Evolution top league scores

pier4r
Skynet
Skynet
Posts: 3389

Evolution top league scores

Post#1 » 30 Jan 2017, 09:39

Inspired by viewtopic.php?f=7&t=82&p=639#p639

2017-01-30 09_33_55-gladiabots_analysis_20161130.ods - OpenOffice Calc.png
2017-01-30 09_33_55-gladiabots_analysis_20161130.ods - OpenOffice Calc.png (47.56 KiB) Viewed 1157 times


This is a (non rigorous) proof that the top league slowly is accumulating points. Likely due to matchmaking rules that alllows gathering points and/or due to the influx of new players that "give points to the top" (1). It also true that sometimes this phenomena is balanced by top players stop playing (see the negative progression).

(1) viewtopic.php?f=22&t=126&p=3880#p3880
Last edited by pier4r on 18 Jun 2018, 21:28, edited 1 time in total.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

pier4r
Skynet
Skynet
Posts: 3389

Re: Evolution top league scores 27.11.2016 - 30.01.2017

Post#2 » 30 Jan 2017, 09:41

Interesting enough, mcompany was right: viewtopic.php?f=7&t=82&p=639#p606
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

ntk
Algorithm
Algorithm
Posts: 75

Re: Evolution top league scores 27.11.2016 - 30.01.2017

Post#3 » 31 Jan 2017, 08:43

I think there is a mistake in your stats

I think u need see stat on every single player. I see your and other stat the top player keep points from low players (unfair games).
U have tag low don't have unfair. Now i don't have time for develop ai, but a do a game for mistake: unfair game i was go down in rank, i played vs a noob, i ,i and other things but game was unfair.

On viewing games i was surprised how one top player lose vs low player becouse his sniper searching sniper of enemy (but enemy didn't have it). His sniper was hidle.

Think on it and build stat on fair games u can get some surprise on top clas an low class

pier4r
Skynet
Skynet
Posts: 3389

Re: Evolution top league scores 27.11.2016 - 30.01.2017

Post#4 » 31 Jan 2017, 09:55

ntk wrote:I think there is a mistake in your stats


I think that you may have interpreted differently. Those are not to glorify top players, but only to show the evolution of the scores. How a 1500 now is not as prestigious as a 1500 months ago.
Moreover I did not do the delta on the top1 because the top1 may be just idle or not so active and just retain the score (and it was the case several times), while in the top5 there is always movement. Then having the top5 delta I decided to do the top20 delta instead of the top10.

But then, one can see whatever in the stats, I cannot force on you a certain interpretation.

And yes the problem of matching strong vs "not so good" exist, but is also true that (a) the strong one will get only one point - compared to -20 if he loses, so the contribution is marginal, and (b) we have unavoidable tournaments to see who is the strongest. In the tournament who plays a little has mostly no chances to succeed, unless the others play a little too or are horribly losing. It is by design.
If you want to prove that you are the strongest, play for the tournament (as I did, but I lost to blitz and mcompany)
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

ntk
Algorithm
Algorithm
Posts: 75

Re: Evolution top league scores 27.11.2016 - 30.01.2017

Post#5 » 31 Jan 2017, 12:55

I'm very busy, very

I can't develop ai for every map, wait private match and i will be happy, very happy.

I don't want write on the forum, i want see where u will go..., but can't read without say nothing:

U R DO WRONG

Look to chess and restart all, never see student play with handicap

pier4r
Skynet
Skynet
Posts: 3389

Re: Evolution top league scores 27.11.2016 - 30.01.2017

Post#6 » 31 Jan 2017, 13:36

ntk wrote:U R DO WRONG

Look to chess and restart all, never see student play with handicap


To say absolute negative statements is nothing difficult, those disqualifies your argument. Especially if written like yelling in caps lock.

Second you completely miss the point of this thread. It is like I#m talking about numbers and you talk about cooking. That's odd.

Changing to your topics: yes I agree that the handicap is not nice and it is not addressed by the matchmaking (the matchmaking can just let players of the same XP level match, for example) but it improved a lot compared to the first days of alpha 7.8 (in those day a 2000 could match up with a 1100), I still think the leagues should settle.

For the many AIs, I understand completely but there are many players using just one AI doing well and there are also decorators: viewtopic.php?f=22&t=126&start=20#p4470
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

pier4r
Skynet
Skynet
Posts: 3389

Re: Evolution top league scores

Post#7 » 18 Jun 2018, 21:33

quick analsys done on telegram, comparing the top5 to the top10 player (to cut away possible top players not that active, especially pre ghost era).

top5 (the 5th score in the leaderboard of active players)
top10 (the 10th score)

2016 12 05
1546
1438
+108

2017 01 05
1603
1558
+45

2017 02 05
1843
1728
+115

2017 03 06
1850
1804
+46

2017 04 04
1936
1899
+37

2017 05 04
1980
1855
+125

2017 06 04
2005
1882
+123

2017 07 05
2036
1900
+136

2017 08 05
2007
1924
+83

2017 08 07
2082
1960
+122

2017 10 07
2112
1946
+166

2017 11 05
2171
1857
+314

2017 12 08
2013
1852
+161

2018 01 05
1958
1773
+185

2018 02 06
1812
1731
+81

2018 03 06
2095
1989
+106

2018 04 07
2113
2020
+93

2018 05 08
2234
2052
+182

2018 06 10
2334
2147
+187

Consideration: forget the absolute scores, rather focus on the difference. The larger the gap, the bigger the skill difference between top5 and top 10.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

mcompany
Autonomous Entity
Autonomous Entity
Posts: 872

Re: Evolution top league scores

Post#8 » 19 Jun 2018, 18:55

Like I said on telegram: I don't see the significance of these gaps as one can argue that the 5th and 10th are arbitrary picks that do not properly represent the top 10 well. Also, people generally care about the first more than the 5th, and the 1st isn't represented at all

Return to “General Discussions”

Who is online

Users browsing this forum: No registered users and 1 guest