G5

Rank Player name                Points EPoints  StdDev   ERank    1  2  3  4  5  6
   1 Buns157                       0.0    8.55    1.16    1.31   73 24  3  0  0  0
   2 GreenTea                      0.0    6.05    1.54    2.72    5 44 31 13  5  1
   3 fireice82                     0.0    4.47    1.46    3.90    0  9 27 33 23  7
   4 PanagiotisTheGreekFreak       0.0    4.20    1.46    4.12    0  7 21 34 29  9
   5 linberson                     0.0    3.52    1.98    4.50    1  8 13 16 37 24
   6 SexMachineRJ                  0.0    3.20    3.87    4.44   20  8  4  3  6 59

GreenTea is calculated from leMarseillais 1v1 data. (I pick the player based on closest to GreenTea's last rating and preferably higher StdDev.) SexMachineRJ has no 1v1 and is treated as unknown.

No switch. This should be unfair.

If Timinator don't want to play, I cannot force him to play, but I will not change the rules for him. This tourney suppose the winner could play (and win) on very different templates. If a player cannot, just he should not go, he is not the best anyway.

Timinator, you should play. If you don't want, this will penalize you, will penalize germany (indirectly), and will penalize the tournament.


TBest, I'm glad to see a prediction that not put french in last places. Poor AI, between two french, he should be very bad.^^ Do you predict that from ladder perf ?

^ Nice to see a reasonable answer :) ^

Template links

Yin'angzhou
https://www.warlight.net/MultiPlayer?TemplateID=818944

Snow Mountain MA
https://www.warlight.net/MultiPlayer?TournamentID=16520

China
https://www.warlight.net/MultiPlayer?GameID=11512630

Greece
https://www.warlight.net/MultiPlayer?GameID=11510645

Guiroma
https://www.warlight.net/MultiPlayer/Tournament?ID=19191

Strat ME
https://www.warlight.net/MultiPlayer?GameID=9851117
This one but in WR (didn't found the good link. Someone have ?)

Strat MME with commander
https://www.warlight.net/MultiPlayer?GameID=10678718

Turkey LD
https://www.warlight.net/MultiPlayer?GameID=5417740

French Brawl
https://www.warlight.net/MultiPlayer/Tournament?ID=19194

Treasure Map
https://www.warlight.net/MultiPlayer?GameID=11512322

Qina
https://www.warlight.net/MultiPlayer?TemplateID=825274

Battle Island
https://www.warlight.net/MultiPlayer?GameID=11511549

Edited 7/4/2016 14:17:34

Commanders are bullshit from a strategic view.
i completely stopped playing rt ladder because commanders are there. and i will keep boycotting commanders.

So you can decline the commander tourney and play the other one. I don't think you could be qualified for round 2 with that, but maybe take some points for germany. As you want.

I'm not strong with commanders, but I don't think it's not strategic... it's recent, maybe you just don't understand it for now, as a lot of players (and probably me).

5th place is a heavy prediction for me, I would be glad if I could even win one game :P

if spamming every opponent's territory with an attack of 1 without risk sounds strategic to you... well, can't help you then :)

G4

Rank Player name   Points EPoints  StdDev   ERank    1  2  3  4  5  6
   1 Verzehrer        0.0    7.46    1.37    1.75   46 38 12  3  1  0
   2 Beren            0.0    7.09    1.86    2.06   41 32 15  7  3  2
   3 MoD              0.0    4.66    2.37    3.71   11 15 22 16 15 21
   4 Nox              0.0    4.27    1.39    3.97    1  7 27 34 22  9
   5 MisterT          0.0    3.69    1.65    4.42    1  6 17 24 27 24
   6 Summer           0.0    2.83    1.37    5.10    0  1  7 17 31 44

andy903 data' is used for Beren. MarcRoss for MoD.

@ Do you predict that from ladder perf ?

Yes, unexpired games from the 1v1 ladder is the source. (which is why I find a "replacment" player for the once who are not on the ladder, following the principle from last post.)

Edited 7/3/2016 17:16:28

who the hell is Marc ross? ^^ i toght i know almost all good players

haha xD

I have no clue who Marc Ross is. I don't even have his profile link as a matter of fact. Also, MoD, who siad you are a good player ;)

In seriousness through, I used the rating you had when you entered your last played ladder game. In your case that is significantly lower then your peak. (about 1900 compered to 2300) Due to this I picked a player with a high StdDev, (so few games). Point is, the prediction probably undervalues you. (21% chance of last place...) I know it is wrong, but I would rather be consistent with the principle I stated earlier.

To estimate a player with another one ? Furthermore with very suspicious similarities, and with very few and incomplete data... Why you have chosen Marc Ross so TBest ? These predictions have no sense. There is no good tool without common sense even if your simulation tool seems to be interesting. ;)

Timi, you are a very good player, but you seem to be very mulish too. Maybe it's why you're not the best. I organize that and try to do this pleasant to all. I didn't choose templates I like or I'm good. I didn't do that without asking for some advices. I'm not here to debate longer of all or a part of them now (too late) and still less to hear your condescension.

I do not get tired more. Play or do not play, you will be invited to the two tournaments anyway.

Edited 7/3/2016 23:52:36

First match series have been created.

Tournament links:

https://www.warlight.net/MultiPlayer/Tournament?ID=19721
https://www.warlight.net/MultiPlayer/Tournament?ID=19726
https://www.warlight.net/MultiPlayer/Tournament?ID=19725
https://www.warlight.net/MultiPlayer/Tournament?ID=19723
https://www.warlight.net/MultiPlayer/Tournament?ID=19724
https://www.warlight.net/MultiPlayer/Tournament?ID=19727

To estimate a player with another one ? Furthermore with very suspicious similarities, and with very few and incomplete data... Why you have chosen Marc Ross so TBest ? These predictions have no sense. There is no good tool without common sense even if your simulation tool seems to be interesting. ;)

I could have treated MoD as an unknow player. But this makes even less sense /: then MoD's prediction would have looked like this "

6 SexMachineRJ                  0.0    3.20    3.87    4.44   20  8  4  3  6 59"

So that option is no good. I choose MarcRoss becoes he was closest to the Last rating MoD had on the 1v1 ladder. So while inaccurate, better then nothing. But yes, the tool is only good for the players who have unexpired games on the 1v1 ladder. And this is alos the first time I am trying this, so I am curious as to how accurate it will turn out to be. (Besides, accuracy is never there anyway. Skill on one template don't equal skill on another.)

What is the tool you use for your simulations ? I don't think ladder stat is a very good thing to make predictions, so all this seems crap to me, but maybe the tool is ok. I would be curious to see that.

I wrote about it here. https://www.warlight.net/Forum/159812-clan-league-8-division-c-official-thread?Offset=160 , and the next page in that thread.
You will see a few more predictions there. And you also see what happens when you have unknown players... which is why I choose to use an existing player here.

http://www.remi-coulom.fr/Bayesian-Elo/

^the tool that is used for making the ratings on the ladder. This also includes a RR prediction tool. It is that prediction tool that I am using. This means it is very easy to take the ladder data http://data.warlight.net/Data/BayeseloLog0.txt and predict things. After all the Data is already in the right format.

Edited 7/4/2016 22:13:09

TY

Update after one week.




- hedja got two boots already and will have hard time to reach round 2.
- Summer beats Verzehrer who seems not so in shape currently.
- Wazz beats INSIDE
- MIFRAN beats AI
- Panagiotis beats fireice82 and fireice82 beats SexMachineRJ.

No surprise for now. The round 2 stay open for everyone.

Edited 7/11/2016 13:20:52

Update

#G1
MotD 2-0
Krzychu 1-0
KKND 1-0
Sultan 1-1
Tbest 0-0
hedja 0-4

#G2
timon92 1-0
Mirror 0-0
Timinator 0-0
FlyingBender 0-0
Farah 0-0
Rakleader 0-1

#G3
Wazz 1-0
Mike 0-0
USA Bitches 0-0
Smoove 0-0
Ryiro 0-0
INSIDE 0-1

#G4
Mister T 2-1
Nox 2-1
Summer 1-1
Verzehrer 1-1
MoD 0-0
Beren 0-2

#G5
Buns157 2-0
GreenTea 2-1
PanagiotisTheGreekFreak 1-0
fireice82 1-2
SexMachine 0-1
linberson 0-2

#G6
Phaeril 2-0
MIFRAN 2-1
AI 1-1
Nackickwind 0-0
Quicksand 0-1
Peixoto 0-2

Edited 7/15/2016 10:54:46