Meta observations and statistics

From Gladiabots Wiki
Revision as of 17:17, 25 February 2018 by Pier4r (Talk | contribs)

Jump to: navigation, search

Meta statistics

Recorded best scores

This is a list of all players that managed to reach an impressive league score of 2000 or more in career. The list is currently gathered manually from Gladiabots stats and unofficial Gladiabots stats. The league score inflation distorts the result slightly. If the date of a league score can only be determined approximately the best league score found is added in brackets.

Creation date: 2017-10-12

Rank
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Player League score Date of league score
Miojo 2432 2017-10-07
Milkhunter 2391 2017-10-06
mrchris 2337 2017-10-10 (2280)
Ritter Runkel 2285 2017-07-28 (2161)
LuBeNo 2256 2017-08-19
Pier4r Nvidia k1 2251 2018-02-25
Wozza 2237 2017-04-16 (2177)
Buzzefall 2221 2017-06-05 (2198)
TcT 2219 2017-04-30 (2127)
DH! 2216 2017-10-05
aoi 2194 2018-02-25
chilio2 2191 2017-10-11 (2169)
B_head 2175 2017-07-24 (2169)
acho 2175 2018-02-25
Castlevania 2163 2017-06-23 (2146)
yopi 2152 2017-12
Mcompany 2145 2017-07-10 (1939)
Arcueid_57 2123 2017-09-07 (2114)
T-800 (nullpointer) 2122 2017-05-03 (2050)
Cynic_Alex 2122 2017-09-09 (2113)
Kheops Mk II 2109 2017-04-25 (1929)
Mumpsimus 2074 2017-10-09 (2016)
NullPointer 2068 -
pikey 2062 2017-04-16 (1992)
Tortuga 2062 2017-07-16 (1904)
Kdinos 2057 2017-09-09 (2010)
Don Quijote (Ritter) 2055 -
Tralalo on PC 2053 2017-07-16 (2045)
sollniss 2049 2017-10-04
PEW 2046 2017-06-25 (1920)
MGBlitz81 2029 -
joni la pojat 2025 2018-02-19
0x1de 2025 2017-06-26 (1997)
hairy phil 2012 2018-02-19
Ox1de 2011 2017-09-14 (1982)
Darth Zannah 2004 2017-05-21 (2001)

Historical notes

The core score formula is unchanged since November 2016. The following parameters changed:

  • From November 2016 to January 2017, the top league had a K-factor of 25.
  • From January 2017 to October 2017, the top league had a K-factor of 20.
  • In the first week of October 2017, the top league had a K-factor of 30. Additionally the Master and Grandmaster leagues were merged into one new top league.
    • The changed K-factor from 20 to 30 and the larger league including the ~20 players from Grandmaster and the ~15 players from Master league produced a blossoming of scores for the top players Miojo and Milkhunter that quickly reached 2400 points. This was possible due to more points available for players that dominate even from players with several hundreds of points less. If should be difficult to obtain the same with the reverted K-factor.
  • After the first week of October, the top league had a K-factor of 20. The merged leagues were again split.
  • In February 2018 alpha 12 was released and with it the idea of inactive deployments. See here and here. Those are like "new players" that starts with the score of the last active deployment, so they bring a lot of points in the system that are created out of thin air. Before this, every player had to start with 1000 points, now new "ghosts" players start with 1700, 1900, 2000 and more points according to the last active deployment. In this case it is likely that there will be an inflation effect. An inflation effect was also in the game during alpha 7, 8 and 9 (feb 2017 - Fall 2017) but now it should be a bit quicker so we may see quickly new best scores even with the kfactor equal at 20 in the top league. As a quick explanation: inflation can happen quickly because if a player find out how to win all the "ghosts" in his matchmaking range, he can just get point from them over and over until getting more points is difficult. So those ghosts in this case may work as a ladder to reach higher scores. This is not valid if in the meanwhile, the other player release better versions of their AI.