These trials don't prove anything anyway when cases are so low. At, say, 2k case per day, assuming you're infectious for, say, 10 days, that's 20k people in the country who are infectious, so only a measly 0.03% of the population. So even if you have 20,000 people in Wembley (or enroute), the chances are that only 6 or 7 of them were infectious, so the chances of any kind of "spike" arising are miniscule, especially if most are wearing masks and socially distanced. Same with the beaches and protests last Summer, when cases are so low, these kind of "gatherings" really aren't going to cause spikes.
Same with a nightclub or other "high risk" indoors event. If there are 3,000 people attending, chances are that only 1 of them would be infectious. A place with so many people inside would be pretty large, even if cramped, and there's a limit to how many people you can come into contact with in just a few hours. I seem to recall figures that 1 infectious person could infect around 25 others in the right conditions. So in the scenario of 3,000 people, only even 25 of them catching it still wouldn't cause any noticeable "spike", unless they all went out and infected another 25 each a few nights later. In reality, most of the potential 25 would only pass it to 1 or 2 other people, so no obvious spike. You'd need a few dozen infected people in that nightclub of 3,000 to make any noticeable spike.
The time to do these trials are when case numbers are much higher when there are plenty of infectious people around. I really don't see any point in doing it when case numbers are so low as the results can't be reliable if you don't know how many people were infected before hand. The only use would be if they did know how many were infectious, i.e. if they tested everyone upon entry using accurate tests (i.e. the ones that take a day or so to process, not the fast ones). Yes, the event will be over by the time they know who was infected, but at least it gives more accurate results as to the spread of infection in these venues.