So I’m only commenting on this because of your data and methodology. I have no valid experience playing pyro at a high level and can not speak to the discussion about why it should be banned/unbanned.
All data should be done and presented with context. The context here being the scorch shot is overpowered…. But you never explain why it is overpowered. It appears from your posted research that you believe the scorch shot is overpowered because it does a lot of damage and it hard counters sniper. What you should have done and opened your post with is the discussion that the invite-level pyro mains are discussing in this very thread (and the ban thread previously posted). They seem to feel the overpowered nature stems from the knockback, so a better analysis might have been to see the scout deaths pre and post (for a crude example). Your reasoning for the data presented is the FOUNDATION of your analysis.
Your first presented experiment is to determine the effectiveness of scorch shot in high-level matches. What about mid-tier? Low-tier? Look at the amount of players in each tier and see how skewed this data is when you only select the cream of the crop. You also claim to have a small sample size in order to not “muddy the waters.” I know I’m going to sound like a joke for saying this and I know this is just a video game…. but claims like this a fantastic way to get kicked out of a research institute. Your justification (albeit not great) should have been these are all the invite grand finals before and after the ban. If the metric gets exponentially worse as you increase the sample size, it’s probably not a good metric to base a claim on. Perhaps look at ALL matches on product pre and post ban and present trends (or lack thereof) off of that much larger data set.
Your next experiment consisted of an MGE between a sniper and a pyro. You negate all outside variables present in this fictitious matchup that might almost never happen in an in-game scenario without explicitly stating what they might be. A good researcher tries their hardest to prove their own research wrong. I’m not saying this experiment was not useful, but you should try to determine the pitfalls of this experiment. If you can’t, get it… peer-reviewed (ha) by someone with experience on sniper, pyro, HL, etc.
Research presented without foundation or validation is a dangerous weapon. I know this is simply a test of ban vs unban, but I just want to make everyone a little bit better at conducting research!