Fair question. I think most scouts' rankings change "astronomically" from month to month during the season, which is why in years past I have only published preseason, midseason and final rankings rather than monthly rankings. However, a lot of these prospects weren't even playing in May, yet experienced fairly big rises or falls. There are a few reasons for that.
For one, I base some of my rankings on my own personal eye test and others on feedback from trusted sources. So once a monthly ranking is published, I will field several texts/emails/DMs on here, etc., saying "this guy is too high" or "that guy is too low" or "you forgot this guy" and more often than not I will adjust accordingly if I trust the source and my follow-up research checks out, especially for the non-WHL prospects that I have only seen on TV, YouTube, GIFs, etc. But I never send my whole list to a source prior to publishing, so once it comes out, they often reach out again and say "I didn't mean that high for Player A" or "Player B is still too low" and that impacts the following month's ranking. I will always try to follow up and get more than one source's opinion on each player that I rank while also doing my own research on every prospect that appears on my lists. So that part applies to a guy like O'Reilly.
Secondly, I requested input from more people for my final rankings than my May rankings, which may have been too heavily weighted on my eye test from the under-18 tournament, which I watched quite closely from afar. I also heard back from more people this time, getting good feedback on the regions that they specialize in and tweaking from there — in some cases "astronomically".
Lastly, as mentioned, I put an added emphasis on career potential for my final rankings. That was always my main focus, but I blocked out some of the "what have you done for me lately" opinions and tried to project what these prospects will be doing in 3-5 years in their prime and bumped up the players that I feel possess the higher upside for the future, not necessarily for next season. That's why a guy like Brady Tkachuk dropped a couple more spots, even though he could make more of an immediate impact in the NHL.
Tough to explain that rationale sometimes, but hopefully it sheds a little light on my thinking and makes a bit of sense as to why my rankings varied "astronomically" from May to June but also from September to February and from February to April and from April to May, etc. Some prospects had really up-and-down years and others were impacted by feedback that I received and applied to my rankings at certain stages throughout the season (even when they were no longer playing games).