A New Points Record

RJ set a new points record against Hartman in the toilet bowl semi-final. I thought he was a shoe-in for last place this year. His team stinks. But he shattered Hartman’s record against Hartman! That should teach Hartman never to set records again. So here is what the updated leaderboard looks like for most points in a game:

Owner Team Year Week Points
Thieneman 12.5 PSI 2015 R2 192.6
Hartman Name Your C. Johnson 2015 03 182.6
Shokunbi Team Shokunbi 2014 13 182.0
Skrzyszewski Oozing With Machismo 2013 13 181.5
Harrington Cameron’s Cuties 2011 13 175.1
Hartman Peter Pan’s Poor Nanna 2013 02 173.0
Skrzyszewski Oozing With Machismo 2013 06 171.5
Regan Losman’s Gamblers 2009 06 171.0
Ready Detroit Ride Or Die 2013 R2 170.8
Harrington Cameron’s Cuties 2015 02 169.6

Aside from the gaudy point total, the surprising thing here is who is setting the record. It’s RJ and his 12.5 PSI in what has quickly become a dated reference for a fantasy football team name. RJ was in contention for setting the all-time season low point total until a few big weeks at the end pulled him out of it.

How can we square this nearly historically bad team with a points record? Let’s consider the percentage of a season’s total points achieved in each game. If you score roughly the same amount each week, you’d score about 7% or 8% of your season’s point total in a given week. Due to not having played Round 3 this year yet, and some Round 1 BYE teams with non reported scores from early in league history, I took the percentage of points each game counted toward season totals for Weeks 1-13 and Round 2. In the below graph, each dot represents one of 1,148 games played and how much of that owner’s season point total that game was responsible for. So how big of an outlier was RJ last week? See for yourself:

He’s way, way out there! If Hartman ends up taking the SAT he’ll be cursing the D. Johnson, D. Woodhead combo. Did you know that:

  1. RJ didn’t have a single week this year in which he would have beaten Hartman’s 136.4 points? (until last week)
  2. RJ once went three consecutive weeks this season without scoring 192.6 total points? (Weeks 3, 4 and 5: 184.9 total points)
  3. If we assume these data are normally distributed, (seems reasonable when you don’t consider RJ’s data point) that last week we saw a 5 sigma event? The chances that a team would score nearly 15% of their season’s point total in a single game is less than 1 in 100,000,000,000,000!

Hartman does now!!

An Update on Playoff Probabilities

Several weeks ago I posted some playoff odds for each team. As a check, I cross-referenced previous years’ predictions under the same methodology. I postulated that my scoring was better than random guessing, if you score by how many teams who were the most likely to finish in a given place actually earned that place. Most of my scores for each were 2/12 or 3/12. Turns out my method is no better than random guessing under similar constraints (seeds 7-12 finish in place 7-12, 1 and 2 seeds must finish in 1st-4th place).

Rather than calculate it directly I simulated data in order to determine, that, even randomly assigning teams to places the most likely outcome is to get two right! (27% of the time). Anyway, next year I commit to better methodology and better validation before I post something. Clunky and inefficient R code is here if you’d like to see the details of the calculation/simulation.

Farewell Ola

As many of you saw, Ola announced his retirement from fantasy football last weekend. Suprising most everyone, I would say it was a little Barry Sanders-ish. But let’s not get ahead of ourselves, it’s more Ricky Williams-ish.

I’m giving Ola a proper Stats Corner send-off. Many are wondering if he’s riding off into the sunset at the ripe old age of 28. But

  1. You have to win your last game in order to ride off into the sunset.
  2. He doesn’t have anything to ride off in. Remember?

Ola, this is your fantasy life:

So long, Ola. This league will miss you.