The RSPS Vote System That Protects Real Votes

The Half of the Problem Nobody Talks About

Every RSPS list will tell you it blocks fake votes. It is the easy claim to make and the easy half of the problem to attempt. Tighten the filters, reject anything that looks unusual, and you can honestly say you are stopping manipulation.

The half that almost nobody talks about is what happens to the real votes while you are doing it.

This is where vote verification actually gets hard, and it is where the difference between a mature system and an immature one becomes obvious. A list can block manipulation and still produce rankings that are completely wrong, because in the process of blocking it, it destroyed a large share of the genuine votes real players cast. The manipulation stopped. The rankings are still meaningless. Nobody notices, because the failure is invisible from the outside and the list can still truthfully say it fought fake votes.

We have spent over two decades on this problem, and protecting real votes has always been treated as equally important as blocking fake ones. Not as an afterthought. As half the actual job.

 

Why This Failure Mode Is So Common

The reason so many verification systems end up stripping genuine votes is not carelessness. It is a predictable consequence of how these systems evolve under pressure.

A list starts with reasonable filtering. Manipulation attempts increase, as they always do, and some start getting through. The natural response is to tighten. Reject more aggressively, widen what counts as suspicious, lower the threshold for flagging. This feels responsible and it does reduce the fake votes getting through.

What it also does is make the system blunt. Aggressive filtering cannot distinguish carefully. It catches genuine unusual activity along with the manipulation, because from the outside they can look similar. Multiple players voting from the same household network. A legitimate surge after a popular video. Players voting from mobile data and then from a computer. A server with a large enthusiastic community producing high volume precisely because the support is real. All of this can trip an aggressive filter, and increasingly does as the filter gets more aggressive.

So the tightening continues, the false positives accumulate, and the system quietly transitions from filtering manipulation to filtering activity. Real votes disappear in volume. And the servers that suffer most are the ones with the largest genuine communities, because their real support generates exactly the kind of volume the blunt filter has learned to distrust. The honest servers get penalized while the manipulation, run by people specifically studying how to appear normal, often still slips through. The filter ends up catching the honest signal and missing the dishonest one.

This is the trap, and it is not hypothetical. It is what happens to any verification system that responds to pressure by getting more aggressive instead of more precise. Once a list is on that path, the rankings degrade steadily and there is no obvious moment where anyone notices it happened.

 

Precision Is Harder Than Aggression

The distinction that matters is between a system that is strong and a system that is merely strict, and they are not the same thing.

Strictness is easy. Anyone can reject more. If your only goal is to stop fake votes and you do not care what else you destroy, the problem is trivial. Reject everything that is not perfectly typical and you will block nearly all manipulation. You will also block an enormous amount of legitimate support, but the manipulation number will look great.

Precision is hard. Distinguishing a genuine unusual pattern from a manufactured one requires understanding the difference at a level that surface signals cannot provide. It requires having seen enough of both over enough years to know how each actually presents. It requires a system built to make graduated judgments rather than binary ones, and the accumulated knowledge to make those judgments correctly. There is no shortcut to this. It comes from long experience with the specific problem, in this specific scene, against the specific ways manipulation shows up here.

That is the work we have put in, continuously, for as long as this list has existed. Not building a stricter filter. Building a more accurate one. The goal has never been to reject the most votes. It has been to reject exactly the right ones and leave everything else untouched.

 

What We Will Not Do

We do not compensate for verification weakness by stripping genuine votes.

This is a commitment with real consequences, because the alternative is genuinely tempting. When manipulation pressure increases, the fastest way to make your numbers look clean is to start rejecting more broadly. It works immediately, it requires no additional engineering, and the servers whose real votes you destroyed have no way to prove what happened. It is the path of least resistance and it is the path a verification system takes when it cannot solve the problem properly.

We have never taken it. When manipulation patterns change, and they change constantly, the response has always been to improve the system's accuracy rather than widen its net. That is slower, it requires far more ongoing work, and nobody outside can see the difference. But it is the only approach that produces rankings that actually mean anything, and it is the reason a vote cast by a real player on this list is a vote that counts.

For obvious reasons we do not publish the specifics of how the system makes these determinations. Detailing the mechanics would be handing a roadmap to exactly the people trying to defeat it. What we can say is that the system is layered, that it is continuously developed against current manipulation patterns rather than left static, and that accuracy in both directions is the standard it is held to.

 

Why This Matters for Servers

If you run a server, this distinction affects you directly and in ways that are easy to miss.

Under a weak verification system, your position depends on whether you are willing to compete with whoever manipulates hardest. Building a good server and earning genuine support does not help you, because genuine support is not what the rankings are measuring. That is a race no honest server can win.

Under a blunt one, you face something worse in a way, because it is invisible. Your real votes get discarded, your position drops, and there is no explanation and no recourse. You did nothing wrong. Your community did nothing wrong. The filter simply decided your genuine activity looked suspicious, and you have no way to know that is what happened or to appeal it. Servers with the strongest communities are the most exposed to this, because their real support produces the volume that blunt systems distrust.

What you actually want is a system strong enough that manipulation does not pay and precise enough that your genuine support is never treated as collateral damage. Those two things together are what make a ranking worth competing for. Either one alone produces a list that does not reward you for building something good.

 

Why This Matters for Players

For players it reduces to one question. When you look at the rankings, do the positions tell you anything true?

If fake votes get through, the answer is no, because the top is a measure of who manipulated best. But if real votes are being destroyed, the answer is also no, and this is the part most people never consider. A list that discarded a large share of genuine support produces rankings just as disconnected from reality as one that let fakes through. The distortion runs in a different direction. It is still distortion.

Rankings are only informative when both halves are handled correctly. Fake votes blocked, real votes preserved. When you see a server near the top of our list, it is there because real players who actually play it chose to support it, and because that support was recognized rather than filtered away. That is what makes the position mean something and what makes it worth using as a signal for where to spend your time.

 

The Standard We Hold

Vote verification is invisible work. Players never think about it. Server owners only notice when something goes wrong, and often not even then, because the failure mode we have described here leaves no trace anyone outside can see.

That invisibility is exactly why it requires an explicit standard rather than a best effort. It would be easy to let accuracy slip, because nobody could prove it. The only thing preventing that is a list deciding that getting it right matters more than getting it easy, and then holding to that decision year after year as the pressure and the sophistication of manipulation keep increasing.

That has been the standard here since the beginning and it has never been relaxed. Not when manipulation volume spiked. Not when it would have been simpler to tighten and move on. The system gets more accurate, never just stricter, and genuine votes stay genuine votes.

 

Where to Look

If you want rankings built on that standard, browsing the RSPS list is the most direct place to start. What you see reflects verified votes from real players, with a full monthly reset so the picture stays current rather than historical. The servers near the top are there because their communities put them there and because that support was recognized properly.

Blocking fake votes is the half everyone claims. Protecting real ones is the half that separates a list you can trust from a list that merely sounds trustworthy. Doing both correctly, continuously, against pressure that never lets up, is the entire job. It is unglamorous and nobody applauds it, but it is the foundation everything else depends on, and it is why these rankings still mean something.