Mining centralization pressure from non-uniform propagation speed

Would it be easy to code in another variable? I think it would be useful to know how a per-node imposed max block size limit (with this limit, a node would still accept all incoming blocks, but refuse to relay any above a specified size) would affect orphan rates and reorgs. Example: What would the orphan rate be of an 8MB block if 80% of all nodes refuse to relay it? The resulting data would provide insight into how nodes can enforce a dynamic maximum block size in absence of a hard limit.

I would do this myself but I am not a coder. I'm hoping someone here who is capable finds this data worth pursuing.