Here is a brief expansion of the network science point that, subject to certain assumptions (!), the number of possible ties between people in a group rises supra-linearly with the size of the group. So transmission risk (e.g. of a virus) rises rapidly with group size. 1/
That is, bigger groups provide more and more opportunities for a virus to spread between pairs of people. 2/
A small funeral of 10 people offers the virus about 50 chances to spread between pairs of people, but a small party of 100 people (10 times the size) offers the virus about 5,000 such chances (that is, 100 times the number of chances). 3/
Let N be the number of people in a group. If we assume (key assumption!) that every person in a group has a chance to meet every other person, the number of possible connections within a group of N people is given by the formula N(N-1)/2, or, approximately, N^2/2. 4/
So bigger groups offer explosive potential for virus spread, as summarized in the following simplified and illustrative table. 5/ 
Of course, there is a lot of epidemiology that is relevant here too! Outdoor events, or those in the summer, or those where people are wearing masks or keeping their physical distance, or where people have higher levels of immunity, all reduce the risk. 6/
But with such a large number of possible connections, as groups get larger, even a very small risk of transmission (e.g., 5 minutes being very near someone who is wearing a mask, outside) means that a large number of cases could result, given the possible multiplier. 7/
To be clear, this simplified calculation (!) is about the *opportunities* large groups present for transmission. Risk could saturate at much smaller group sizes (so N=10,000 may be no different than N=100,000). This will depend on crowd flux, mixing, choke points, etc.. 8/
