it might be slightly diff train of thoughts in terms of understanding the communication patterns as i think data analysis is more passive and the formulae based variables would be more of an active reaction - but data analysis is an excellent start and would be a great study in understanding the existing behaviors of relays and will provide a lot of input to devs here.
There are some publicly available data on Primal and Nostr.Band but I don't know enough of Nostr or existing data analysis programs to pursue this conversation, but let me tag nostr:npub1nlk894teh248w2heuu0x8z6jjg2hyxkwdc8cxgrjtm9lnamlskcsghjm9c who has been very helpful in providing tech guidance.
Hi nostr:npub1nlk894teh248w2heuu0x8z6jjg2hyxkwdc8cxgrjtm9lnamlskcsghjm9c , Ken and friend are keen on contributing to Nostr from a math-skillset and the conversation is bouncing off various ideas, one is on the potential data analysis to understand the current behaviors of Nostr. I don't know if there is a publicly available data but would you be able to point him in the direction ?