Isn't that the same as "last updated?" When I update I get all results to date.

(BTW: Just added a last updated line)

Last updated is when the page was updated, while current to reflects at what point you took results from (e.g. you updated today, but the results are only as current as last week... I don't know how your system works).

I've gone and challenged a bunch of people who were above me

Is the rating used in working out the score the one they had when you challenge them, or the one they have when you win/lose? I suspect the later, but...

If the results aren't current, they aren't current because the source of the game info isn't giving me the most up to date data. I have no way of telling this, but it seems that they get the data directly from the games database, and therefore should be up to date up to the second. I was given this data without a user's manual

Actually, come to think of it, the score may be calculated in the order games were challenged rather than finished, as that is how they are probably stored in the database. The only information I have about each game is the two players and who won, so theres no way to reorder it correctly. That is a problem, but I don't think it should skew the results horribly.

Any sign of the next update? Some filter options would be nice. From looking at the placings so far, I'd say > 100 battles and < 100 RD makes for some fairly accurate statistics.

The other option is to sort by Score-RD (Score-MaxChange).

Scores will hardly ever match, nor will RDs, so there's not much benefit to using either as a secondy filter. If I changed anything, it would be to add a new column showing how high or low someone might REALLY rank depending on their RD, assuming all opponents are accurate.

Yeah, I'd like to see just how far my recent losing binge has dropped me.

Mael, you should also probably not count games that are won by timeout on non-Emerald maps when a Trial player is involved. I know that I have almost a dozen games on Antarticus, Rust, etc. that are almost guaranteed wins because I was playing a person who quit after the free beta.

Unfortunately, I have very limited access to data for this. I would love to do all sorts of filtering options, and I would especially love to have results sortable by map, etc, but there are a few problems:

1) I don't know PHP all that well yet. And I don't have the time to learn it with my job, etc, and my free time is often spent on the clan war.

2) I have limited access to the data. All I get for each game is who plays what side, and who won. I get no information about which map it is played on, how long the game lasted, and whether the game was won by timeout.

The program I use to generate this list was actually written by Qohnil, who also is very busy. Many additional options could be implemented with the data we have now (such as showing only people that have RD < 100 etc), but I don't have the expertise in web development to do that sorting.

So if anyone has the extra time and expertise, let me know, I'll be glad to have some additional help

List is updated. Looks like we have over a thousand players that have finished games now! A lot of newer players have clawed their way to the top of the Glicko scoring.

Woot! Only lost 6 spots!

Just so you know, the "last updated" line still reads April 5

::keeps on clawing::

I don't think my claws are as sharp as Tiger's tho.

Youch, quite the fall for me. I think I'm starting to find my true place

Updated again

Now I know I shouldn't be that high on the list, but how does Zippo get top spot with only 14 wins? Did he beat Tiger 14 times or something?

No wait, that's impossible, because Tiger doesn't have that many loses, heh.

Zippo has an RD of 203, and with only 14 battles under his belt I'd say that's not a great figure for statistical analysis. What it does mean that he has potential and is obviously beating better players... question is, can he keep it up?

The position alone is meaningless, you need to take the RD and number of battles into account if you want to get anything from these figures. Personally, I look for people with an RD < 100 and battles > 100.

in this case....
may be make another glicko rating and include there only players wth over 100 wins
or over 50...this way exclude those "super-n00bs" and make it more accurate..
if this statistical system works well only with large numbers so why include there everybody...

Zippo has a completely undefeated record as of yet. Two of those victories were against Tiger. That will explain his high score.

BTW, just updated the list again.

Why over 50/100 wins and not simply over 100 games regardless of outcome?

Mrakobes: Nice new avatar, by the way.

In other news....25th! Kickass!

See Arty? Those few wins I had over you didn't set you back that far.