I’ve been running the same test multiple times with the same seed and network profile and diffing the JSON. The timings move, but the event structure doesn’t, which helped me reason about cause vs effect more clearly.
If there are existing tools or workflows that already give you this kind of repeatable structure, I’d genuinely love pointers.