Banned from reviewing without failing an audit

Question

I have done about 20 reviews today and didn't fail a single one. How can it happen that I'm banned for 7 days? Have the criteria changed somehow? Sometimes the review audits are not 100% clear (there is not only black and white but also gray).

EDIT: according to the very interesting discussion in the comments below I seem to have been manually banned from reviewing. There seem to be a lot of unwritten rules about the rewiewing process which should be made publicly known in some help text (an answer with these rules would be helpful).

Answer 1

There seem to be a lot of unwritten rules about the rewiewing process which should be made publicly known in some help text (an answer with these rules would be helpful). The general text on a failed review is not always helpful.

Yes, two of them:

1. Don't repeatedly goof up and make the wrong call.

2. When in doubt, skip the review and let someone else handle it.

In case that's not good enough, there's also this.

I found the following rules, which where new to me:

• Do not correct code samples, even if your solution runs

This is what we call common sense. It's not about whether your solution runs. If you think you have a better solution, then post a new answer along with a full explanation of how it solves the problem being asked in the question.

Leave other people's answers alone. They wouldn't have posted them unless they thought they were correct. It is more likely that you are misunderstanding what their code is supposed to do or how it is supposed to be used.

If you feel very strongly, you could always post a comment to their answer, explain why you think their code is wrong, and make some suggestions on how to fix it. This way, the person can either say "oh you're right, I goofed up!" and fix the post themselves, or "no, you're doing it wrong; here's how you're supposed to use this code". Now you learned something!

And you especially should never change code in questions, because doing so is almost guaranteed to obscure some sort of problem. Even if you don't think it's the source of the problem discussed in the question, it's still something that answerers might want to comment on (and may already have done so). Silently fixing it doesn't help anyone because it doesn't explain why it was wrong and why your fix is better. If you don't have an actual answer to post, leave this information as a comment to the question.

• all questions about password recovery are SPAM, because they could attract SPAM

By virtue of the use of the word "all", this statement is basically guaranteed to be wrong. I assume it's hyperbole, but hyperbole is a poor way to establish the types of rules that you seek.

The truth is, whether something is spam or not is often a judgment call. Experienced users can often make this judgment rather quickly and accurately, based on their experience. If you aren't sure, refrain from making a judgment.

And as for your particular example:

• The answer is spam because it consists of little more than marketing text and a link to a third-party tool. We don't welcome these on the site. Worse, it violates the clear rules in the FAQ about self-promotion and how to write a good answer.

• The question is not spam, it's just extremely poor. It should be closed for any number of reasons, but not deleted or flagged as spam.

• if I have a different opinion than others, I must be wrong

Hyperbole again? Not constructive.

Brace yourself, though: what you're about to hear may come across as harsh… Yes, in general, if your opinion differs from the commonly-held consensus of others who have been active on the site for a long time, you are probably wrong.

It goes without saying that this doesn't mean you aren't allowed to have your own opinion. You're even allowed to share it. There's nothing wrong with being wrong, per se. Just keep in mind that you're not likely to carry the day unless you can provide a very compelling and persuasive argument in favor of your unique perspective. You've got a lot of expertise and experience to overcome, and you need to articulate why that should be dismissed out of hand. As they say where I'm from: you've got some 'splainin' to do!

Besides, by design, lots of things on the site are vote-based (e.g. question closures and re-opening, and suggested edits). This is inherently a majority-rules system, and there's good reason for that. There are times that I'm convinced I've made the right call based on my experience, but yet I get overruled. Sometimes I remain convinced that I'm right and everyone else is wrong. But I'm a radical.

Answer 2

My question had a lot of response, mainly in comments. Therefore I summarize the results in my answer.

It is possible that a moderator bans you from review, even if you don't fail any audits. In my case the reasons were probably the following (I never got a proper answer):

• I tended to reopen questions after they have been edited and improved, even if the question is not perfect yet.
• I accepted a corrected code sample.

I learned the following lessons:

• before you review, read this. It contains a lot of useful information which is not mentioned in the help.
• do not edit code (unless minor fixes) to improve it, add a comment for the author or write your own answer
• check from time to time how other people vote. If you are the only one voting for or against, there might be something wrong.
• and not to forget: when in doubt, skip the review (of course I knew that before)