Week 9 Rando Rankings

[Guest]

Week 9

1. Morgantown (7-0)
Score: 1.625 Last Week: 1 In Game: 1

Morgantown holds their place at the top of the rankings while taking a week off. They did, however suffer a slight loss in overall score, as a result of the their former victims going 2-4 this week Up next for the undefeated Maniacs, the winless Hitmen.

2. Memphis (7-1)
Score: 1.477 Last Week: 2 In Game: 3

A 10 point win over Motor City, not enough to put them ahead of Morgantown, but it was enough to keep them ahead of the team at #3.

3. St. Paul (6-1)
Score: 1.430 Last Week: 3 In Game: 4

They beat Chicago by 29, but it still wasn't impressive enough for them to move past Memphis, but a win next week over Yuma should be more than enough.

4. Utah (6-1)
Score: 1.399 Last Week: 7 In Game: 2

A win over 4-3 Denver helped push them back up the rankings here, but it didn't help as much as the numbers 4, 5 and 6 teams from last week all losing this week.

5. Yuma (5-2)
Score: 1.218 Last Week: 11 In Game: 5

The Aces are the only addition to the Top 10 this week and are here by way of a 31-6 drubbing of last week's number 4 team, North Liberty Enforcers.

6. North Liberty (5-2)
Score: 1.203 Last Week: 4 In Game: 7

The team on the receiving end of the above mentioned drubbing. After getting of to as hot a start as anyone in the league, the Enforcers are now 1-2 in their last 3 games. They should be able to bounce back versus Pearl Harbor this week, but looming on the horizon, a week 1 rematch against St. Paul.

7. San Francisco (5-3)
Score: 1.176 Last Week: 5 In Game: 8

After dropping their last 2 in a row, the Bayhawks get a week off to try and regroup and get ready for a trip to division leaders Yuma next week.

8. Las Vegas (5-3)
Score: 1.170 Last Week: 8 In Game: 10

No matter how I tried, I couldn't find the .007 points that the Outlaws needed to move past the team they beat last week, San Fransisco, but a win this week versus...oh wait, bye week. Well, maybe if the Bayhawks lose to... damn. Well, enjoy the view from 8 guys.

9. Anchorage (5-3)
Score: 1.157 Last Week: 5 In Game: 9

They fall 4 spots after a loss last week, but the good news for the Gladiators? They're done playing Kalispel this year (unless they meet them in the playoffs.)

10. Gainesville (5-2)
Score: 1.150 Last Week: 10 In Game: 11

The Gaels finish their sweep of the NC East, but fail to move up in the rankings here. The reason why? The NC East is a combined 9-22.

And the rest.

11. Kalispel
12. Denver
13. Port City
14. Lake County
15. Portland
16. Phoenix
17. Dakota
18. New Orleans
19. Los Alamos
20. Punxsutawney
21. Columbus
22. Camden
23. Ohio Valley
24. Washington
25. Motor City
26. Charlotte
27. S. Maryland
28. Waco
29. Boston
30. Pearl Harbor
31. Chicago
32. Brooklyn

[Guest]

I had some requests last week for the formula used in calculating the Rando rankings so I figured I'd do one better, and try an offer an explanation on not only how it works, but why it works that way (most common reason: I'm lazy).

As with most Mathematical or Computer Rankings, the formula may determine the ranking, but that formula itself is determined by the subjective definitions of success of the person who created it.

The final Rando Ranking score is based on three main components, Adjusted Record (40% of the score), Performance Metrics (48%), and Momentum Score (12%).

The first of these, Adjusted Record is the easiest to calculate and explain. It is simply (Winning Percentage * Opponent Win Percentage). I had tried several, more complicated formulas. Things that took into account the winning percentage of the teams you had beaten versus the winning percentage versus the teams you had lost to, and even one that weighted wins and losses based on the overall ranking of the teams you had played, but in the end none of these methods produced noticeably different, or subjectively (I says so) better results. And they all took a lot more work to figure, so I went simple. The five teams and bottom team in term of Adjusted Record:

1. Morgantown 0.444
2. San Fransisco 0.426
3. Memphis 0.423
4. St. Paul 0.396
5. Anchorage 0.384
...
32. Brooklyn 0.000

The next component, Performance Metrics, is the most complicated. I'm not even sure I can explain it fully. It starts out with three measures of "success" other than W-L ratio, Points Differential (Points Scored minus Points Allowed), Total Turnover Differential (Opponent's Fumbles, Interceptions and Punts minus Fumbles, Interceptions and Punts) and YPA Differential (Yards per Pass Attempt minus Opponent's Yards per Pass Attempt). Teams that do well in W-L also (not surprisingly) tend to do will in these areas, although it is debatable if this is a chicken or egg effect (do winning teams do well in these areas because they are winning teams, or do they win because they do well in these areas?), but there is a correlation. Each resulting differential is then weighted upwards equally so that there are no negative values and then multiplied by your opponent's win percentage (running up the score on bad teams isn't as impressive and winning big over other winning teams). Finally these three numbers are weighted so that they are equal, 1/3 components of a final score. ( Point Dif. And TTO Dif. are divided by 100, YPA Dif. by 25. Not perfect, but close enough for me, and once again, I'm lazy) The top five and bottom teams in the Performance Metrics:

1. Morgantown 0.493
2. San Francisco 0.469
3. St. Paul 0.457
4. Anchorage 0.443
5. Memphis 0.448
...
32. Brooklyn 0.108

Familiar bunch, isn't it? You'll notice though that San Francisco is #2 in both rankings but is only 5-3 overall and #7 in the final rankings. This is because their Adjusted Record and Performance Metric score are both being modified by their league best 0.614 Opponent Win Percentage. They've played the toughest schedule in the league. But just in case you're worrying that the system over values OWP to highly, the top 5 all have OWPs below .500 and Chicago who has played the second toughest schedule at 0.604 OWP is still ranked 31st despite their high modifier.

The other reason San Fran isn't in the top 5 like thier Adjusted Record and Performance Metrics would indicate is thier Momentum Score. This is the "What have you done for me lately?" part of the rankings. It awards teams a flat .05 points for every win in the last 3 try (I don't care who you beat a month ago) and bonus points for you last victory (0.1*(OWP+0.01) on a penalty for your last loss (-0.025/(OWP+0.01). This formula was actually just changed from last week to fix a possible divide by zero error that might occur should Brooklyn ever win a game. And in case you are wondering the base values for the win or loss were chosen so that the sum of all the bonuses for each week would be very close to 0, that is to say that the sum of all of the bonuses for winning would equal the negative sum of all the penalties for losing.

The top 5 and bottom teams in term of Momentum.

1. Morgantown 0.189
2. Memphis 0.189
3. Gainesville 0.189
4. Port City 0.189
5. Yuma 0.172
...
32. Brooklyn -0.065

So there's the explanation, bit more complicated than just giving you for formula I suppose, but without it the following would be worthless.

score=(WP*OWP)+(((PS-PA+20)*OWP)/100)+((OF+OI+OP-F-I-P+20)*OWP)/100)+((YPA-OYPA+5)*OWP)/25)+((0.05*WILTG)+ either (0.1*OWP+0.01) or (-0.025)*OWP+0.01)