User Pass
Home Sign Up Contact Log In
Forum > Oceania > Oceania AAA Power Rankings - Week 3
Page:
 
shutd0wn
offline
Link
 
Oceania AAA Power Rankings

1. 1.667 - Brantford Mustangs (3-0-0)
2. 1.000 - Tonga Ikale Tahi (2-1-0)
3. 0.667 - PUMPED UP PISTONS (3-0-0)
3. 0.667 - The Bomb Squad (3-0-0)
5. 0.666 - FFLW Caninites (3-0-0)
5. 0.666 - So Cal BDC (3-0-0)
5. 0.666 - Gold Coast Crocs (1-1-1)
8. 0.334 - Springfield Sewer Rats (1-2-0)
9. 0.333 - Arctic Panthers (3-0-0)
9. 0.333 - Sydney Sharks (3-0-0)
9. 0.333 - Dunedin Beavers (3-0-0)
9. 0.333 - Cajun Crushers (2-1-0)
13. 0.167 - Sth Queensland Crushers (2-0-1)
14. 0.001 - Melbourne SWG (1-2-0)
15. 0.000 - Utopia Kangaroos (2-1-0)
15. 0.000 - Toowoomba (OTF) Centurions (2-1-0)
15. 0.000 - Tonga Strikers (2-1-0)
15. 0.000 - Sunshine Coast Wallabies (1-2-0)
15. 0.000 - Canberra Stingers (1-2-0)
20. -0.333 - Sparta Spartans (1-2-0)
20. -0.333 - Los Cabolo Sharks (0-3-0)
20. -0.333 - Fiji Storm (0-3-0)
20. -0.333 - Avondells Raiders (0-3-0)
24. -0.334 - Samoa Flying Foxes (2-1-0)
25. -0.666 - Le Thor Bulls (0-3-0)
25. -0.666 - Oregon Flat Backers (0-3-0)
27. -0.667 - Sydney Thunder (0-3-0)
27. -0.667 - Newcastle Surfers (2-1-0)
27. -0.667 - District 5 Ducks (0-3-0)
27. -0.667 - Port Moresby Demons (1-2-0)
31. -0.999 - The Practice Squad (0-3-0)
32. -1.667 - Barossa Grapestompers (0-3-0)

Source: http://www.sysfx.org/GLB/rank.php


Formula:
Now I know some people are looking at this and might be scratching their heads, especially seeing some 2-1 teams ranked so low. Let me briefly explain how it works.

The ranking numbers are based on the record of the opponents you win and lose to. So someone who beats winning teams is going to be ranked higher than someone who beats teams who haven’t won a game. The kicker is when you get upset against a losing team. Your ranking drops when you lose to teams you should have beat.

Kind of like this:

W = YOUR RANK + (OPP W%)
L = YOUR RANK – (1.000 - (OPP W%))

So for example:
So Cal BDC (GCC vs. BDC) - Win) (GCC WP: 0.333) 0.333
So Cal BDC (BDC vs. SCW) - Win) (SCW WP: 0.333) 0.666
So Cal BDC (BDC vs. LCS) - Win) (LCS WP: 0.000) 0.666

It is a formula ive seen work before and appears to be working good now. Maybe we will fine tune it later if anyone has better ideas.
 
ougijoe
offline
Link
 
The formula is sound... but not after week 3.

Run it again about half way through the season
 
shutd0wn
offline
Link
 
Its automatic so it can be ran anytime. All the data is live and pulls from a database.
Last edited Apr 30, 2008 20:58:31
 
ougijoe
offline
Link
 
But if it's all based on win percentage... No
 
turnit643
offline
Link
 
Nice job shutdown. Would be cool to see it factor in some scoring for & against and/or some yardage stuff too, but hey I won't argue with the #2 spot for the Ikale Tahi.
 
AdamD210719
offline
Link
 
Cool post! Its good to see our team name on something! =)
 
ekspa
offline
Link
 
It has the Panthers who are 3-0-0 ranked lower than teams who are 1-2-0?
 
Leionidas
offline
Link
 
Originally posted by AdamD210719
Cool post! Its good to see our team name on something! =)

Good to see your supporting the Beavers that Avatar must've cost you a fortune at the Beaver shop.
 
McGruffHawk
offline
Link
 
Originally posted by ekspa
It has the Panthers who are 3-0-0 ranked lower than teams who are 1-2-0?


Your opponents are a combined 1-8. That's why.
 
shutd0wn
offline
Link
 
Originally posted by McGruffHawk
Originally posted by ekspa

It has the Panthers who are 3-0-0 ranked lower than teams who are 1-2-0?


Your opponents are a combined 1-8. That's why.


Exactly, it pretty much confirms you haven't played anyone tough yet. With this formula, you dont get highly rewarded on beating teams you should have beat. This is why Tahi is ranked above a bunch of 3-0 teams. Their only loss is coming from a team that has not lost a game yet, while their wins are coming from teams who have a couple of wins themselves.

As ougijoe said, it only being the 3rd game of the season really doesnt help the numbers much. We still have 8 undefeated teams in the league. But as the season progresses you will see the numbers grow stronger. The better teams will rise to the top and the bottom half will continue to sink.

I am open for discussions on maybe a better formula as long as it doesnt involve the tedious task of gathering a lot of stats.
Last edited May 1, 2008 10:32:11
 
stevegamer
offline
Link
 
The problem with doing this so early is that teams could be losing because they play good teams, or because they are inherently bad.

The winless teams ahead of teams with 2 wins is much more problematic to my thinking.

I'd round the numbers to just 2 digits to remove the .666/.667 variances, and then add like 0.05 per win to get the ties broken in favor teams without more wins to reduce any manual work.
 
turnit643
offline
Link
 
I was playing with it a bit using your "Strength of Victory" type of ranking as one factor combined with a "Production" Ranking that looks at Points Scored/Allowed and Yardage Gained/Allowed. I'd love to figure out how to automate it, but I have no idea how to do that, so I just grabbed the numbers manually and played with them in Excel.

For the Production Ranking, I ranked each of the 32 teams in the four categories I mentioned above then gave each team an Offensive Production Rank (combined Points_Scored and Yardage_Gained) and a Defensive Production Rank (combined Points_Allowed and Yardage_Allowed). From there I just added those two ranks to your SOV Ranks and came up with an Overall Score for each team. It's pretty simple, maybe too simple, but I didn't want to get into weighting factors or anything else much beyond that.

For fun, here's what it produced ...

Overall Rank, TEAM, (record), SCORE [SOV_RK, OFF_RK, DEF_RK]

1. The Bomb Squad (3-0-0) , 8 [ 3 , 1 , 4 ]
2t. FFLW Caninites (3-0-0) , 10 [ 5 , 4 , 1 ]
2t. So Cal BDC (3-0-0) , 10 [ 5 , 2 , 3 ]
4. Tonga Ikale Tahi (2-1-0) , 13 [ 2 , 7 , 4 ]
5. PUMPED UP PISTONS (3-0-0) , 17 [ 3 , 12 , 2 ]
6. Cajun Crushers (2-1-0) , 22 [ 9 , 7 , 6 ]
7. Dunedin Beavers (3-0-0) , 23 [ 9 , 3 , 11 ]
8. Arctic Panthers (3-0-0) , 24 [ 9 , 5 , 10 ]
9. Sydney Sharks (3-0-0) , 31 [ 9 , 16 , 6 ]
10. Springfield Sewer Rats (1-2-0) , 33 [ 8 , 12 , 13 ]
11. Brantford Mustangs (3-0-0) , 35 [ 1 , 19 , 15 ]
12t. Gold Coast Crocs (1-1-1) , 36 [ 5 , 9 , 22 ]
12t. Toowoomba (OTF) Centurions (2-1-0) , 36 [ 15 , 12 , 9 ]
14. Sth Queensland Crushers (2-0-1) , 38 [ 13 , 17 , 8 ]
15. Tonga Strikers (2-1-0) , 40 [ 15 , 5 , 20 ]
16. Utopia Kangaroos (2-1-0) , 42 [ 15 , 11 , 16 ]
17. Sparta Spartans (1-2-0) , 52 [ 20 , 20 , 12 ]
18. Newcastle Surfers (2-1-0) , 53 [ 27 , 10 , 16 ]
19. Samoa Flying Foxes (2-1-0) , 58 [ 24 , 18 , 16 ]
20t. Melbourne SWG (1-2-0) , 59 [ 14 , 21 , 24 ]
20t. Avondells Raiders (0-3-0) , 59 [ 20 , 23 , 16 ]
22. District 5 Ducks (0-3-0) , 68 [ 27 , 27 , 14 ]
23. Sunshine Coast Wallabies (1-2-0) , 69 [ 15 , 29 , 25 ]
24. Port Moresby Demons (1-2-0) , 70 [ 27 , 15 , 28 ]
25. Canberra Stingers (1-2-0) , 72 [ 15 , 28 , 29 ]
26. Los Cabolo Sharks (0-3-0) , 73 [ 20 , 26 , 27 ]
27. Le Thor Bulls (0-3-0) , 75 [ 25 , 29 , 21 ]
28. Fiji Storm (0-3-0) , 76 [ 20 , 25 , 31 ]
29. Oregon Flat Backers (0-3-0) , 77 [ 25 , 21 , 31 ]
30. Barossa Grapestompers (0-3-0) , 79 [ 32 , 24 , 23 ]
31. Sydney Thunder (0-3-0) , 83 [ 27 , 31 , 25 ]
32. The Practice Squad (0-3-0) , 93 [ 31 , 32 , 30 ]

edit: Since this "score" is a combo of three 1-32 ranks, the best possible is 3 and the worst possible is 96.
Last edited May 1, 2008 12:49:54
 
shutd0wn
offline
Link
 
That one is actually pretty cool and displays similar results like Ranking the Flying Foxes as the worst 2-1 team and putting the Sewer Rats up in the top 10 even though they're 1-2.
 
turnit643
offline
Link
 
Weighting your SOV Rank 1.5:1 as compared to the Production Rank would smooths it a bit further I'd guess, but I just posted this one as an example.
 
shutd0wn
offline
Link
 
What is your formula for SOV Rank? PF-PA?
 
Page:
 


You are not logged in. Please log in if you want to post a reply.