Week 5 Rankings

I've refined my ranking formula and this will probably be the start of further refinements.  I'm still looking at actual scores from the last 4 weeks of FLOF and assigning those scores an expected win percentage based on the last 2 years worth of data.  In the past, I roughed this by figuring out general groupings.  So I knew that an average week would have 2 scores above 86, sometimes more, sometimes less.  But that tells us that if you scored 87 or more you would likely beat all but one other team that week.  But scoring 87 is not the same as scoring 120.  In fact, an 87 gives you roughly an 82% chance of victory while a 120 is more like 99%.  So, now I'm assigning much more specific expected win percentages to everyone.


All stats are from the last 4 games only.  EWP stands for expected win percentage.  Variance is calculated as the standard deviation of that teams scores.

1) Mimes (no change)
median EWP = 93%, avg EWP = 92%, variance = 6%
I think this really shows how crazy it is that Mimes went 2-2 over these 4 games.  Not only do they have the highest EWP they also have the lowest variance by far.  Hamburglers have been the next most consistent team and their variance is more then twice Mimes'.

2) Malleus (up 4)
median EWP = 89%, avg EWP = 73%, variance = 38%
On the other hand Malleus have had real close to the highest variance over the last 4 games.  Their week 2, 53 pt game scored an EWP of 18% and is the reason for that high variance.  Over their last 3 they have been Mime-like.

3) Drow (down 1)
median EWP = 80%, avg EWP = 77%, variance = 20%
They posted a better avg EWP than Malleus but I've always valued medians over averages in this space.  While Malleus' worst scores were from weeks 1 & 2, Drow followed up two 93% EWP games in weeks 2-3 with a 54% week 4 and 67% week 5.

4) Lagers (no change)
median EWP = 78%, avg EWP = 63%, variance = 39%
Another high variance team, Lagers week 2 was worth 91% EWP but their week 3 was worth 6%.  Although that week 3 low is the only big blemish this season.

5) Beavers (up 2)
median EWP = 64%, avg EWP = 71%, variance = 17%
Beavers have been real consistent after their rocky week 1 start.  In fact, if I was going to rank teams based on their best, worst game Beavers would be 2nd to Mimes.  Beavers' worst game over their last 4 was a 72 pt, 59% EWP week 5 outing.  Based on their average EWP, Beavers should have won 3 of their last 4 which makes their 1 win over that time the least lucky in FLOF.

6) CAPO (down 1)
median EWP = 66%, avg EWP = 56%, variance = 6%
CAPO could make an argument for 5th but their median EWP just edge out Beavers' while their average is far below.  CAPO look like an unlucky team given that they lost both games when they scored 80 pts (76% EWP) but they won week 3 with only a 56% EWP and tied week 4 with a 17% EWP.

7) Rednecks (up 1)
median EWP = 58%, avg EWP = 63%, variance = 21%
Rednecks really broke out last week with their first game above 67% EWP (96pts, 92% EWP).  This also skewed their variance with had been very small before that game.

8) Old No. 7 (down 5)
median EWP = 51%, avg EWP = 48%, variance = 42%
What happened to this team?  That gigantic variance is the result of a drop from week 2 & 3 EWPs of 84% and 85% to the last 2 weeks' EWPs of 18% and 6%.

9) Bill (no change)
median EWP = 37%, avg EWP = 42%, variance = 39%
Bill has often been a lucky team in FLOF.  At least it seems that way.  He went 2-1-1 by winning any game he had a positive EWP in and picking up a tie despite a 18% EWP in week 4.  His week 2 5% EWP just drags all his stats down.

10) Hamburglers (up 1)
median EWP = 35%, avg EWP = 38%, variance = 14%
Hamburglers really wish they weren't consistent.  At least not like this.  They only had one game with a positive EWP all year (week 2) it was only 56%.

11) Ramjets (up 1)
median EWP = 27%, avg EWP = 32%, variance = 36%
After a good week 2 (80pts, 76% EWP) Ramjets had two consecutive games under 10% EWP before "recovering" in week 5 with 67 pts (49% EWP).

12) Flannel (down 3)
median EWP = 21%, avg EWP = 34%, variance = 42%
Their 4-game history looks exactly like Ramjets except their week 2 was a bit better (96pts, 92% EWP) and their week 5 was a bit worse (61pts, 35% EWP).