This week's edition of The Weekly League features:
1. Some notes on the peculiar joy derived from the current iteration of the San Diego Chargers, a team whose generic win probability (GWP) surpasses considerably its actual win-loss record.
2. An examination -- via some white-hot tables -- of the NFL's luckiest and unluckiest teams (as measured by GWP versus current, actualy win-loss records) through Week Five.
and
3. A table that proposes to offer something called "EPA Standings" -- itself an idea of minimal importance -- followed by even less important comments.
Let The Weekly League into your life, America!
As usual, a glossary of all unfamiliar terms can be found here.
Idle Thoughts on the Chargers
As our fearless leader Brian Burke noted recently in these electonic pages -- and as reader James noted even more enthusiastically in the comment section -- the San Diego Chargers are frigging awesome right now. The team currently sports a GWP of 0.81 -- this, despite an underwhelming strength of schedule (as suggested by the team's 0.42 opponent GWP). They feature one of the league's better defenses and the best offense.
For the footballing nerd, though, the Chargers' excellence isn't even the most interesting thing about the current interation of that team. No, the most interesting thing is that the team currently stands at 2-3. For those fans who merely take the standings at face value, the Chargers are a middling team.
This is one of those moments when the advanced metrics allow us to see -- as either Melville or Chuck D (I forget which) once said -- "a little lower layer." As you walk the streets of your fair city, talk beside the water cooler at your workplace, enjoy an adult beverage at your local drinketeria, you can look about yourself and say with some authority, "Basically all these people think the Chargers are merely an average team. Imagine how surprised they'll be when the Chargers end up being awesome!"
Of course, you, the footballing nerd, will decidedly not be surprised. For he (she?) knows that, of the 106 points the Chargers have conceded so far this season, an incredible 44 of them have come by means which are more a function of random variance than any kind of actual weakness. The Chargers have been the victims of two kick-return TDs, two fumble-return TDs, a punt-return TD, a blocked-punt TD, and a safety. Of course, those things have happened -- there's no denying that -- but research shows that they're hardly the product of repeatable skills (or lack-of skills, as it were). The things that are repeatable -- pass efficiency, limiting pass efficiency -- the Chargers are really good at those things.
GWP Standings and Luck
As you're very probably aware, Mr. Burke publishes the GWP standings in these pages each week. While we will, I'm sure, all agree that his research -- and the presentation thereof -- is infallible, one question I sometimes find myself asking is how the GWP standings might look when placed beside the actual current standings.
The table below represents an attempt to capture that relationship. In said table, you'll see wins, losses, and winning percentage. You'll also find GW and GL -- that is, wins and losses as determined by GWP. Finally, you'll see GLUCK -- or the numbers of wins by which a team has been unlucky (in green) or lucky (in red) relative to their GWP.
• While the New York Jets aren't bad, per se, and certainly won't be hurt by the return of Santonio Holmes and Calvin Pace, they are very likely not the best team in New York.
• San Francisco coach Mike Singletary has given very clear indications that he's willing to hand over quarterbacking responsibilities to David Carr. While I, a tiny guy writing from a tiny netbook computer, am not necessarily in a position to contradict him, I feel comfortable stating that the Niners are not as bad as their 0-5 record.
• Chicago is 4-1? For those of us who say Jay Cutler get almost assassinated two weeks ago, this is hard to believe.
An "EPA Standings" of Sorts
Because I (a) have free time and (b) own Microsoft Excel, I occasionally like to get all intimate with the numbers here at Advanced NFL Stats. Recently, I wondered how closely EPA and Opponent EPA might correlate with actual points scored and allowed.
The following table represents and attempt to capture that relationship. For each team, I've included not only points scored and allowed per game (PTS/g and OPTS/g, respectively), but also something I'll call EPA Points and Opponents EPA Points (EPTS/g and OEPTS/g, respectively). To calculate EPA Points, I've added each team's EPA to the league-average points scored times a constant (in this case, 0.93) to normalize the per-game numbers to the actual league averages. The equation looks like this: EPA + (LgAvg PTS * 0.93). LUCK and OLUCK represent the difference between each team's PTS/g and their EPTS/g. EPW, EPL, and PWIN represent pythagorean wins, losses, and win percentage based on EPTS and OEPTS (using 2.37 as an exponent). Finally, LUCK2 is similar to GLUCK from the previous table -- i.e. is the number of wins by which the team has been lucky or un-.
• It's unlikely that the Chargers should be allowing only 10 points a game -- although that's probably more representative of their true-talent level than the 21.2 points they're actually allowing per game.
• It's a fact!: Arizona's bad.
• As with GWP, the Giants come out well by this measure, too. Though their pythagorean record would be somewhere around .500, their EPA and Opponent EPA reveals the relative strength of their team defense.
The Chargers fall to the Rams. There's still a long way to go if they want to make the playoffs.
It'd be preferable if the charts were sortable charts (especially the last one) instead of pictures. I know it's a fair bit of extra work/coding, but it would be better.
Very helpful, thanks for this post.