Would it be a geometric distribution? That seems to simple an answer. I believe there is a huge spike of 0 subscriber channels (viewers only) and then a small Poisson-like bump somewhere, or possibly large spikes around numbers equal to and smaller spikes around numbers divisible by $10^n$, since people would be drawn to grow their subscriber counts toward numbers they like. In the limit $N \rightarrow \infty$, I would expect the growth of popularity cause the amount of channels with subscriber count $\in [N, N+\Delta N$] to exponentially decline.

Where is the truth? I've googled for a while and found nothing of value. Also, if you have no data, what would you predict?

  • $\begingroup$ This is an empirical question, not theoretical. $\endgroup$ Nov 2, 2020 at 1:42
  • $\begingroup$ This is a little pedantic, but in the limit $N\to\infty$ the number of channels is $0$, because the number of subscribers is finite and less than some maximum. $\endgroup$
    – tromben
    Nov 2, 2020 at 2:10


You must log in to answer this question.

Browse other questions tagged .