• Log InLog In
  • Register
Liquid`
Team Liquid Liquipedia
EDT 00:35
CEST 06:35
KST 13:35
  • Home
  • Forum
  • Calendar
  • Streams
  • Liquipedia
  • Features
  • Store
  • EPT
  • TL+
  • StarCraft 2
  • Brood War
  • Smash
  • Heroes
  • Counter-Strike
  • Overwatch
  • Liquibet
  • Fantasy StarCraft
  • TLPD
  • StarCraft 2
  • Brood War
  • Blogs
Forum Sidebar
Events/Features
News
Featured News
[ASL19] Finals Recap: Standing Tall9HomeStory Cup 27 - Info & Preview18Classic wins Code S Season 2 (2025)16Code S RO4 & Finals Preview: herO, Rogue, Classic, GuMiho0TL Team Map Contest #5: Presented by Monster Energy6
Community News
[BSL20] Non-Korean Championship 4x BSL + 4x China2Flash Announces Hiatus From ASL63Weekly Cups (June 23-29): Reynor in world title form?13FEL Cracov 2025 (July 27) - $8000 live event22Esports World Cup 2025 - Final Player Roster16
StarCraft 2
General
Program: SC2 / XSplit / OBS Scene Switcher The SCII GOAT: A statistical Evaluation Statistics for vetoed/disliked maps Weekly Cups (June 23-29): Reynor in world title form? PiG Sty Festival #5: Playoffs Preview + Groups Recap
Tourneys
RSL: Revival, a new crowdfunded tournament series FEL Cracov 2025 (July 27) - $8000 live event Sparkling Tuna Cup - Weekly Open Tournament WardiTV Mondays Korean Starcraft League Week 77
Strategy
How did i lose this ZvP, whats the proper response Simple Questions Simple Answers
Custom Maps
[UMS] Zillion Zerglings
External Content
Mutation # 481 Fear and Lava Mutation # 480 Moths to the Flame Mutation # 479 Worn Out Welcome Mutation # 478 Instant Karma
Brood War
General
SC uni coach streams logging into betting site Player “Jedi” cheat on CSL Flash Announces Hiatus From ASL BW General Discussion Practice Partners (Official)
Tourneys
CSL Xiamen International Invitational [BSL20] Non-Korean Championship 4x BSL + 4x China The Casual Games of the Week Thread [BSL20] Grand Finals - Sunday 20:00 CET
Strategy
Simple Questions, Simple Answers I am doing this better than progamers do.
Other Games
General Games
Stormgate/Frost Giant Megathread Path of Exile Nintendo Switch Thread What do you want from future RTS games? Beyond All Reason
Dota 2
Official 'what is Dota anymore' discussion
League of Legends
Heroes of the Storm
Simple Questions, Simple Answers Heroes of the Storm 2.0
Hearthstone
Heroes of StarCraft mini-set
TL Mafia
TL Mafia Community Thread Vanilla Mini Mafia
Community
General
US Politics Mega-thread Russo-Ukrainian War Thread Summer Games Done Quick 2025! Trading/Investing Thread Things Aren’t Peaceful in Palestine
Fan Clubs
SKT1 Classic Fan Club! Maru Fan Club
Media & Entertainment
Anime Discussion Thread [Manga] One Piece [\m/] Heavy Metal Thread
Sports
Formula 1 Discussion 2024 - 2025 Football Thread NBA General Discussion TeamLiquid Health and Fitness Initiative For 2023 NHL Playoffs 2024
World Cup 2022
Tech Support
Computer Build, Upgrade & Buying Resource Thread
TL Community
Blogs
Culture Clash in Video Games…
TrAiDoS
from making sc maps to makin…
Husyelt
Blog #2
tankgirl
StarCraft improvement
iopq
Trip to the Zoo
micronesia
Customize Sidebar...

Website Feedback

Closed Threads



Active: 650 users

So I made a rating system

Blogs > TheBB
Post a Reply
TheBB
Profile Blog Joined July 2009
Switzerland5133 Posts
Last Edited: 2012-12-02 19:06:30
December 02 2012 16:37 GMT
#1
Edit: Here is a PDF with some math.

Well, if you've followed my blog and my work recently, you may have noticed that I have been complaining a bit about the available rating systems in SC2.

I need a reliable rating system to predict tournament results. (You may have seen me try to do this lately in LR threads.) A good rating system for SC2 provides the following:
  • Accurate results.
  • Matchup-specific indicators, and not just general ratings.
The two I have been using are TLPD Elo and SC2Charts.net (hereafter called SC2C).

TLPD provides an implementation of the Elo rating system, which is old and tested, and has been seen to perform reasonably well in chess.

SC2C instead implements the Glicko rating system, which in addition to the rating also tracks the uncertainty. This allows the rating of new players to be adjusted much more quickly than Elo, while the rating of veterans which we are very sure of is more stable. This is a significant improvement over the Elo system.

Both these systems are intended to be used in a periodized fashion, i.e. the rating of players is only updated in discrete time intervals. The length of the periods must be adjusted to give a good balance. If they are too short, the system will tend to over-react, and if they are too long, the system will have poor resolution in time.

From what I can see, TLPD is not periodized, and SC2C may or may not be.

Moreover, TLPD has a sort of half-assed attempt at giving matchup-specific ratings, done in a way that I'm pretty sure the Elo system was never intended to do, but at least it's better than SC2C which does not provide any matchup-specific ratings at all.

SC2C also produce ratings on a different scale than the traditional Glicko algorithm, and they do not publish rating uncertanties at all, which makes it difficult to use these ratings without magically conjuring up some numbers. (Which is what I'm doing at the moment.)

Add to this the fact that I think the SC2C website is terrible and has a search functionality that must have been new in 1910, and that TLPD is being updated once in a blue moon, I wasn't very happy with either of them. (They know about it but the international database still only lists a handful of games after September.)

So what I've done now is that I have read Mark Glickman's 1999 paper where he describes the Glicko algorithm in detail, and I have tried to generalize this to a system that is specifically designed with matchup-specific ratings in mind.

So, I'm pretty sure you want to see some results! Please note that the below have been produced with a prototype implementation, where the parameters in my method have not been fitted. I chose "sensible" values, but some testing is needed to check if other values would be better.

My ratings are usually in the interval -1.0 to +1.5 or so. Probably later I want to change the scale so that it looks like something similar to the Elo scale, but for now this will have to do.

The ratings below are based on a snapshot of the SC2C database from a week or so ago. I divided the games into two-week periods. For each period, I give you the top ten.

Enjoy!

(Check out MarineKing in period 55.)

+ Show Spoiler [Period 1: 2010-02-25 to 2010-03-10] +
0 returning and 27 new players played 32 matches.

1. DIMAGA 0.5559
2. Orly 0.3365
3. LaLuSh 0.3058
4. KaaZ 0.3039
5. MorroW 0.2976
6. NightEnD 0.2976
7. Fantom 0.1691
8. Kirtar 0.1080
9. BeNSeN 0.1058
10. White-Ra 0.0000

+ Show Spoiler [Period 2: 2010-03-11 to 2010-03-24] +
11 returning and 16 new players played 32 matches.

1. LucifroN 0.7604
2. DeMusliM 0.7023
3. DIMAGA 0.6135
4. Orly 0.4874
5. KaaZ 0.3039
6. MorroW 0.2505
7. LaLuSh 0.2050
8. Nazgul 0.1824
9. ret 0.1775
10. Fantom 0.1691

+ Show Spoiler [Period 3: 2010-03-25 to 2010-04-07] +
23 returning and 28 new players played 56 matches.

1. DeMusliM 0.5993
2. LucifroN 0.5742
3. DIMAGA 0.5470
4. Mardow 0.5356
5. SLiDeR 0.3921
6. HasuObs 0.3633
7. GoOdy 0.3583
8. HayprO 0.3436
9. White-Ra 0.3191
10. TheLittleOne 0.3066

+ Show Spoiler [Period 4: 2010-04-08 to 2010-04-21] +
33 returning and 17 new players played 69 matches.

1. DIMAGA 0.5683
2. Mardow 0.5356
3. DeMusliM 0.5268
4. BratOK 0.4667
5. HasuObs 0.3946
6. SLiDeR 0.3921
7. LucifroN 0.3818
8. TheLittleOne 0.3804
9. ZpuX 0.3681
10. Naniwa 0.3450

+ Show Spoiler [Period 5: 2010-04-22 to 2010-05-05] +
29 returning and 13 new players played 60 matches.

1. LucifroN 0.5184
2. DeMusliM 0.5084
3. Strelok 0.5079
4. ZpuX 0.4936
5. HasuObs 0.4184
6. DIMAGA 0.4156
7. Mardow 0.4057
8. HayprO 0.4010
9. SLiDeR 0.3921
10. Naniwa 0.3376

+ Show Spoiler [Period 6: 2010-05-06 to 2010-05-19] +
24 returning and 14 new players played 46 matches.

1. Mardow 0.6938
2. HuK 0.6842
3. TheLittleOne 0.6489
4. LucifroN 0.5184
5. DeMusliM 0.4956
6. DIMAGA 0.4819
7. BratOK 0.4460
8. Sen 0.4317
9. HayprO 0.4010
10. SLiDeR 0.3921

+ Show Spoiler [Period 7: 2010-05-20 to 2010-06-02] +
28 returning and 22 new players played 55 matches.

1. TheLittleOne 0.8266
2. FruitDealer 0.6109
3. HuK 0.6042
4. LucifroN 0.5184
5. Dayfly 0.5068
6. DeMusliM 0.4956
7. BratOK 0.4941
8. ZpuX 0.4600
9. Gerrard 0.4562
10. Tarson 0.4543

+ Show Spoiler [Period 8: 2010-06-03 to 2010-06-16] +
8 returning and 3 new players played 9 matches.

1. TheLittleOne 0.8266
2. BratOK 0.6571
3. FruitDealer 0.6109
4. LucifroN 0.5184
5. Dayfly 0.5068
6. DeMusliM 0.4956
7. ZpuX 0.4600
8. Gerrard 0.4562
9. Strelok 0.4445
10. HuK 0.4380

+ Show Spoiler [Period 9: 2010-06-17 to 2010-06-30] +
0 returning and 0 new players played 0 matches.

1. TheLittleOne 0.8266
2. BratOK 0.6571
3. FruitDealer 0.6109
4. LucifroN 0.5184
5. Dayfly 0.5068
6. DeMusliM 0.4956
7. ZpuX 0.4600
8. Gerrard 0.4562
9. Strelok 0.4445
10. HuK 0.4380

+ Show Spoiler [Period 10: 2010-07-01 to 2010-07-14] +
15 returning and 13 new players played 36 matches.

1. TheLittleOne 0.8654
2. BratOK 0.6571
3. FruitDealer 0.6109
4. LucifroN 0.5184
5. Dayfly 0.5068
6. DeMusliM 0.4956
7. MorroW 0.4946
8. HuK 0.4938
9. ZpuX 0.4600
10. Gerrard 0.4562

+ Show Spoiler [Period 11: 2010-07-15 to 2010-07-28] +
31 returning and 16 new players played 57 matches.

1. BratOK 0.6571
2. teSteR 0.6270
3. FruitDealer 0.6109
4. IdrA 0.5941
5. TheLittleOne 0.5634
6. LucifroN 0.5076
7. Dayfly 0.5068
8. DeMusliM 0.5014
9. MorroW 0.4946
10. Loner 0.4736

+ Show Spoiler [Period 12: 2010-07-29 to 2010-08-11] +
58 returning and 51 new players played 140 matches.

1. SjoW 0.7457
2. TheLittleOne 0.6499
3. teSteR 0.6270
4. FruitDealer 0.6109
5. IdrA 0.5941
6. BratOK 0.5833
7. Strelok 0.5581
8. LucifroN 0.5076
9. Dayfly 0.5068
10. Ensnare 0.4886

+ Show Spoiler [Period 13: 2010-08-12 to 2010-08-25] +
66 returning and 20 new players played 143 matches.

1. MorroW 0.8472
2. SjoW 0.7944
3. TheLittleOne 0.6924
4. Socke 0.6519
5. teSteR 0.6270
6. FruitDealer 0.6109
7. Mana 0.5667
8. HasuObs 0.5621
9. DeMusliM 0.5539
10. drewbie 0.5424

+ Show Spoiler [Period 14: 2010-08-26 to 2010-09-08] +
69 returning and 60 new players played 203 matches.

1. IdrA 0.8608
2. Socke 0.8443
3. MorroW 0.8217
4. SeleCT 0.7708
5. BratOK 0.7434
6. TheLittleOne 0.6924
7. Naniwa 0.6899
8. FruitDealer 0.6833
9. DeMusliM 0.6576
10. ThorZaIN 0.6322

+ Show Spoiler [Period 15: 2010-09-09 to 2010-09-22] +
97 returning and 45 new players played 203 matches.

1. DeMusliM 0.9571
2. MorroW 0.8489
3. KiWiKaKi 0.7142
4. Socke 0.7055
5. BratOK 0.6806
6. teSteR 0.6608
7. Mana 0.6556
8. HuK 0.6328
9. Strelok 0.6236
10. GoOdy 0.6151

+ Show Spoiler [Period 16: 2010-09-23 to 2010-10-06] +
95 returning and 30 new players played 191 matches.

1. DeMusliM 1.0380
2. Naniwa 0.8300
3. FruitDealer 0.7881
4. SjoW 0.7839
5. BratOK 0.6806
6. GoOdy 0.6728
7. PredY 0.6689
8. HasuObs 0.6612
9. Naama 0.6375
10. MorroW 0.6273

+ Show Spoiler [Period 17: 2010-10-07 to 2010-10-20] +
120 returning and 44 new players played 214 matches.

1. Socke 1.0831
2. SjoW 0.9840
3. DeMusliM 0.9668
4. BratOK 0.9037
5. FruitDealer 0.8504
6. Naama 0.7989
7. TOP 0.7321
8. IdrA 0.7256
9. PredY 0.6689
10. ajtls 0.6638

+ Show Spoiler [Period 18: 2010-10-21 to 2010-11-03] +
126 returning and 41 new players played 232 matches.

1. Socke 0.9563
2. BratOK 0.9339
3. SjoW 0.8846
4. FruitDealer 0.8156
5. TOP 0.7321
6. MarineKing 0.7189
7. SLiDeR 0.6822
8. Naama 0.6717
9. SeleCT 0.6706
10. Kas 0.6649

+ Show Spoiler [Period 19: 2010-11-04 to 2010-11-17] +
136 returning and 20 new players played 260 matches.

1. SjoW 1.0511
2. BratOK 0.9236
3. NesTea 0.8494
4. FruitDealer 0.8156
5. MarineKing 0.8035
6. TOP 0.7321
7. elfi 0.7061
8. White-Ra 0.6777
9. Socke 0.6655
10. ajtls 0.6638

+ Show Spoiler [Period 20: 2010-11-18 to 2010-12-01] +
148 returning and 41 new players played 333 matches.

1. NesTea 0.8392
2. eNvious 0.7852
3. FruitDealer 0.7369
4. MarineKing 0.7279
5. SjoW 0.7155
6. LucifroN 0.7127
7. DeMusliM 0.6989
8. Satiini 0.6752
9. ajtls 0.6638
10. PredY 0.6266

+ Show Spoiler [Period 21: 2010-12-02 to 2010-12-15] +
138 returning and 25 new players played 265 matches.

1. Naniwa 0.8838
2. Strelok 0.8226
3. SjoW 0.8131
4. Rain 0.7738
5. NesTea 0.7421
6. PainUser 0.7303
7. Socke 0.7010
8. BratOK 0.6852
9. DIMAGA 0.6733
10. ajtls 0.6638

+ Show Spoiler [Period 22: 2010-12-16 to 2010-12-29] +
104 returning and 16 new players played 225 matches.

1. MC 0.9451
2. Kas 0.8938
3. Naniwa 0.8603
4. SjoW 0.7436
5. Rain 0.7123
6. Strelok 0.6925
7. Mana 0.6889
8. ajtls 0.6638
9. Tarson 0.6544
10. Tefel 0.6257

+ Show Spoiler [Period 23: 2010-12-30 to 2011-01-12] +
190 returning and 32 new players played 431 matches.

1. Naniwa 0.9703
2. MC 0.7946
3. Mvp 0.7816
4. Kas 0.7722
5. TOP 0.7409
6. Strelok 0.7262
7. BratOK 0.7118
8. TheLittleOne 0.6908
9. Rain 0.6801
10. Socke 0.6783

+ Show Spoiler [Period 24: 2011-01-13 to 2011-01-26] +
177 returning and 33 new players played 468 matches.

1. White-Ra 1.0845
2. Naniwa 0.9549
3. Mvp 0.9151
4. MarineKing 0.8519
5. DeMusliM 0.8052
6. SjoW 0.7864
7. MC 0.7663
8. TOP 0.7491
9. IdrA 0.7367
10. Mana 0.7273

+ Show Spoiler [Period 25: 2011-01-27 to 2011-02-09] +
146 returning and 22 new players played 349 matches.

1. Naniwa 1.0609
2. Kas 0.9754
3. Mvp 0.9473
4. TOP 0.9312
5. DIMAGA 0.8848
6. SjoW 0.8586
7. DeMusliM 0.8052
8. MarineKing 0.7923
9. MC 0.7663
10. TheLittleOne 0.7557

+ Show Spoiler [Period 26: 2011-02-10 to 2011-02-23] +
187 returning and 45 new players played 487 matches.

1. Kas 0.9685
2. TOP 0.9312
3. Mvp 0.8536
4. Naniwa 0.8428
5. Bomber 0.8403
6. ret 0.8337
7. HerO 0.8182
8. DeMusliM 0.8052
9. MarineKing 0.7923
10. Nerchio 0.7478

+ Show Spoiler [Period 27: 2011-02-24 to 2011-03-09] +
168 returning and 21 new players played 286 matches.

1. HerO 1.0805
2. Kas 0.9765
3. HasuObs 0.9626
4. Mvp 0.8536
5. MC 0.8449
6. Naniwa 0.8443
7. Bomber 0.8403
8. Axslav 0.8305
9. DeMusliM 0.8052
10. BratOK 0.8044

+ Show Spoiler [Period 28: 2011-03-10 to 2011-03-23] +
164 returning and 19 new players played 349 matches.

1. Kas 1.2662
2. HerO 1.0805
3. MC 0.9324
4. Axslav 0.8305
5. BratOK 0.8206
6. Socke 0.8179
7. DeMusliM 0.8052
8. Naniwa 0.7816
9. GoOdy 0.7612
10. Bomber 0.7520

+ Show Spoiler [Period 29: 2011-03-24 to 2011-04-06] +
185 returning and 14 new players played 444 matches.

1. Naniwa 1.0591
2. HerO 1.0579
3. Bomber 0.9378
4. HasuObs 0.9363
5. MC 0.8567
6. BratOK 0.8479
7. Ryung 0.8213
8. DeMusliM 0.8052
9. KiWiKaKi 0.7934
10. SuperNoVa 0.7894

+ Show Spoiler [Period 30: 2011-04-07 to 2011-04-20] +
174 returning and 18 new players played 318 matches.

1. Naniwa 1.1113
2. Nerchio 1.1052
3. HerO 1.0579
4. Kas 1.0391
5. Stephano 0.9426
6. BratOK 0.9008
7. Bomber 0.8965
8. Mvp 0.8395
9. MC 0.8256
10. DeMusliM 0.8052

+ Show Spoiler [Period 31: 2011-04-21 to 2011-05-04] +
213 returning and 18 new players played 473 matches.

1. Nerchio 1.1544
2. Naniwa 1.1244
3. IdrA 1.0464
4. Strelok 1.0446
5. Bomber 0.9807
6. HerO 0.9534
7. MC 0.9138
8. sC 0.8661
9. HasuObs 0.8655
10. MarineKing 0.8264

+ Show Spoiler [Period 32: 2011-05-05 to 2011-05-18] +
192 returning and 17 new players played 436 matches.

1. Strelok 1.1646
2. MC 1.0865
3. Bomber 1.0310
4. IdrA 0.9795
5. ThorZaIN 0.9581
6. HerO 0.9223
7. Naniwa 0.9165
8. Attero 0.8926
9. Nerchio 0.8834
10. DongRaeGu 0.8767

+ Show Spoiler [Period 33: 2011-05-19 to 2011-06-01] +
223 returning and 19 new players played 374 matches.

1. Nerchio 1.0083
2. Strelok 1.0075
3. MC 0.9748
4. HerO 0.9223
5. Kas 0.9184
6. DongRaeGu 0.9159
7. KiWiKaKi 0.9019
8. Attero 0.8926
9. Bomber 0.8877
10. ThorZaIN 0.8759

+ Show Spoiler [Period 34: 2011-06-02 to 2011-06-15] +
206 returning and 17 new players played 488 matches.

1. MC 1.1227
2. MMA 1.0752
3. HerO 0.9223
4. BratOK 0.9171
5. DongRaeGu 0.9123
6. IdrA 0.9015
7. Attero 0.8926
8. Bomber 0.8877
9. Nerchio 0.8613
10. ThorZaIN 0.8495

+ Show Spoiler [Period 35: 2011-06-16 to 2011-06-29] +
202 returning and 14 new players played 442 matches.

1. MC 1.2042
2. HuK 1.1393
3. Bomber 1.1263
4. Nerchio 1.0841
5. DongRaeGu 1.0219
6. Naniwa 0.9320
7. BratOK 0.9134
8. Happy 0.9016
9. Attero 0.8926
10. HerO 0.8856

+ Show Spoiler [Period 36: 2011-06-30 to 2011-07-13] +
194 returning and 14 new players played 383 matches.

1. Kas 1.2485
2. MC 1.2167
3. PuMa 1.1829
4. Naniwa 1.1804
5. Bomber 1.1778
6. MarineKing 1.1406
7. HuK 1.0925
8. DongRaeGu 0.9761
9. GoOdy 0.9143
10. Sen 0.9016

+ Show Spoiler [Period 37: 2011-07-14 to 2011-07-27] +
214 returning and 15 new players played 441 matches.

1. PuMa 1.1829
2. Naniwa 1.1804
3. Stephano 1.1385
4. HuK 1.1204
5. Bomber 0.9963
6. DongRaeGu 0.9761
7. Nerchio 0.9671
8. MC 0.9406
9. MarineKing 0.9217
10. Sen 0.9016

+ Show Spoiler [Period 38: 2011-07-28 to 2011-08-10] +
243 returning and 21 new players played 568 matches.

1. PuMa 1.1829
2. MMA 1.1453
3. Mana 1.1006
4. Mvp 1.0862
5. DongRaeGu 1.0802
6. Stephano 1.0690
7. sC 1.0405
8. ret 1.0359
9. GanZi 1.0206
10. NesTea 0.9885

+ Show Spoiler [Period 39: 2011-08-11 to 2011-08-24] +
189 returning and 27 new players played 357 matches.

1. PuMa 1.4502
2. Mvp 1.1352
3. GanZi 1.0631
4. Nerchio 1.0574
5. DongRaeGu 1.0130
6. Stephano 0.9917
7. MC 0.9749
8. Mana 0.9571
9. ret 0.9486
10. NesTea 0.9219

+ Show Spoiler [Period 40: 2011-08-25 to 2011-09-07] +
195 returning and 17 new players played 392 matches.

1. Nerchio 1.1760
2. Mvp 1.1619
3. PuMa 1.1048
4. Bomber 0.9994
5. GanZi 0.9617
6. DIMAGA 0.9389
7. NesTea 0.9219
8. Stephano 0.8903
9. CoCa 0.8784
10. MC 0.8752

+ Show Spoiler [Period 41: 2011-09-08 to 2011-09-21] +
218 returning and 36 new players played 402 matches.

1. Mvp 1.2210
2. MMA 1.0820
3. Mana 1.0138
4. Bomber 1.0104
5. Nerchio 1.0006
6. GanZi 0.9617
7. MarineKing 0.9496
8. NesTea 0.9484
9. sC 0.9340
10. PuMa 0.9290

+ Show Spoiler [Period 42: 2011-09-22 to 2011-10-05] +
230 returning and 19 new players played 471 matches.

1. Nerchio 1.3387
2. MarineKing 1.1711
3. Stephano 1.1494
4. Mvp 1.1072
5. Mana 1.0559
6. MMA 1.0174
7. DongRaeGu 0.9918
8. GanZi 0.9855
9. Bomber 0.9577
10. CoCa 0.9193

+ Show Spoiler [Period 43: 2011-10-06 to 2011-10-19] +
265 returning and 20 new players played 592 matches.

1. Mvp 1.1995
2. Stephano 1.0716
3. TheStC 1.0584
4. MC 1.0530
5. BratOK 0.9986
6. Nerchio 0.9853
7. NaDa 0.9804
8. sC 0.9726
9. Polt 0.9468
10. DongRaeGu 0.9401

+ Show Spoiler [Period 44: 2011-10-20 to 2011-11-02] +
209 returning and 22 new players played 526 matches.

1. Stephano 1.3835
2. Kas 1.1286
3. MarineKing 1.1195
4. Mvp 1.1008
5. TheStC 1.0584
6. Nerchio 1.0119
7. sC 1.0079
8. MMA 0.9981
9. NesTea 0.9909
10. NaDa 0.9804

+ Show Spoiler [Period 45: 2011-11-03 to 2011-11-16] +
240 returning and 18 new players played 478 matches.

1. Nerchio 1.3305
2. Stephano 1.3228
3. MarineKing 1.1971
4. Mvp 1.1452
5. IdrA 1.1277
6. MMA 1.0322
7. PuMa 0.9900
8. TheStC 0.9598
9. BratOK 0.9407
10. sC 0.9381

+ Show Spoiler [Period 46: 2011-11-17 to 2011-11-30] +
293 returning and 15 new players played 651 matches.

1. Leenock 1.4653
2. Mvp 1.2281
3. HerO 1.1891
4. MarineKing 1.1871
5. PuMa 1.1568
6. Nerchio 1.1400
7. DongRaeGu 1.1108
8. Tails 1.0088
9. Naniwa 0.9917
10. Curious 0.9887

+ Show Spoiler [Period 47: 2011-12-01 to 2011-12-14] +
232 returning and 25 new players played 567 matches.

1. Mvp 1.3784
2. PuMa 1.3263
3. MC 1.1381
4. DongRaeGu 1.1212
5. HerO 1.1092
6. Kas 1.0753
7. MarineKing 1.0448
8. GanZi 1.0057
9. Jjakji 0.9973
10. MMA 0.9841

+ Show Spoiler [Period 48: 2011-12-15 to 2011-12-28] +
210 returning and 10 new players played 372 matches.

1. Mvp 1.5414
2. PuMa 1.3263
3. MMA 1.2740
4. MC 1.1929
5. Kas 1.1746
6. Polt 1.1572
7. DongRaeGu 1.1455
8. MarineKing 1.1391
9. GanZi 1.0443
10. TheStC 1.0364

+ Show Spoiler [Period 49: 2011-12-29 to 2012-01-11] +
128 returning and 6 new players played 212 matches.

1. DongRaeGu 1.3762
2. Mvp 1.3358
3. PuMa 1.3263
4. MMA 1.2740
5. MC 1.1882
6. MarineKing 1.1769
7. Kas 1.1514
8. GanZi 1.0712
9. Mana 1.0682
10. Polt 1.0644

+ Show Spoiler [Period 50: 2012-01-12 to 2012-01-25] +
184 returning and 7 new players played 318 matches.

1. Mvp 1.3358
2. MarineKing 1.3014
3. DongRaeGu 1.2956
4. MMA 1.2771
5. PuMa 1.2198
6. Stephano 1.2170
7. MC 1.1882
8. NesTea 1.0335
9. JYP 1.0184
10. DIMAGA 1.0102

+ Show Spoiler [Period 51: 2012-01-26 to 2012-02-08] +
255 returning and 13 new players played 490 matches.

1. Polt 1.3159
2. DongRaeGu 1.3077
3. MarineKing 1.2709
4. MMA 1.2376
5. Mvp 1.1787
6. MC 1.1765
7. PuMa 1.1326
8. Stephano 1.1015
9. Life 1.0359
10. NesTea 1.0335

+ Show Spoiler [Period 52: 2012-02-09 to 2012-02-22] +
229 returning and 12 new players played 409 matches.

1. DongRaeGu 1.5639
2. MarineKing 1.3975
3. MMA 1.2504
4. Stephano 1.2163
5. Mvp 1.1787
6. HerO 1.1574
7. Genius 1.0559
8. PuMa 1.0547
9. MC 1.0385
10. NesTea 0.9706

+ Show Spoiler [Period 53: 2012-02-23 to 2012-03-07] +
244 returning and 14 new players played 585 matches.

1. MarineKing 1.5544
2. PuMa 1.3298
3. DongRaeGu 1.3150
4. MMA 1.3107
5. Polt 1.1975
6. HerO 1.1635
7. Parting 1.1338
8. Genius 1.1252
9. viOLet 1.1142
10. Nerchio 1.0784

+ Show Spoiler [Period 54: 2012-03-08 to 2012-03-21] +
206 returning and 8 new players played 433 matches.

1. MarineKing 1.3459
2. Stephano 1.3288
3. PuMa 1.3249
4. MC 1.3137
5. HerO 1.3099
6. DongRaeGu 1.2913
7. MMA 1.2862
8. Genius 1.1252
9. Parting 1.0894
10. Leenock 1.0388

+ Show Spoiler [Period 55: 2012-03-22 to 2012-04-04] +
196 returning and 7 new players played 404 matches.

1. MarineKing 1.7145
2. MMA 1.2951
3. HerO 1.2039
4. Jjakji 1.1543
5. DongRaeGu 1.1383
6. GanZi 1.1060
7. Leenock 1.0703
8. PuMa 1.0584
9. TaeJa 1.0569
10. Stephano 1.0501

+ Show Spoiler [Period 56: 2012-04-05 to 2012-04-18] +
165 returning and 36 new players played 359 matches.

1. aLive 1.2795
2. MarineKing 1.2357
3. Squirtle 1.2305
4. Stephano 1.1793
5. DongRaeGu 1.1685
6. NesTea 1.1560
7. HerO 1.1059
8. TaeJa 1.0569
9. Life 1.0421
10. Naniwa 1.0374

+ Show Spoiler [Period 57: 2012-04-19 to 2012-05-02] +
183 returning and 18 new players played 389 matches.

1. aLive 1.4443
2. Stephano 1.4057
3. ThorZaIN 1.2937
4. Squirtle 1.2849
5. HerO 1.1872
6. MarineKing 1.1704
7. NesTea 1.1560
8. TaeJa 1.1112
9. DongRaeGu 1.1100
10. Life 1.0634

+ Show Spoiler [Period 58: 2012-05-03 to 2012-05-16] +
177 returning and 30 new players played 291 matches.

1. Squirtle 1.4195
2. Stephano 1.3406
3. MMA 1.2058
4. MarineKing 1.2052
5. HerO 1.1616
6. Parting 1.1026
7. DongRaeGu 1.0794
8. Mvp 1.0710
9. TaeJa 1.0631
10. LucifroN 1.0586

+ Show Spoiler [Period 59: 2012-05-17 to 2012-05-30] +
193 returning and 23 new players played 380 matches.

1. Nerchio 1.2610
2. Stephano 1.2577
3. viOLet 1.2262
4. Mvp 1.1592
5. Squirtle 1.1508
6. Symbol 1.1399
7. HyuN 1.1336
8. HerO 1.1210
9. Life 1.1094
10. Puzzle 1.1029

+ Show Spoiler [Period 60: 2012-05-31 to 2012-06-13] +
226 returning and 33 new players played 528 matches.

1. Nerchio 1.3573
2. DongRaeGu 1.3440
3. MarineKing 1.3215
4. Alicia 1.2926
5. Monster 1.2822
6. Symbol 1.2700
7. Stephano 1.1455
8. Mvp 1.1355
9. PuMa 1.1162
10. Squirtle 1.0972

+ Show Spoiler [Period 61: 2012-06-14 to 2012-06-27] +
251 returning and 79 new players played 745 matches.

1. Stephano 1.3708
2. HerO 1.2916
3. DongRaeGu 1.2722
4. Symbol 1.2614
5. Monster 1.2039
6. Nerchio 1.1701
7. MC 1.1615
8. SaSe 1.1352
9. MarineKing 1.1268
10. Alicia 1.1130

+ Show Spoiler [Period 62: 2012-06-28 to 2012-07-11] +
238 returning and 35 new players played 501 matches.

1. MC 1.3733
2. Nerchio 1.3103
3. DongRaeGu 1.2846
4. Stephano 1.2388
5. HerO 1.2362
6. Symbol 1.1928
7. Alicia 1.1816
8. HyuN 1.1492
9. Polt 1.1347
10. aLive 1.0971

+ Show Spoiler [Period 63: 2012-07-12 to 2012-07-25] +
250 returning and 27 new players played 489 matches.

1. MC 1.4168
2. Stephano 1.3301
3. HyuN 1.3261
4. Nerchio 1.2969
5. Oz 1.2826
6. TaeJa 1.2544
7. MarineKing 1.2122
8. DongRaeGu 1.1942
9. TITAN 1.1478
10. Symbol 1.1145

+ Show Spoiler [Period 64: 2012-07-26 to 2012-08-08] +
120 returning and 14 new players played 214 matches.

1. Stephano 1.3301
2. TaeJa 1.2857
3. Oz 1.2826
4. MarineKing 1.2122
5. DongRaeGu 1.1787
6. Creator 1.1673
7. Seed 1.1562
8. MC 1.1478
9. HyuN 1.1181
10. Nerchio 1.0838

+ Show Spoiler [Period 65: 2012-08-09 to 2012-08-22] +
120 returning and 11 new players played 246 matches.

1. Stephano 1.3301
2. Creator 1.3254
3. TaeJa 1.2799
4. fOrGG 1.2353
5. MarineKing 1.2122
6. Seed 1.1876
7. Mvp 1.1825
8. SuperNoVa 1.1642
9. Nerchio 1.1554
10. DongRaeGu 1.1459

+ Show Spoiler [Period 66: 2012-08-23 to 2012-09-05] +
177 returning and 28 new players played 486 matches.

1. Leenock 1.3762
2. Nerchio 1.3739
3. SuperNoVa 1.2782
4. HerO 1.2384
5. Scarlett 1.1947
6. First 1.1919
7. HyuN 1.1500
8. DongRaeGu 1.1379
9. Creator 1.1354
10. TaeJa 1.1307

+ Show Spoiler [Period 67: 2012-09-06 to 2012-09-19] +
158 returning and 13 new players played 279 matches.

1. Nerchio 1.4710
2. Leenock 1.3762
3. fOrGG 1.2895
4. HerO 1.2384
5. TaeJa 1.2047
6. Creator 1.1972
7. Scarlett 1.1947
8. First 1.1601
9. Life 1.1601
10. viOLet 1.1536

+ Show Spoiler [Period 68: 2012-09-20 to 2012-10-03] +
175 returning and 29 new players played 422 matches.

1. TaeJa 1.4342
2. Stephano 1.4067
3. By.Rain 1.3601
4. Leenock 1.3011
5. Scarlett 1.2595
6. HerO 1.2162
7. Nerchio 1.1960
8. Life 1.1924
9. Yonghwa 1.1791
10. First 1.1601

+ Show Spoiler [Period 69: 2012-10-04 to 2012-10-17] +
241 returning and 21 new players played 735 matches.

1. Scarlett 1.5251
2. Stephano 1.4381
3. By.Rain 1.4298
4. Life 1.3893
5. BaBy 1.3777
6. Soulkey 1.3294
7. Leenock 1.3277
8. RorO 1.2395
9. Last 1.2126
10. Creator 1.2041

+ Show Spoiler [Period 70: 2012-10-18 to 2012-10-31] +
206 returning and 29 new players played 494 matches.

1. Stephano 1.5296
2. Life 1.5058
3. Scarlett 1.4848
4. Leenock 1.3676
5. By.Rain 1.3431
6. Soulkey 1.2785
7. soO 1.2556
8. BaBy 1.2426
9. Creator 1.2041
10. Parting 1.1919

+ Show Spoiler [Period 71: 2012-11-01 to 2012-11-14] +
256 returning and 32 new players played 516 matches.

1. Life 1.6925
2. Leenock 1.5163
3. Bomber 1.4616
4. By.Rain 1.3186
5. HyuN 1.2846
6. Bogus 1.2403
7. Flash 1.2327
8. fOrGG 1.2302
9. BaBy 1.2107
10. Creator 1.2037

+ Show Spoiler [Period 72: 2012-11-15 to 2012-11-28] +
143 returning and 6 new players played 260 matches.

1. Bomber 1.6862
2. Life 1.5577
3. HyuN 1.4510
4. Leenock 1.4070
5. HerO 1.3790
6. Parting 1.3000
7. Ryung 1.2602
8. BaBy 1.2190
9. Trap 1.1849
10. Soulkey 1.1823


*****
http://aligulac.com || Barcraft Switzerland! || Zerg best race. || Stats-poster extraordinaire.
snively
Profile Blog Joined August 2011
United States1159 Posts
December 02 2012 16:47 GMT
#2
wow thats amazing! 5/5
that must of taken soo muc work

also LOL at marineking.
why is his rating so high?
im trying to remember what happened in march/april 2012
My religion is Starcraft
TheBB
Profile Blog Joined July 2009
Switzerland5133 Posts
Last Edited: 2012-12-02 17:10:04
December 02 2012 17:08 GMT
#3
On December 03 2012 01:47 snively wrote:
im trying to remember what happened in march/april 2012

I think MarineKing won an MLG in that time period? Apparently he went 24-6 against players like MC, Ganzi, Sase, Thorzain, Parting, DRG, Zenio, July and Leenock.
http://aligulac.com || Barcraft Switzerland! || Zerg best race. || Stats-poster extraordinaire.
docvoc
Profile Blog Joined July 2011
United States5491 Posts
December 02 2012 18:07 GMT
#4
Jeez O.o this must have taken forever dude.
User was warned for too many mimes.
jrkirby
Profile Blog Joined August 2010
United States1510 Posts
December 02 2012 19:38 GMT
#5
On December 03 2012 03:07 docvoc wrote:
Jeez O.o this must have taken forever dude.


While I'm sure he put tons of work into this, I'm guessing most of it was in programming, and he probably wrote a short piece of code to format the data into all those spoilers. He probably didn't copy paste 70 times, just once.
dzovan
Profile Joined May 2011
Serbia37 Posts
December 02 2012 20:38 GMT
#6
Great job! I wish I had time to read through the paper and your implementation. I'm really interested in the dynamical model in the paper. I liked the example of the strongest chess players in history.
dzovan
Profile Joined May 2011
Serbia37 Posts
December 02 2012 20:48 GMT
#7
I have recently been studying Markov chains and its implementations. My understanding is quite basic but I think it could be used to examine the sc2 pro scene, it would fun to try at least... It is different from your work but it would be fun to play with the probabilities, but it would be hard to sort through the data ...
opterown *
Profile Blog Joined August 2011
Australia54784 Posts
December 02 2012 21:11 GMT
#8
wow, nice job man!
ModeratorRetired LR Bonjwa
TL+ Member
Luepert
Profile Blog Joined June 2011
United States1933 Posts
December 02 2012 21:24 GMT
#9
Neither Kas nor Strelok ever deserved to be number 1, ever.
esports
Entirety
Profile Blog Joined April 2012
1423 Posts
December 02 2012 22:05 GMT
#10
Perhaps this is due to the system not adequately differentiating between the difficulty level of international tournaments vs. GSL, but I feel like foreigners are ranked too highly whereas GSL is ranked too low.

It makes sense though, since the first GSL season was in August 2010. Before that, the international scene produced a few results. Thus, the people who did well before August 2010 were catapulted to the position of "World's Best" by default... meaning that another international tournament where some dude beat up the current "World's Best" immediately skyrocketed up the rankings. On the other hand, FruitDealer, who beat up a bunch of "no-namer" Koreans, did not go very high up in the rankings.

I know that this seems highly subjective, the opposite of hard data/statistics which you want, but I think you should add a lot more weight to the first GSL and GSLs after that. Maybe you have to subjectively designate the first GSL participants as highly ranked. I don't know the best solution to this.

Here's my idea: Extend your program which predicts tournament results and apply that program before every tournament. (For example, in August 2010, use the rankings from right before the tournament begins, plug it into your program, and calculate the expected winners of the tournament.) Then, compare your predicted results with the actual results (perhaps use something like a Chi-Squared Test) and find your error. Then sum up all the error for every tournament until now.

Then, you can adjust your parameters and methods to minimize your error. Hopefully that can help improve your ranking system?
IMMvp (정종현) | Fan Club: http://www.teamliquid.net/forum/viewmessage.php?topic_id=211431
ThunderGod
Profile Blog Joined February 2009
New Zealand897 Posts
December 03 2012 00:36 GMT
#11
I support this 5/5
"Certain forms of popular music nowadays, namely rap and hip hop styles, are just irritating gangsters bragging about their illegal exploits and short-sighted lifestyles." - Shiverfish ~2009
TheBB
Profile Blog Joined July 2009
Switzerland5133 Posts
Last Edited: 2012-12-03 22:50:05
December 03 2012 22:39 GMT
#12
On December 03 2012 07:05 Entirety wrote:
Then, you can adjust your parameters and methods to minimize your error. Hopefully that can help improve your ranking system?

I was planning to do this, but not quite so complicated as that. Basically what I wanted to do was to go through all the matches in my database and use the ratings at the time the match was played to predict the result. Then I can group all the games into batches, i.e. those which were favoured between 50-55%, 55-60% and so on. For each group I can then calculate what the actual winrate was. When plotted, these should ideally make a straight diagonal line so that predicted winrate = actual winrate.

So... drumroll:

[image loading]


Not half bad if I say so myself! The black line shows the data, the red line shows the ideal relationship that we want to achieve, and the blue line shows a linear fit for the data. The plot is based on almost 50,000 games worth of data.

This shows that my system (at the moment) somewhat underestimates the chances of the stronger player. If anything, I was afraid it would do the opposite, and now I'm very confident that this system can be really good.

Next step is to tweak the parameters so that the lines coincide as much as possible.

Does this make sense or am I fooling myself?
http://aligulac.com || Barcraft Switzerland! || Zerg best race. || Stats-poster extraordinaire.
opterown *
Profile Blog Joined August 2011
Australia54784 Posts
December 04 2012 11:35 GMT
#13
On December 04 2012 07:39 TheBB wrote:
Does this make sense or am I fooling myself?

makes sense, i look forward to seeing the results!~
ModeratorRetired LR Bonjwa
TL+ Member
heartlxp
Profile Joined September 2010
United States1258 Posts
December 10 2012 18:24 GMT
#14
well done!
number01
Profile Joined December 2012
203 Posts
December 10 2012 18:46 GMT
#15
All i can say is thanks man this is tight!
Idra is the reason I play SC
KillerDucky
Profile Blog Joined July 2010
United States498 Posts
Last Edited: 2012-12-12 22:35:43
December 12 2012 22:35 GMT
#16
On December 04 2012 07:39 TheBB wrote:
Show nested quote +
On December 03 2012 07:05 Entirety wrote:
Then, you can adjust your parameters and methods to minimize your error. Hopefully that can help improve your ranking system?

I was planning to do this, but not quite so complicated as that. Basically what I wanted to do was to go through all the matches in my database and use the ratings at the time the match was played to predict the result. Then I can group all the games into batches, i.e. those which were favoured between 50-55%, 55-60% and so on. For each group I can then calculate what the actual winrate was. When plotted, these should ideally make a straight diagonal line so that predicted winrate = actual winrate.

Does this make sense or am I fooling myself?



When you say "ratings at the time the match was played", do you mean ratings after that period or ratings from the previous period? I think it should be from the previous period right? The papers I've read about rating systems they separate the data set into two parts. One to use as input, and one to measure prediction accuracy. If you use the same data for both input and measuring prediction accuracy don't you get a sort of self-fulfilling prophecy?

MarineKingPrime Forever!
TheBB
Profile Blog Joined July 2009
Switzerland5133 Posts
December 14 2012 14:46 GMT
#17
On December 13 2012 07:35 KillerDucky wrote:
Show nested quote +
On December 04 2012 07:39 TheBB wrote:
On December 03 2012 07:05 Entirety wrote:
Then, you can adjust your parameters and methods to minimize your error. Hopefully that can help improve your ranking system?

I was planning to do this, but not quite so complicated as that. Basically what I wanted to do was to go through all the matches in my database and use the ratings at the time the match was played to predict the result. Then I can group all the games into batches, i.e. those which were favoured between 50-55%, 55-60% and so on. For each group I can then calculate what the actual winrate was. When plotted, these should ideally make a straight diagonal line so that predicted winrate = actual winrate.

Does this make sense or am I fooling myself?



When you say "ratings at the time the match was played", do you mean ratings after that period or ratings from the previous period? I think it should be from the previous period right? The papers I've read about rating systems they separate the data set into two parts. One to use as input, and one to measure prediction accuracy. If you use the same data for both input and measuring prediction accuracy don't you get a sort of self-fulfilling prophecy?

I didn't consider that, but I don't think it applies, does it? I use the ratings from the previous period, which does not depend on the given match in any way.
http://aligulac.com || Barcraft Switzerland! || Zerg best race. || Stats-poster extraordinaire.
KillerDucky
Profile Blog Joined July 2010
United States498 Posts
December 14 2012 18:09 GMT
#18
On December 14 2012 23:46 TheBB wrote:
Show nested quote +
On December 13 2012 07:35 KillerDucky wrote:
On December 04 2012 07:39 TheBB wrote:
On December 03 2012 07:05 Entirety wrote:
Then, you can adjust your parameters and methods to minimize your error. Hopefully that can help improve your ranking system?

I was planning to do this, but not quite so complicated as that. Basically what I wanted to do was to go through all the matches in my database and use the ratings at the time the match was played to predict the result. Then I can group all the games into batches, i.e. those which were favoured between 50-55%, 55-60% and so on. For each group I can then calculate what the actual winrate was. When plotted, these should ideally make a straight diagonal line so that predicted winrate = actual winrate.

Does this make sense or am I fooling myself?



When you say "ratings at the time the match was played", do you mean ratings after that period or ratings from the previous period? I think it should be from the previous period right? The papers I've read about rating systems they separate the data set into two parts. One to use as input, and one to measure prediction accuracy. If you use the same data for both input and measuring prediction accuracy don't you get a sort of self-fulfilling prophecy?

I didn't consider that, but I don't think it applies, does it? I use the ratings from the previous period, which does not depend on the given match in any way.


Ah previous period -- that should work then. I just read too much into "ratings at the time the match was played'.
MarineKingPrime Forever!
Please log in or register to reply.
Live Events Refresh
Next event in 6h 25m
[ Submit Event ]
Live Streams
Refresh
StarCraft 2
Nina 297
ProTech64
Ketroc 29
StarCraft: Brood War
Leta 503
Shine 97
Noble 45
Mind 31
ajuk12(nOOB) 13
Bale 4
Dota 2
monkeys_forever1201
Counter-Strike
Stewie2K1312
Super Smash Bros
Mew2King168
Other Games
summit1g8665
ViBE236
Maynarde151
Organizations
Other Games
gamesdonequick39435
StarCraft 2
Blizzard YouTube
StarCraft: Brood War
BSLTrovo
sctven
[ Show 18 non-featured ]
StarCraft 2
• Berry_CruncH296
• Hupsaiya 50
• practicex 27
• Kozan
• Migwel
• AfreecaTV YouTube
• sooper7s
• intothetv
• IndyKCrew
• LaughNgamezSOOP
StarCraft: Brood War
• Azhi_Dahaki37
• Diggity4
• STPLYoutube
• ZZZeroYoutube
• BSLYoutube
Dota 2
• masondota2508
League of Legends
• Lourlo1250
• Rush782
Upcoming Events
Wardi Open
6h 25m
Replay Cast
19h 25m
Sparkling Tuna Cup
1d 5h
WardiTV European League
1d 11h
MaNa vs sebesdes
Mixu vs Fjant
ByuN vs HeRoMaRinE
ShoWTimE vs goblin
Gerald vs Babymarine
Krystianer vs YoungYakov
PiGosaur Monday
1d 19h
The PondCast
2 days
WardiTV European League
2 days
Jumy vs NightPhoenix
Percival vs Nicoract
ArT vs HiGhDrA
MaxPax vs Harstem
Scarlett vs Shameless
SKillous vs uThermal
uThermal 2v2 Circuit
2 days
Replay Cast
2 days
RSL Revival
3 days
ByuN vs SHIN
Clem vs Reynor
[ Show More ]
Replay Cast
3 days
RSL Revival
4 days
Classic vs Cure
FEL
4 days
RSL Revival
5 days
FEL
5 days
FEL
5 days
Sparkling Tuna Cup
6 days
RSL Revival
6 days
FEL
6 days
Liquipedia Results

Completed

BSL Season 20
HSC XXVII
Heroes 10 EU

Ongoing

JPL Season 2
BSL 2v2 Season 3
Acropolis #3
KCM Race Survival 2025 Season 2
CSL 17: 2025 SUMMER
Copa Latinoamericana 4
Jiahua Invitational
Championship of Russia 2025
RSL Revival: Season 1
Murky Cup #2
BLAST.tv Austin Major 2025
ESL Impact League Season 7
IEM Dallas 2025
PGL Astana 2025
Asian Champions League '25
BLAST Rivals Spring 2025
MESA Nomadic Masters
CCT Season 2 Global Finals
IEM Melbourne 2025

Upcoming

2025 ACS Season 2: Qualifier
CSLPRO Last Chance 2025
CSL Xiamen Invitational
2025 ACS Season 2
CSLPRO Chat StarLAN 3
K-Championship
uThermal 2v2 Main Event
SEL Season 2 Championship
FEL Cracov 2025
Esports World Cup 2025
StarSeries Fall 2025
FISSURE Playground #2
BLAST Open Fall 2025
BLAST Open Fall Qual
Esports World Cup 2025
BLAST Bounty Fall 2025
BLAST Bounty Fall Qual
IEM Cologne 2025
FISSURE Playground #1
TLPD

1. ByuN
2. TY
3. Dark
4. Solar
5. Stats
6. Nerchio
7. sOs
8. soO
9. INnoVation
10. Elazer
1. Rain
2. Flash
3. EffOrt
4. Last
5. Bisu
6. Soulkey
7. Mini
8. Sharp
Sidebar Settings...

Advertising | Privacy Policy | Terms Of Use | Contact Us

Original banner artwork: Jim Warren
The contents of this webpage are copyright © 2025 TLnet. All Rights Reserved.