport or royalty occur, society 'syncs up' and so do electrical loads. This can result in higher costs of provision, higher emissions or even in challenges to our security of supply altogether. Better for some less important uses to be suspended than for everyone to stand in the dark. What constitutes 'less important' is a societal decision, but it is one that we can only make if we understand the patterns of use and the reasons behind them. That is part of EDOL's mission.