I typed a whole thing really fast and was going to post until I realized that was really dumb and sadly, that's what most of us are doing here. I say us because I definitely contributed to this when I posted my one liner just because it was on my mind at the moment. This isn't a chatroom, so we can take a while to make proper posts after considering our options. Here's my attempt at that.
Problem 1: data shows that "metas on the second page of the OMOTM voting thread were receiving significantly fewer votes than those on the first page"
Problem 2: bandwagoning
Offered solutions for problem 1:
1.
Randomizing order (mamp)
Rejection reason was provided in his suggestion; makes things hard to find [ctrl+f is a thing tho]. also doesn't solve bandwagonning.
2.
Changing between alphabetical and reverse alphabetical order (mamp)
Issue with it is also provided in his suggestion; middle metas stay the same so the overall effect isn't that different.
3.
Going reverse alphabetical for a few months to see if the page theory is BS or not (me in the post i never posted)
This was going to be the subject of my old post, but upon going back and reading the past conversation, it's really not different enough from the current system (middle metas staying the same means the variation isn't large enough) to be a metric to bust this theory.
4.
Increasing page capacity to 50(+) for every OMotM threads/capping at 25. (ev)
Requires xernforo manipulation + it's just for one thread every month? Yeah, not happening. TI doesn't like a 25 cap because first come first serve could lead to mostly metas disliked my the majority of the community being nominated due to the cap being as small as it is.
5.
Using smogon's polls (kit)
Not enough options iirc. Also, i assume checking people's joindate will be annoying. Idk shit about smogon polls.
6.
Randomly picking 24 metas from the ones nominated: (suggested by me in private months ago during omotm)
This eliminates first come first serve and is fair because rng. Rejected by evg because it could end up having an omotm voting up where the majority don't like the options.
Offered Solutions to both:
1.
strawpoll (mamp)
Easy to rig using proxies.
2.
Setting up a site like Smogon Awards (mamp)
Not possible.
3.
Something I just thought up which is probably how blind votings are done
Creating a new subforum like submissions and have users submit threads to vote. After deadline's over, the threads would be merged into one thread and votes would be counted.
Pros:
-Eliminates the probable page 1 bias, eliminates bandwagonning
-Definitely doable
Cons:
-It's a fuckton of work for the user counting and none for the voters. Workload consists of going through every single entry and noting down the votes in a spreadsheet, adding new columns for a new meta. Furthermore, finding and eliminating voters whose votes do not count. This is assuming 3 votes per person.
-Host would need project mods in said forum to be able to view it/merge threads, so adding new hosts is a pain.
Now you have all the options present. I personally think the only one even worth trying to get is option 3 from "offered solutions to both". However, I would not be looking forward to the workload at all and idk how willing others would be to host it. There's also going to be a need to vet and approve hosts.
Also, if you have other well thought out alternatives, please post them.