Fixing Uarenas

#7 Again, not really relevant unless they're abusing the system.

On your second point, that would be considered abusing the system, and it's already not allowed and will be punished.

On your third point – no it's not. Unfortunately it is practically impossible to have a perfect system where nobody will be able to cheat in any way. Everything you've mentioned that shouldn't be allowed, already isn't allowed.

#8 Then that would indeed be a bug. I doubt that's what happened, though. Their lowest rating for the day was a thousand points under their highest. What most likely happened is they made a smurf account and sanbagged, joined a tournament then played some games with their actual strength before the tournament started. Again, this is not allowed, they were breaking the rules by sandbagging.

#9 That is false. You can't join U arenas several days ahead, it's more like 6 hours or so max. This is another thing that makes the system work (for fair players). Increasing your rating after joining a tournament gives you no advantage, it'll only give you tougher pairings. If we're talking actually boosting here, then of course, that is already not allowed (but still isn't an advantage; the advantage you get is by joining a U arena that's capped below your strength, presumably by sandbagging).

#10 I disagree. Ratings can naturally increase and decrease significantly over time. You could choose some arbitrary limit, but even if that was worth considering as an idea (I don't think it is), it absolutely wouldn't be 200 points. Ratings simply aren't that stable.

If long term natural changes are a concern, then calculate maximums over a limited period. Let's say if in the past 3 months you exceeded the tourney limit by 200 points, then you are not allowed in the tourney. I see no reason why somebody who reaches 2000 rating should in 3 months forget everything, but even if this is the case, after some cool down this maximum is forgotten and the player can join an U1800 tourney again.

You can argue about point and time limits, but there should be some reasonable limits which make both abuse very difficult and at the same time don't exclude legitimate palyers.

It may help if exact limits are not public. For example at Stack Overflow exact limits for banning someone from asking questions are not public, but if your questions are downvoted sooner later you will find that you cannot ask any more. In the same vein if a sandbaggers knew that there are some limits on maximum rating, but those limits cannot be exactly known, they would think twice before committing to abusive practices.

@DoTheMath OK, I don't know the details of exact time you can join to the Uarenas, but my point is that you can actually join it beforehand and then get lots of rating points. So the 1900 in a 1500- at the moment of its start is not technically a bug caused by the blatant incorrect algorithm of checking, it's rather exploiting particular aspects of the system combined with sandbagging.

#12 There are many reasons why players would drop over 200 points, and I see it all the time. Lichess knows how to handle sandbaggers. I might be for the idea of using what you're describing as some sort of a flag (which it might be already), but straight up restricting play seems iffy, to say the least. Why not just auto-ban them as well, tournament or not, since we're talking in the context of those who abuse the system? I'm not a fan.

#13 Sure, I just don't consider some players sandbagging to be a bug. What was mentioned in #12 could be a good way to reduce such instances, if in a form of a flag or something of sorts, not restriction. But I don't know how lichess' anti-cheat system works and what it does/doesn't do.

If "𝐥𝐢𝐜𝐡𝐞𝐬𝐬 𝐤𝐧𝐨𝐰𝐬 𝐡𝐨𝐰 𝐭𝐨 𝐡𝐚𝐧𝐝𝐥𝐞 𝐬𝐚𝐧𝐝𝐛𝐚𝐠𝐠𝐞𝐫𝐬" then what are we talking about in this thread?

We are then embracing 𝐋𝐞𝐢𝐛𝐧𝐢𝐳𝐢𝐚𝐧 𝐨𝐩𝐭𝐢𝐦𝐢𝐬𝐦 that we live in the best of all possible worlds.

On the other hand, to claim that there exists no formal system that can decide based on a rating history that this history belongs to a sandbagger is in the 𝐟𝐢𝐧𝐚𝐥 𝐚𝐧𝐚𝐥𝐲𝐬𝐢𝐬 saying that either sandbaggers do not exist, or that physics is not computable, so humans possess some computation skill that a Turing machine does not, to decide who is a sandbagger and who is not.

I'd argue it's not a bug, the entire Uxxxx concept is flawed. I protested the idea before it was implemented because unsolvable problems like this would occasionally occur.

Why is that guy still the winner?

Give the prize to the actual U1500 champion!

#15 We're talking a bunch of nonsense, evidently. But to be clear, I'm not saying sandbaggers aren't a problem, they clearly are. Still, you can't get rid of them, you can only manage them – that's what "handling" is to me.

I don't think the current system can't be improved. I don't even know what the current system is. I also don't think your suggestion is good, but that's just me. Imo, the system in place is fine, and I don't see how a cap you've suggested would make the situation better.