First and most important: hello there!
I didn't add any personal opinion in my Reddit post about the date that I've found in RtMI web page. It's just a date mentioned is some JSON code and I had no information on how that field is used by Devolver Digital, so it didn't feel right to feed any crazy theory to those nice people in the subreddit.
Here is different, of course, and there is no ethical conundrum in using Mojo users as guinea pigs for a social experiment, especially considering that some of you have already taken, spontaneously and sadly, the unhealthy path that leads to deranged speculations.
So, let me add some more information on those dates and my opinion, which is: that is not the release date of the game and even the month might well be incorrect.
For already released games, that field consistently contains their release date, assuming that the sample of games that I've checked was random enough.
For upcoming games, that field always contain a date set in the future, except for the game "Terra Nil", where the date is set to the 1st of June 2022, and "Wizard with a Gun", that has a date of 1st of August 2022.
All future dates show the first day of a month.
One question that would actually provide some useful information is if already released games showed at least the correct month in that field before being released. According to a quick check on Archive.org, the answer is "no".
For example, "Card Shard" has been released in June and before its release the date in that field was the 1st of April. "Trek to Yomi" has been released in May and the date before its release was the 1st of March. "Cult of the Lamb" has been released just a few days ago and the date before its release was the 1st of May.
Here is my take: when the page of a game is created by Devolver Digital, they mention some very vague intended period in that field. When that date passes and the game is not already out, they don't even bother to update that field, suggesting that this is not a reliable field to consider to make sound predictions.
So, was this investigation good for nothing? Absolutely not!
You have to consider the positive relational outcomes of unwillingly sending people (or even Mojo users) on a wild goose chase. During their naive attempts of making sense of completely garbage data, they enjoy making predictions and they interact witch each other reinforcing their social bonds... which takes me back nicely to the beginning of this post: hello there!