This is mostly unrelated, I'm just a mega fag who actually watches actual Minecraft speedruns in his free time, and I'm not talking the edited videos of the completed runs. I'm the degenerate sitting in the 13 hour long streams watching reset after reset.
Anyway I took a look at the document the speedrun mods put out when they first released it and there is a fairly large flaw in it that people who don't give a fuck about speedruns seem to have glossed over. Their sample size wasn't an even sample of his runs, they selected his 5 best runs from a period of about 2 weeks. When you stream Minecraft speedruns its not uncommon to have hundreds of runs in a single night all completed to various levels. I think it goes without saying if you grab 5 out of hundreds its not hard to find anomaly's
Now I dont watch dreams runs so I don't really have a dog in this fight, I just don't wanna see bullshit math/statistics being used as the next weapon of cancel culture
I'm about the farthest thing from a scientist but I remember being taught something in 7th grade biology about having to be able to repeat data for it it be reliable but what the fuck do I know right