the *decreasing ELO gain if you win against the same players rule*

[RazorOz]

I'm pretty sure the majority of players support removing this rule for all games, it has like the longest running thread on the site, but BGA have shown no interest in doing so, even though the rule does quite literally the opposite of some of the idea behind the rule. Eg. They put forward the idea that ELO isn't a trophy to be sat on, yet this rule encourages not playing, and punishes active players, the best way to keep your ELO high under this formula is to play a few games a month.

It's a ridiculous rule, given if you are 50:50 to beat someone, 7 game win streaks won't even be that rare, and if you're playing all the time you get max punished by this rule, which then defies the point of what ELO is all about, as it becomes no longer as accurate measure of a player's skill.

[Andrei3009]

All this rule does is encourage players to temporary red thumb opponents in case of win streak. It's very annoying and sometimes feels bad but there are no other choice

[joyoy]

I can’t tell if you are being deliberately obtuse, but on the benefit of the doubt let me draw you a picture:

One person owns two accounts, A and B. Account A only ever plays account B, nobody else. And B always throws the game to A. Meanwhile, B also plays a wide variety of other people and thus keeps their Elo up at their true level, but by always losing to A they are siphoning Elo to A and putting them at a vastly elevated level.

None of this makes your game any different from any other. If you can throw the game, this applies.

It is a classic illustration of Goodhart’s Law: “any measure that becomes a target ceases to be a good measure.” Elo has become a target in and of itself.

what do you mean with with one player "throwing" a game? I assume you mean he loses on purpose. that again is cheating and not relevant to what is handled here.
if he doesn''t lose on purpose again specific players, everything mentioned so far applies again. a part of the rule can and should remain: that exact (smaller) part that exclusively addresses cheating.

All that matters for elo is who wins and who loses, player styles and strategies are irrelevant.

All the reason for the consecutive-wins reduction is based on who wins and who loses, player styles and strategies are irrelevant.

There are many reasons against the consecutive-wins reduction, and probably most people are against it, but the extent of interaction in games is not one of them, because it has nothing to do either with elo or with the reduction. All that matters is that someone wins and someone loses.

1. I know that Elo is based on who wins/loses only. I dont know how many more times I should try to explain the same...
the game result determines the Elo - but who wins or loses is determined or influenced by personal style or approach in a lot of / most games!! the more so the more interaction there is.

if we conisder 4 players, a b c d, with the same Elo. if their relative styles or strenghts work in a way that:
a > b > c > d > a
(a wins most against b, who wins most against c, who wins most against d, who wins most against a)

do you still maintain that Elo does not depend on playing style?
- if a never plays against b, and most against d..
- compared to a plays most against b and never against d..
his Elo will be much much higher in the second case. even though initially the Elo of his opponents were the same, but a's compatibiliy was much worse or better.
here the rule is to limit that effect.

on the other side -- now if we have *no interaction games*, it is futile and detrimental. even more reasons by other users were mentioned in this topic.

I'm pretty sure the majority of players support removing this rule for all games, it has like the longest running thread on the site, but BGA have shown no interest in doing so, even though the rule does quite literally the opposite of some of the idea behind the rule. Eg. They put forward the idea that ELO isn't a trophy to be sat on, yet this rule encourages not playing, and punishes active players, the best way to keep your ELO high under this formula is to play a few games a month.

It's a ridiculous rule, given if you are 50:50 to beat someone, 7 game win streaks won't even be that rare, and if you're playing all the time you get max punished by this rule, which then defies the point of what ELO is all about, as it becomes no longer as accurate measure of a player's skill.

All this rule does is encourage players to temporary red thumb opponents in case of win streak. It's very annoying and sometimes feels bad but there are no other choice

thank you. these are even more objections to the rule, that I hadnt even mentioned so far.

It is so obvious to me, and Id say objectively there's so many disadvantages + unfair aspects of the rule, that Im wondering if some others who rejected it in here don't exactly see what is meant or if they really advocate for the rule (while actually presenting almost no reasons yet).
(I'm sorry anyway if I sounded annoyed in a few of my replies)

[Emerson_Matt]

It is so obvious to me, and Id say objectively there's so many disadvantages + unfair aspects of the rule, that Im wondering if some others who rejected it in here don't exactly see what is meant or if they really advocate for the rule (while actually presenting almost no reasons yet).
(I'm sorry anyway if I sounded annoyed in a few of my replies)

I don't want to speak for others, but I'm not advocating for the rule, just saying that your original justification does not hold water. On balance, I'd rather scrap the decay, or at least moderate it so that it's less severe.

The main reason to keep it, to my mind, is to mitigate cheating, which you don't seem to accept. Given that Plan A---people not cheating of their own accord---and Plan B---direct enforcement of the cheating behaviour---don't seem to work, I think Plan C---Elo decay---is not a bad solution.

[BarnardsStar]

what do you mean with with one player "throwing" a game? I assume you mean he loses on purpose. that again is cheating and not relevant to what is handled here.

It is completely relevant. This change was implemented as an anti-cheating mechanism! There is no other purpose to it!

And I should make clear, I do not agree with it. I think it is stupid. But when proposing changes it is important to understand why things are the way they are.

(And yes, that is what it means to “throw” a game—intentionally losing.)

[teallite]

The problem that the Elo reduction is solving is when someone wants to boost their Elo by repeatedly playing the same much-weaker player (possibly fixed, possibly not). This has nothing to do with interaction-levels in the game.

If it is not fixed, then it is not cheating. People did it before Elo reduction, people do it now. There is no shortage of weak players. Also, the margins are so thin in this case, that even 50% reduction is not that hurtful.

[Emerson_Matt]

Agreed, it catches lots of games that should be counted as full Elo, certainly more than are dishonest. The question is whether the trade-off between slowing down cheaters easily is worth the corruption of Elo for honest games. I think there are good arguments on both sides. The OP did not (does not?) engage with this though.

[h_illes]

The question is whether the trade-off between slowing down cheaters easily is worth the corruption of Elo for honest games. I think there are good arguments on both sides.

Another question that should be asked is if there is a better solution. And yes, several better solutions have been proposed in the ELO topic (completely ignored by BGA). My personal favorite is reducing the K factor. Minimal difference in implementation, but does not distort the ELO difference, just slows down changes.

[RazorOz]

Agreed, it catches lots of games that should be counted as full Elo, certainly more than are dishonest. The question is whether the trade-off between slowing down cheaters easily is worth the corruption of Elo for honest games. I think there are good arguments on both sides. The OP did not (does not?) engage with this though.

Cheaters should just be banned, and if you're caught ELO cheating you lose that ELO, indeed isn't this what already happens, so I'm not even sure why this is needed.

There are also so many ways you could code this to hurt cheaters where it less affects ordinary players. For example currently it's linked to time, which is a terrible way to do it, because it's just punishing people who play more, if you linked it to activity, ie. # of games played it would still hit the cheating element, and much less affect honest players, which is the vast majority of players.

[nik592]

Cheaters should just be banned, and if you're caught ELO cheating you lose that ELO, indeed isn't this what already happens, so I'm not even sure why this is needed.

If they are caught, yes. But I think the issue was that "too many" accounts could do this under the radar, so they implemented this to try and automatically deal with it. Like many solutions done by BGA in recent times, it mostly hurts ordinary players, as those who wish to cheat simply find ways to work around the rule change. As h_illes says, other solutions were suggested, but again, BGA doesn't seem that interested in correcting missteps, they just keep on the same path, even when users express issues with it, and/or suggest better alternatives.