This is a good start on list of problem areas to be addressed.
"1 - problem with the scoring system. Meaning the way net points are awarded.
2 - perceived problems with player rating manipulation commonly called sandbagging.
3 - putting clan goals ahead of personal goals. In other words, players resigning clan games where the outcome has been decided.
4 - collusion amongst amongst several clans to generate points towards one clan."
I would add to the above list: untimely timeouts.
What would anyone think if a player repeatedly ended playable games by timing out? For example (chosen at random):
Game 11523406
Game 11531856 (time out with a bishop pair and a connected passer)
Game 11523407
Game 11531855
Game 11524624
Game 11780024 (timeout on move 7)
Game 11723590 (timeout on move 13)
Maybe there is a perfectly harmless explanation for each of those time outs. Maybe the player was ill, in which case I wish him a speedy recovery. Maybe all those games were lost anyway; he's a stronger player than I am and sees things I don't. But timing out on move seven?! It begins to look like deliberate rating manipulation. Or maybe it's 'clan management.'
So, to no. 3 "resigning clan games where the outcome has been decided" should be added: "untimely timeouts" as another prima facie suspicious means of manipulating ratings (either individual or clan).
"I find it amusing that some waste all their breath wanting to talk about "tossing" games that don't affect the outcome of a challenge… What's wrong with this picture?"
It is not about the result of that one challenge. It is about the effect that tossed game has on the individual's rating, especially if done repeatedly: it artificially lowers his rating and therefore puts him at an unfair advantage against some future player whose rating has not been artificially lowered. Calling it 'clan management' is subterfuge; it's still
sandbagging, whatever the rationalization for it may be.
Regarding a possible solution:
"Change the net points scoring to be the sum of all rating changes for every clan game. For example if you win a clan game against your opponent and your rating goes up 6 points, your clan is awarded 6 points. Every game will contribute to the clan score. So resigning a game will cost the clan."
I've been suggesting something similar to this for a long time. The way standings have been calculated until now
(not collusion) is the fundamental flaw in the clan system. It was flawed from inception, but it became clear only recently, when some clans started blatantly leveraging that flaw.
Getting away from net points is essential to creating a level playing field. The clans who were disrupting the system last year had a grievance which was never until now addressed. Whether that grievance was
justified is another matter; but it must at least be
heard and addressed. Otherwise there will continue to be guerrilla warfare here instead of chess.
Whether a revised clan rating system is to be ELO-based or some other is open to discussion, but in any case, a system in which
every thrown game decreases clan standing is essential to undo the damage caused by a) collusion and b) sandbagging in all its forms including 'clan management' and untimely timeouts. There was a lot of resistance to the suggestion of an ELO-based clan rating system, partly because some people did not understand it and therefore did not understand how it would render sandbagging irrelevant. I emphasize "irrelevant," because some posters here are stuck on the idea that sandbagging must be
prevented and punished. It cannot be
prevented; but it can be rendered
ineffective as a means of manipulating clan standings, and that is sufficient for our purposes of getting on with playing chess here.
The proposal Russ has offered (clan ratings but not based on individual ratings) is not clear on what exactly is being measured. If it is just another way to calculate net points, then it will not solve any of the problems noted above.
A win-ratio-based system would achieve the desired result and is much easier to implement, understand, and explain, than an ELO-based system.
Regarding collusion:
I see no means of preventing it. I do, however, believe that setting the rankings on a fairer system will remove the motivation which originally led to it.
We can nonetheless envision a multi-stage response to it, if it continues after a fairer clan rating system has been implemented and given a fair chance to work.
Stage 1.: set a low limit to the number of same-clan-same-player challenges which will count towards the clan standings. A limit of, say, two challenges in any 12-month period. Any number of same-clan-same-player challenges may be played (some clans just like each other), but only 2 would count towards the standings.
That does not address the issue of multiple feeder clans gifting points to some one clan. So, stage 2.: there should an ad hoc roll-back of ill-gotten points, to be implemented either by Russ (who probably does not want to police the clans) or some sort of adjudication committee (with authority to rule on suspicious cases). This has been suggested before and I have never objected to it; what I objected to was the idea that some people here seem to have that this alone would be sufficient to solve the problems here without addressing the other issues.
If that still does not resolve the issue of gift points, then stage 3.: disqualification of offending clans and removal of all their results (including legitimate ones) from the clan table. This too has been suggested before and I have never objected to it; what I objected to was the idea that some people here seem to have that this alone would be sufficient to solve the problems here without addressing the other issues.
Cheers,
moonbus