Concourse queues at half-time are treated as a staffing failure and are usually a mathematical inevitability. The arrival pattern, not the effort of the people serving, produces them.
Demand arrives as a spike
Almost everyone who wants to buy something does so within a short window after the whistle. The arrival rate briefly exceeds any plausible service capacity, while the same crowd generates almost no demand for the preceding forty-five minutes.
Queueing behaviour changes sharply once arrivals exceed service rate, with waiting time growing rapidly rather than gradually.
A system sized for the average would be idle for most of a match and overwhelmed for fifteen minutes of it.
Service time matters more than server count
Total throughput is the number of service points divided by the time each transaction takes. Halving transaction time doubles capacity without adding a single position, and staff numbers are limited by counter space in a way that transaction time is not.
This is the actual argument for cashless payment and pre-poured products, both of which remove seconds from every transaction.
Seconds compound heavily in a spike, so small reductions produce large changes in the length of the queue.
Choice is expensive at the counter
Every decision a customer makes at the point of service adds to transaction time. Long menus therefore reduce throughput even when everything is in stock, and variety that is valuable at every other moment of the match becomes costly at this one.
Venues respond by narrowing what is available at peak, moving decisions earlier through ordering apps, or separating ordering from collection.
Each approach shifts the decision away from the moment when the queue is being served.
Queue design changes perception and fairness
A single line feeding multiple servers produces shorter average waits and removes the frustration of choosing a slow line. It also prevents the situation where one counter stands idle beside a long queue.
Many concourses cannot accommodate that layout because of space, so parallel queues persist despite performing worse.
Perceived waiting time also depends on whether people can see progress, which is why visible queue structure matters beyond its throughput effect.
Shifting demand is more effective than serving it faster
Encouraging purchases before kickoff or during the first half moves demand out of the spike entirely, which is worth more than any counter improvement.
In-seat delivery and collection points spread arrivals across a wider window and reduce concourse density at the same time.
The clubs that have made real progress on half-time queues did so by flattening the arrival pattern rather than by trying to serve a spike that cannot be served.