Fine tuning of the rating-system, SL Options

[sl.ErdnussJack]

It is fact that many players in the Top100 play only once within 7 days for a short time. They don't want to mess up their ranking. Unfortunately, this isn't according to a ranking and to offer such players an incentive for playing and not penalize active players respectively we have made a "short-lived" ranking now.
This means one doesn't play, the rating is made go down easily so (after 24 hours of inactivity). The more inactive player and the higher his rating, the more steeply he falls in the ranking. The whole isn't as bad as it listens to it here.

For example, we have a 20-rating player: If he plays every two days then he loses every two days 0.03 rating points. If he plays only every 5 days then he loses 0.33 each within 5 days. Within 14 days abstinence he loses already 3.02 points. However, these are all values which one can easily make up again. How we recognize does the new modification apply primarily to players who don't play.
Active players are virtually untouched of this and that is also not bad.

In addition, we have reduced the minimum player limit from 8 to 6 players test wise and therefore more players should get a valid session. If the practice tests are positive we keep the new player limit.

[Warmonger]

How the hell can player A who plays xpsave servers only for example where he will have all his skills etc everytime he connects be compared fairly to player B who plays only ETpro servers and always starts at 0 xp

This is exactly my point. Player A in your example will NEVER be compared DIRECTLY to player B, because they never play together. They will only be compared INDIRECTLY.

Player A's rating will be adjusted according to how he performs on HIS server compared to those he's playing against during THAT SESSION, and player B's rating will be adjusted according to how he performs in HIS server compared to THOSE players.

A simplified example:

Player A (XPsave) performs poorly (20 SPM) compared to the people he is playing with (they all got 30 SPM ... same map, same spawntimes, same XPsave same everything because they are all playing together) ... he LOSES points.

Player B (no XPsave) performs well (10 SPM) against the people HE is playing with (5 SPM) ... He GAINS points.

Thus, the NON XP-save guy has just gained ground on the XPsave guy, even though his SPM was 50% lower.

The problem you refer to is how the ladder was BEFORE the new system!!!!!!!!!!

if u have a concrete suggestion for improvement, let it us know

I am sorry if I have offended, I did not mean to do that. I only want to help. The only concrete suggestion I can make is one I already have: the provisional rating. When a new name is scanned (A name with a rating of 0.00), immediately assign that new name a rating that is the ladder's average. for example, a 12...or whatever the average rating of the whole ladder is. This rating is more likely to be closer to that players actual skill level than zero is. THEN, do not allow this players rating to enter into the calculations of others until he had accumulated some rating time, for example: 10 hrs. During this 10 hrs of play, the new players rating can be adjusted as normal.

This post has been edited by Warmonger: Nov 29 2006, 03:43 PM