Mentos- wrote:I think the best way to stop these extreme low or high point gains would be to change the point formula, but I'm sure that will never happen.
If it did though, I think this would be the best way to go:
Current formula: (loser's points)/(winner's points)
New formula: 10 + (.5)(loser's points)/(winner's points)
This new formula would cut the impact of points in half, the minimum point gain/loss would be 10, and if the cap for (loser's points)/(winner's points) remained at 100, and the maximum gain/loss would be 60.
I don't have a problem with the current point system, just thought I'd throw this out there.
This is genius. Not everyone that agrees with the OP is looking for a free ride. Take down the max loss and increase the minimum gain because there is too much volatility ... you hit the nail on the head here in that you should start with a flat rate, and then introduce the regressive correction factor on top of there.
If you guys think playing a cadet for 10 points rather than 2 is somehow going to give unfair advantage, you are just trying to insulate the system against the people (like me) who generally do not pre-judge game quality by rank and will play with people of lesser rank. And that is precisely the reason why the numerous suggestions here to add a feature to limit rank of who can join a game have been rejected by the powers that be - because the site wants to purportedly foster more inter-level play.
Mentos formula would be an all around win. If you are worried about people who join 8-player games with people who deadbeat, then change the formula so that from each deadbeat, you get no points if the game doesn't go a certain number of minimum rounds. And, remember that if you are talking about people who pwn noobs, noobs have 1000 points, so for anyone up to colonel, they're already still getting 8-10 points from this type of conduct even under the current system.
And also it's highly unlikely that anyone on this site will have <10 points. Right now the lowest score on the site is 65.