I have a theory about "weird" rejection reasons.
It has become very common for people to complain here and in other forums about their wayspot candidates being rejected for "weird" or "wrong" reasons. It's such a frequent refrain that it's impossible to overlook. Rather than jumping to the conclusion that reviewers have become stupider (which is certainly possible, I'll admit) I decided to think about it systemically. What could have changed? Of the possible causes, which ones are most plausible? Which ones best fit the evidence? I came up with a list of four possible causes:
- Reviewers have deliberately changed their behavior to choose random reasons
- Niantic has changed something in their UI so that people accidentally click the wrong reason more frequently
- Niantic has changed the way they decide which rejection reasons to include in email
- There's a bot or an autoreview script being used that chooses random rejection reasons
Are there any that I'm missing? Maybe. If so let me know what I left out. Here's my analysis of those options.
Reviewers have changed their behavior: I think this is the least likely explanation. If it was limited to a few small geographic areas I would be willing to consider this possibility, but reports are coming in from all over the world and it appears to be a widespread change. It seems entirely implausible that reviewers all over the world have decided in unison to change their behavior.
Niantic has changed the Wayfarer UI: This is a more plausible explanation than reviewers worldwide suddenly losing their minds. If Niantic changed the interface and made it easier for reviewers to misclick/mistap then this change in output of the system wouldn't be surprising. I'm not aware of Niantic having made any changes, but I only review on a desktop computer and can't speak to the mobile experience. Is it plausible? Yes, but only if there have been UI changes.
Niantic generates rejection email differently: This is quite possible. Niantic's system for deciding what goes into rejection email is a magical black box from our perspective. I'm going to make up some numbers as an example. Let's say that a year ago you would get a specific rejection reason included in the email if five reviewers chose it. If Niantic changed the algorithm so that it now only requires two reviewers to pick a reason then people would get a lot more email with "weird" rejection reasons without player behavior changing. Is it plausible? Absolutely, with one underlying assumption: that there have always been reviewers giving silly rejection reasons. Does it fit the evidence? Yes. Remember that Niantic doesn't tell us when they change things under the hood, though we often reverse engineer them by crowd-sourcing information.
Of the above three I think this is by far the most likely as it would be a systemic change and it's one that players wouldn't know happened. In fact, I was pretty convinced that this was the answer... then I thought of a fourth possibility.
Reviewers are using a script or bot: I've heard plenty of talk about local reviewers sharing their submissions with each other and voting in unison on them. I've also heard anecdotal tales of communities that would reject anything that was submitted by someone outside the group. I found a couple of vague references to such a tool but no concrete evidence. It seems certain that such a tool exists, and that a few groups are using it. Is something like this being used widely enough to cause an epidemic of "wrong rejection reasons"? I have no way of knowing. Is it plausible? Yes, though there would have to be a large enough community engagement with it to get this level of outcry. Does it fit the evidence? Yes, modulo the following caveat.
So what's the bottom line? I completely reject the first one. I don't believe that enough reviewers worldwide just started changing their behavior at the same time in a silly way... it doesn't make sense. I reject the second one based on having no evidence of a recent UI change that would cause these sorts of errors... but would re-evaluate if I learned about such a change. Number three seems completely believable to me, though it requires me to accept that there was always some level of rejection-reason noise hidden in the system. Number four seems completely believable to me, though it requires me to believe that there is a system or systems for centralizing data and giving it back to local reviewers. As an engineer I can easily believe this. I can imagine it either giving reviewers instructions for how to vote, or maybe filling in the votes automatically. Do I have hard evidence of a system like this? I do not.