Wednesday, November 12, 2014

AFFL Tables and Stuff

So...there is no review this week because I took off work yesterday to play golf on what was probably the last nice day of the year. Plus, you all know what happened already, so whatever.

But I had some free time today because work is winding down for the year, so I thought I'd analyze each team's scoring for and against on the season, and see if we could prove or disprove some of the theories that are flying around. 

One doc with a variety of tables on it is attached here.

First, I looked at each team, took the points scored in their wins, totaled that up and got an Average Points Scored Per Win. Then I took their losses and did the same for Average Points Scored Per Loss. Then I took all points scored in the league, divided by 12 teams, then divided by 10 weeks and figured out that average points scored per game is 83.1.

The conclusions seem to be:

  • For the most part, teams have the records they deserve based on their points scored, with a handful of teams noticeably skewing the results. 
  • As expected, you can still rack up wins with below average points for, but above average points against are very hard to overcome.
  • The top 6 teams in Points Per Win have 31 Wins this year, the bottom 6 have 29 Wins.
  • The top 6 teams in Points Per Loss (most points scored per loss) have 27 Wins, the bottom 6 have 33 wins.
  • HGAC's 133-115 loss to Animal House really skews the Points Scored In Losses table for them. Take that wild game away and they are averaging 73.6 points scored in their 3 (adjusted) losses.
  • With that adjustment, ISIS has the highest average points scored in losses, followed by Riggins'. 
  • Did you know that Bri averages 94.3 points in his wins, but only 60.2 in his losses?
  • Using our league average of 83.1 points, I then looked at each team's wins and losses. Figuring that being above or below 83.1 points should give you a better or worse chance then average of winning or losing, I gave everybody a Lucky Win for a game they won while scoring under 83.1, and an Unlucky Loss for a game they lost while scoring over 83.1. 
  • Chris, amazingly, has all Lucky Wins as part of his 3-7 record. He could conceivably be 0-10 right now.
  • On the other hand, he also has 3 Unlucky Losses, so he could be 6-4. Or 3-7, like he is.
  • Since you're wondering, Scott has 3 Lucky Wins and 0 Unlucky Losses.
  • If you net out Lucky/Unlucky by this formula, HGAC moves up to 7-3. Riggins also moves to 7-3. Most other teams jostle around to 6-4 or 5-5. Scott falls to 4-6, and Brian Barker falls to 3-7. The Rippers actually fall to 2-8.

Then, thinking about Gary's preposterous claim that it's better to ignore unpredictable stars in favor of steady guys who score about the same each week, I tried to examine point differentials and see if it correlated to anything.

  • Chris had the smallest differential in Points Scored Per Win and Points Scored Per Loss, with 5.3 points. Then it was ISIS with 11.9, Runny with 12.9, and Rippers with 13.9.
  • The biggest differential in scoring in their wins and losses is HGAC with 36.4, then Kangaroo with 34.1, Jesus with 33.5, and Animal with 30.5.
  • The top 3 teams in win/loss differential (tightest range) are 13-17. The bottom 3 teams (widest range) are 18-12. 

But Gary is a smart guy who hardly ever comes with really wild theories that make absolutely no sense. In the 20+ years I have known him, he has only had like 40-50 ideas that were ridiculous. 

So I looked at the differential between a team's 1 best score and their 1 worst score. Chris, at 3-7, had the smallest differential with 34.5 points. Second best was the Rippers, also 3-7, with 35.0. The widest differentials were Animal at 81.5 and HGAC at 81.4, both teams being 6-4.

But every team can have 1 super good week or 1 super bad one, right? We see it all the time. So then I took each's team average points per week and cross-referenced each individual score. I wanted to see how many times a team scored within 6 points (1 touchdown) of their overall average. Which teams were the steadiest performers?

Well, Gary, of all people is the most varied team in the league. Only 1 time did he score within 6 points of his overall average. Jesus and Kangaroo also did this only 1 times. Those teams are a combined 18-12. Larry, Chris, and M. Barker were the best at scoring within a TD of their average, doing it 6, 5, and 4 times, respectively. Those teams are a combined 9-21.

So it seems like being fairly steady each week will just get you killed. You need to have a wild team that can go off in any given week and beat any opponent and get enough wins that way to ride out your dud weeks.

Then I broke out each team's Average Points For and Average Points Against each week, showing the differentials. All this shows, again, is that Points Against is a bigger factor in team's record than points for. 

Lastly, to increase the sample size, I looked at overall records since I joined the league, commonly known as the AFFL's "Golden Age."

TDHAF: 34-18
RUNNY: 31-21
ANIMAL: 30-22
ISIS: 27-25
RIPPERS: 25-27
RIGGINS: 22-30
SIGMUND: 17-35

But since I'm not an AFFL creationist, I realize there was a savage time long ago when I wasn't in the league. So I looked at all-time W/L records, focusing on the best team in league history, and the worst.

BEST - KANGAROO: 220-156

So there's that.

No comments: