![](/images/forum/corner-1.png) |
![](/images/forum/corner-2.png) |
|
![](/images/forum/corners-inner-top.png)
Member 112 posts
Registered: Nov 2010
Disclaimer #1: These stats contain noise (players aliasing, not trying, goofing off, etc.). Take them with a grain of salt. Disclaimer #2: These daily stats will be from foppa only. (I don't have time to download from all servers). See iELO link for information about ratings2/1/2013 (Edited 2013-02-10, 06:18)
News Writer 646 posts
Registered: Mar 2006
Member 112 posts
Registered: Nov 2010
xterm armor hog? Appears so, maybe heddan is giving them all to him ![](/images/smileys/emoticon_smile.png)
Member 112 posts
Registered: Nov 2010
2/2/2013 Who is amfetaminbengt!? He played one game and had high score, thus his rating is high. More games (data) is needed to make the rating more accurate. Some of those games were between 3p5 and -0-
Administrator 1025 posts
Registered: Apr 2006
Who is amfetaminbengt!? He played one game and had high score, thus his rating is high. More games (data) is needed to make the rating more accurate.
I think its nEpra ![](/images/smileys/emoticon_smile.png)
Member 112 posts
Registered: Nov 2010
NEW! Points System - I added a points system so there are now 2 metrics for looking at the mix results: ratings, and points. I will provide screenshots of both. The points system is calculated as follows: Participation Points: +10 per game Win vs lower rated opponent: +50 per win Win vs higher rated opponent: +100 per win Examples:1. Lose: +10 points 2. Win vs easy team: +10 + 50 = +60 3. Win vs hard team: +10 +100 = 110 2/3/2013RatingsPointsTOTALS (2/1 - 2/3)RatingsPoints
Member 112 posts
Registered: Nov 2010
niw had an amazing mix game. Thus his rating is very high. A few more games and his rating will come down to reality. kapo had a mix marathon this past day, racked up a bunch of points. (14 games, 6 wins). Looks like someone is in it to win it. I wonder if anyone is even looking at these stats and points. 2/4/2013RatingsPointsTOTALS (2/1 - 2/4)RatingsPoints
Member 226 posts
Registered: Jun 2006
would be dope if the stats.quakeworld.nu site had an api so your interface could pull data from there somehow that way people wouldn't have to check your screenshots and could just look at them through your program. no way to do a true player filter though i suppose because we don't have authentication.
Member 112 posts
Registered: Nov 2010
would be dope if the stats.quakeworld.nu site had an api so your interface could pull data from there somehow that way people wouldn't have to check your screenshots and could just look at them through your program. no way to do a true player filter though i suppose because we don't have authentication. I couldn't agree more. I asked pleura for a very basic text API a while back, but we never seemed to sync up. I would go away for weeks/months, then he would. if we use stats.quakeworld.nu, demolyzer would be able to provide #1 and #2 of the following: 1. Ratings 2. Points 3. Stats 4. Stat ratings MVD demos are required for #3 and #4 (which I find very interesting and useful). I would really like to make a version of demolyzer that can spit out a TXT file that some web guru can then use to PHP it and display it nicely instead of my crappy huge PNGs.
News Writer 221 posts
Registered: Jan 2013
Great stuff! Now I can have a purpose with all this QWing: Climbing up from 2nd worst place!
Administrator 1265 posts
Registered: Jan 2006
good stuff. keep it coming ![](/images/smileys/emoticon_smile.png) never argue with an idiot. they'll bring you back to their level and then beat you with experience.
Member 112 posts
Registered: Nov 2010
Great stuff! Now I can have a purpose with all this QWing: Climbing up from 2nd worst place! Sorry but you are now last place. ![](/images/smileys/emoticon_tongue.png) I removed the game with B3... he timed out and had 0 score. I need to fix that sometime in demolyzer. But shortly we will see where you stand after I update the ratings from today's games...
Member 112 posts
Registered: Nov 2010
Nice job erlend you moved up with 3 wins. ![](/images/smileys/emoticon_smile.png) dimman's rating is so high for two reasons: 1. he scored massive with some weak teammates and decent opponents and 2. He only played 1 game so it is something called statistical independence with ELO. Once he plays more his rating will drop. 2/5/2013RatingsPointsTOTALS (2/1 - 2/5)RatingsPoints
News Writer 221 posts
Registered: Jan 2013
Nice job erlend you moved up with 3 wins. ![](/images/smileys/emoticon_smile.png) ![](/images/smileys/emoticon_grin.png)
Member 112 posts
Registered: Nov 2010
Here comes milton. And kapo is a mix machine. 2/6/2013RatingsPointsTOTALS (2/1 - 2/6)RatingsPoints
Member 112 posts
Registered: Nov 2010
2/7/2013RatingsPointsTOTALS (2/1 - 2/7)RatingsPoints
Member 112 posts
Registered: Nov 2010
2/8/2013RatingsPointsTOTALS (2/1 - 2/8)RatingsPoints
Member 112 posts
Registered: Nov 2010
Are there any PHP gurus that would be up to the task of writing the code to display these stats (instead of these large pictures). I can export the data to a txt file that the PHP can read, if that is even possible (I don't know php). I will then host the php page instead of the images.
Member 226 posts
Registered: Jun 2006
Are there any PHP gurus that would be up to the task of writing the code to display these stats (instead of these large pictures). I can export the data to a txt file that the PHP can read, if that is even possible (I don't know php). I will then host the php page instead of the images. cough*phil said he wanted to*cough
Member 112 posts
Registered: Nov 2010
This is the end of 2013 Week 6.I will start the ratings over each week. dimman only had highest rating because he played just 1 game, which consisted of an incredible performance. See iELO link for information about ratings.Check out a milton blowout. He had a 99 fragstreak: 2/9/2013RatingsPointsTOTALS (2/1 - 2/9)RatingsPoints(Edited 2013-02-11, 06:15)
News Writer 646 posts
Registered: Mar 2006
Member 112 posts
Registered: Nov 2010
Member 112 posts
Registered: Nov 2010
This is the start of 2013 Week 7.See iELO link for information about ratings.2/10/2013RatingsPoints
Member 112 posts
Registered: Nov 2010
Week 7 day 2squeeze is going to squat at the top. See iELO link for information about ratings.2/11/2013RatingsPointsTOTALS (2/10 - 2/11)RatingsPoints
Member 5 posts
Registered: Mar 2012
need to level up my game.. :/
News Writer 221 posts
Registered: Jan 2013
need to level up my game ... :/ You're not the only one! I'm going steady at -170 and dreaming of seeing just two numbers ![](/images/smileys/emoticon_smile.png)
Member 112 posts
Registered: Nov 2010
Week 7 day 3Carapace on fire See iELO link for information about ratings.2/12/2013RatingsPointsTOTALS (2/10 - 2/12)RatingsPoints
News Writer 221 posts
Registered: Jan 2013
Woho! 50 point improvement!
Quick question, Cyanide: I played this weeks first mix games yesterday. The day-ratings show -120 and the totals show -111 based on the same two games. Shouldn't those be the same?
Member 112 posts
Registered: Nov 2010
Woho! 50 point improvement!
Quick question, Cyanide: I played this weeks first mix games yesterday. The day-ratings show -120 and the totals show -111 based on the same two games. Shouldn't those be the same? Excellent question. But the answer is no, they should not be the same. The reason why is because iELO is an average of all 100% of every game played for every player. That probably does not make sense so here is an example that should hopefully: Lets say erlend plays milton and loses 10-9. If that is the only game played, then milton's rating might be +5 and erlend would be -5 (just hypothetical). Then lets say erlend goes away for a week, but milton keeps playing and kicks ass all week. And at the end of the week milton has a 300 rating. Since you almost tied milton, and you only have 1 data sample to average (1 game) your rating is still 10 lower than miltons, so you would have 290. So if you play a mix at the beginning of the week, your rating will change up or down throughout the week depending no how well the people you played do during their games, and the people they play, etc. It is all connected, which is nice. An iELO number by itself means nothing. So if your rating goes up or down, it doesnt mean too much. What matters is where you sit in relation to others. Pick out some players with similar ratings, and see if you can do well enough to pass them by in the ratings. Make sense?
News Writer 221 posts
Registered: Jan 2013
Makes perfect sense, thanks!
![](/images/forum/corners-inner-bottom.png)
|
![](/images/forum/corner-3.png) |
![](/images/forum/corner-4.png) |
|