Actually you're not that bad. You only misflag articles 17% of the time.
As for political articles, the test is whether the article is an instance of the ordinary back and forth of debate on some political question, or whether it's about some deeper underlying phenomenon that only happens to be related to politics. E.g. "Politician A accuses Politician B of X" or "Why policy Y is good/bad" are probably both offtopic, whereas "How political campaigns use statistics" is probably ok.
How do you judge misflags? Do you compare against yourself and the other editors kills? Or, do you compare to popular opinion? Which might could be wrong if the community grew too fast. ...Does it help to flag dead articles?
A 17% misflag rate simply means that 17% of the articles you flag don't end up getting killed. Which is pretty good: it means your flags are an 83% accurate predictor of whether an article will be killed.
That has the unfortunate effect of silencing minority constituencies that may have a substantive and HN spirit-compatible reason for thinking that an article is important to get out there and discuss, but whose thoughts do not mirror the prevailing trend of the groupthink.
I suppose it would be a good strategy for those who want their flags counted to flag newer rather than older articles then. The more hours that have passed, the higher the odds are that those who would flag already have.
It would also be a deterrent for those such as myself who flag articles for nauseatingly bad grammar. I don't have that much patience when reading pages with verbless sentences, "there is" vs "there are" mistakes, and that sort of thing. Most 20-somethings do, though. I flagged one just this week after running into a paragraph that started with the following sentence:
Anyway.
The way I had looked at the issue up until now was that this sort of flagging still had some use. Articles would rarely get killed for terrible grammar alone, but if they were also somewhat inappropriate for another reason, then the grammatical errors could push them over the edge.
"As for political articles, the test is whether the article is an instance of the ordinary back and forth of debate on some political question, or whether it's about some deeper underlying phenomenon that only happens to be related to politics. E.g. "Politician A accuses Politician B of X" or "Why policy Y is good/bad" are probably both offtopic, whereas "How political campaigns use statistics" is probably ok."
"Why the US healthcare system is broken" (well written or not) can be judged "politics" by this guideline according to the "Why policy Y is good/bad" part.
That said,
suggestion 1: This would be a great addition to the HN guidelines. Appending the above to "Most stories about politics, or crime, or sports, unless they're evidence of some interesting new phenomenon." would add clarity.
suggestion 2: Can the "misflag rate" be made visible to flaggers as part of their profiles? This would help us improve our flagging.
As for political articles, the test is whether the article is an instance of the ordinary back and forth of debate on some political question, or whether it's about some deeper underlying phenomenon that only happens to be related to politics. E.g. "Politician A accuses Politician B of X" or "Why policy Y is good/bad" are probably both offtopic, whereas "How political campaigns use statistics" is probably ok.