Whereas as another participant, I see your approach as more akin to a 'griefer' than play tester. I think there is a difference between testing a pre-release game, and disrupting others who are trying to find a community in which to communally understand the world. There are some good parts that would be sad to lose, and even without deliberate misrepresentation the situation is fragile. Please be gentle with your research, and don't take for granted the resilience of the community here.
I just don't agree that HN is "equally-biased", and I have the post history to prove it.
Perhaps those of different biases are more or less likely to flag or vote? That is, could there be many here who believe more strongly in the positives of free speech, and thus are reluctant to flag comments they disagree with, such that a small number of flaggers who do not share this view have a disproportionate voice? I don't know what the vote-to-view ratio is for HN, but I'd presume it's very small, with the flag-to-view being miniscule in almost all cases.
HN is like a "release early, release often" sort of software project, which would require active testers on live code. I sincerely doubt that my sole efforts to study/game the system really made much effect overall to moderation. However, I definitely try to take your advice, and will continue to do so.
What I've found about voting/flagging is it's largely laziness, and those who take the opportunity to downvote first. Most of my posts are either downvoted within 5 minutes, or upvoted to +2 or 3, then actively downvoted to -3. There is usually a bottom on meta-moderation, though; once a post gets to -3, a post is rarely seen enough to be flagged. This goes triple for older submissions; if you make a new post on a day-old thread, even the most vitriolic and offensive post possible, you may get a 0 karma for the post. if you're unlucky. Either way your post will be near the bottom, even if +1.
I've had this account for 3 years now, and I had my first two flagged posts today, which tells you how little they are flagged.
If you're 45 minutes late to a topic, you'll likely read for a few minutes before you get to the bottom grey area. Most readers of comments flake out of a thread before they get that far. The ones that stick around usually notice the grey text, and don't downvote or flag, even if they disagree heavily with the comment. This is why it gets interesting when I experience late downvote-bombings, or someone that was so offended by what I said, that they downvote my post history; it means that someone went looking for me to retaliate. These anomalies provide a lot of information about the mean moderation activity, I think.
I'm seeing that all sides of a debate seem to have equal amounts of apathy when it comes to downvoting or flagging, and I attempt to account for that when playing.
I'm tempted to try and see how long a "Trump Supporters for Systemd" post would last in a linux thread this week, but I probably won't.
I was thinking more of the effect on me, and other users. I try to engage with people and understand why they hold the views they do. The greater the tolerance for users who are faking their viewpoints, the easier it is to fall into the habit of treating someone holding a genuine minority opinion as a troll or shill.
Other than being worried about the gaming approach, I agree with most of the rest of what you are saying.