Warrior General Games

Here is our first set of power rankings for our league. The formula used to figure out these rankings come special thanks to Waltang (USA A #5), the format of the rankings are thanks to Sawedust (Europe East AAA#2).

A brief introduction of how the formula works:

Originally posted by Waltang



The ranking is comprised of 3 numbers.

The first is a culmination of record and Strength of schedule. Its forumula is :

RecordValue = (#ofWins x 1.5)+(#ofWins x WinStrength)+(#ofLosses x LossStrength)

where
WinStrengh = ((Total Wins of all opponents that team has beat)/(Total Games Played of all opponents that team has beat)) | (0 if team has not won)
LossStrengh = ((Total Wins of all opponents that team has Lost to)/(Total Games Played of all opponents that team has Lost to))| (1 if undefeated)

The second Value is for Margin of Victory. Its forumla is :

VictoryValue = (AverageMarginofVictory^1)/35 x (#ofWins/#ofGames)*Weekweight

Weekweight = #ofGames * .3 - This keeps the relative value of of margin of victory relatively consistant as the # of games increases.

DefeatValue = (AverageMarginofDefeat^1)/35 x (#ofLosses/#ofGames)*Weekweight*-1

^1 Margin of Victory and Defeats for any single game is limited to 35 points. Blowouts by more than that are as much morale based as actual skill based, so limited to this for powerrankign purposes.

Ranking = RecordValue + VictoryValue + DefeatValue




Early on there is no clear cut "stronger" conference. Out of the top 10 there is a 5 and 5 split between Alpha and Beta. This is likely to change as the season goes on, but for now we are pretty even.


WEEK 4 POWER RANKING


Top 10

1. South Yorkshire Stars (4-0) 8.439
2. Amsterdam Smugglers (4-0) 8.141
3. Belfast Mafia (4-0) 7.85
4. Quebec Colored Trombones (4-0) 6.76
5. Konoha Hokages (4-0) 6.663
6. Antwerp Jewish Diamond Dealers (3-1) 5.194
7. Transylvania Bloodsuckers (3-1) 5.038
8. Hamburg Parrots (3-1) 4.97
9. Amsterdam Artic Blast (3-1) 4.954
10. St. Louis Storm (3-1) 4.843


Contending For A Top Spot

11. Grim Reapers (3-1) 4.637
12. Cardiff Cobras (3-1) 4.56
13. Newcastle Nightmare (3-1) 4.403
14. Dublin Drunken Monkeys (3-1) 4.348
15. Crimson and Cream Sooner Machine (0-4) 3.586
16. Warsaw Knuckleheads (2-2) 3.549


Needs Improvement

17. Vienna Blood Sausages (2-2) 3.394
18. Wrexham Maelor Brionnfhionn (2-2) 3.377
19. Basin City Blues (2-2) 3.295
20. Athens Gators (2-2) 3.261
21. Vatican City Angels (1-3) 2.89
22. Oklahoma (USA) Outlaws (1-3) 2.861


Divine Intervention

23. Dresden Bombers (0-4) 2.703
24. Valencia Bats (1-3) 2.691
25. Istanbul Three Headed Pygmy Dragon (1-3) 2.592
26. Ithaca Space Munkees (0-4) 2.551
27. Vatra Dornei Wraiths (1-3) 2.465
28. Kilkenny Cannibals (1-3) 2.392


Facing Relegation

29. Malmo Dawgs (1-3) 2.362
30. Chanhatten Knights (0-4) 2.314
31. Coventry Thunder (0-4) 2.184
32. London Brawling Tommies (0-4) 2.02




Now.....Discuss

Looks pretty good on the surface. Thanks for compiling this.

Fortunately for London, at the moment we cannot be relegated. This is already the bottom league

This is a good format. Most of my own players are on a team that was in Europe East AAA#2 last season and Sawedust did a great job.

Originally posted by Stehmar
Fortunately for London, at the moment we cannot be relegated. This is already the bottom league

This is a good format. Most of my own players are on a team that was in Europe East AAA#2 last season and Sawedust did a great job.



Sweet just noticed most of your guys are on LBT....glad to see you guys at 4-0 and have turned it around. Agreed, Sawedust is the man....

thx for the work rock

Anyone else find it strange that Sooner Machine is competeing for a top spot after being blown out by 100+ points in their last game and being winless? Even if the margine of defeat is 35 points I don't think a winless team should be hat high. Oh well, I'm not too good with math anyway.

Yeah i was just thinking about the Sooner Machine, something seems off about that

I hate math o.o
But looks like a pretty good work X)

Originally posted by Thom Yorke
Anyone else find it strange that Sooner Machine is competeing for a top spot after being blown out by 100+ points in their last game and being winless? Even if the margine of defeat is 35 points I don't think a winless team should be hat high. Oh well, I'm not too good with math anyway.

My thought exactly. CCSM mauy very well be much better than their record, but damn.... They're head of a heap of 2-2 teams.

Aside from that, everything looks good. Glad the process of mathematics supports our season as much as my team does.

Originally posted by Thom Yorke
Anyone else find it strange that Sooner Machine is competeing for a top spot after being blown out by 100+ points in their last game and being winless? Even if the margine of defeat is 35 points I don't think a winless team should be hat high. Oh well, I'm not too good with math anyway.

Its a numbers game thats partly based on who you've played. Even though they are winless they have played Dublin Monkees, the Reapers, Basin City Blue, and the Stars. Now look where those team rank. Stars are number 1, and the 2 others in the top 15 and the last in the top 20. If they continue to lose to lesser teams they will slide down drastically, but as of right now they are a victim of circumstance. They have played 4 of the "better" teams early on.

Originally posted by DaRockShow

Early on there is no clear cut "stronger" conference. Out of the top 10 there is a 5 and 5 split between Alpha and Beta. This is likely to change as the season goes on, but for now we are pretty even.
...
Now.....Discuss

First, as I posted in the Zeta thread the rankings for the top 8 Zeta teams are almost identical to what I posted as my estimates earlier today. So the formula matches the intuition pretty well.

Second, the conferences don't play each other until the last week, right? That makes the results completely independent for the conferences. You could throw Alpha and USA Pro in and it wouldn't show which is better--it would have the top teams in Alpha ranked along with the top teams in USA Pro, and vice-versa. Until teams play each other (or common opponents) the formula won't sort them relative to each other.

Because of that, it would be more correct to list separately by conference (Alpha 1 through Alpha 16, and Zeta 1 through Zeta 16) rather than mingle them together. The formula is saying the Smugglers are better than the Thunder and the Stars are better than the Tommies, but it's not judging Smugglers vs. Thunder. It's not even saying the Smugglers are better than the Tommies, or the Stars are better than the Thunder.

Originally posted by SantosLHalpar
Originally posted by DaRockShow


Early on there is no clear cut "stronger" conference. Out of the top 10 there is a 5 and 5 split between Alpha and Beta. This is likely to change as the season goes on, but for now we are pretty even.
...
Now.....Discuss

First, as I posted in the Zeta thread the rankings for the top 8 Zeta teams are almost identical to what I posted as my estimates earlier today. So the formula matches the intuition pretty well.

Second, the conferences don't play each other until the last week, right? That makes the results completely independent for the conferences. You could throw Alpha and USA Pro in and it wouldn't show which is better--it would have the top teams in Alpha ranked along with the top teams in USA Pro, and vice-versa. Until teams play each other (or common opponents) the formula won't sort them relative to each other.

Because of that, it would be more correct to list separately by conference (Alpha 1 through Alpha 16, and Zeta 1 through Zeta 16) rather than mingle them together. The formula is saying the Smugglers are better than the Thunder and the Stars are better than the Tommies, but it's not judging Smugglers vs. Thunder. It's not even saying the Smugglers are better than the Tommies, or the Stars are better than the Thunder.

Doing it that way would of course be easier for me since my team is in Zeta. I could cut out 16 Alpha teams and be done a lot faster, but that would suck for them. I figured doing it this way would be more interactive for both conferences. True enough we dont meet eachother until the last week of the season, but ultimately the best Zeta team will face off with the best Alpha team in the championship. So to track the rise (and fall) of teams in both conferences this would be the best way to go about it.

I was mostly just trying to explain your observation " Out of the top 10 there is a 5 and 5 split between Alpha and Beta." Until teams play each other, they're going to be ranked as though the conferences were equally tough.

If you ran it on all the teams in the NFL and the Ivy league this past year, it would spit out something like:

1. New England Patriots
2. Harvard University
3. Indianapolis Colts
4. Green Bay Packers
5. Yale University
...

The relative positions of teams within one league/conference are meaningful, but the positions between conferences don't mean anything.

Originally posted by DaRockShow
Originally posted by Thom Yorke
Originally posted by Thom Yorke
Anyone else find it strange that Sooner Machine is competeing for a top spot after being blown out by 100+ points in their last game and being winless? Even if the margine of defeat is 35 points I don't think a winless team should be hat high. Oh well, I'm not too good with math anyway.

Its a numbers game thats partly based on who you've played. Even though they are winless they have played Dublin Monkees, the Reapers, Basin City Blue, and the Stars. Now look where those team rank. Stars are number 1, and the 2 others in the top 15 and the last in the top 20. If they continue to lose to lesser teams they will slide down drastically, but as of right now they are a victim of circumstance. They have played 4 of the "better" teams early on.

Yeah, I could see that.... if it weren't for the Blues being ranked below CCSM...

Well i mean i think it has alot to do with strength of schedule. Not only did CCSM play some tough teams arleady but when CCSM loses it gives them points in strength of schedule