Redskins Have a 2.8 Percent Chance of Finishing Above .500
My inbox is really filling up with your strange ideas and stranger quests. Keep 'em coming. This morning's installment is courtesy of Washington Post business writer Neil Irwin, currently on leave to study at Columbia University, a fine school that clearly isn't doing enough to keep Neil busy. He writes:
I've been putting my business school classwork to good use by doing some analysis of possible outcomes for the Redskins' season.
I base my analysis on innovative statistics developed by Football Outsiders, the best and smartest football site on the web. Their concept, called DVOA (or, defense-adjusted value over average), attempts to capture the true "quality" of each team, based on a statistical analysis of each play, rather than just total yards gained and allowed or final score or anything so crude as that. It's adjusted for quality of opponent, and it factors out things that are essentially random. (For example, their research has found that success at recovering fumbles is random, meaning past success doesn't predict future success. So is the success or failure of your opponent's field goal attempts.)
I've figured out a way to convert their statistics into a probability of a given team winning a given game, then plotted out the probability of every possible combination of victories and defeats that the Skins could have. I first did that two weeks ago, before the Titans game. I had the probability of the Skins winning that game at 84 percent, and
the probability of their having a winning record for the season at 30 percent.
The last two weeks have been disasters. There are three reasons that the rest of the season looks grim. First, the 84 percent chance of beating the Titans and 34 percent chance of beating the Colts have been replaced by a zero chance of winning each game. Second, those two games give reason to think that the Skins are a worse team than they seemed two weeks ago, lowering their probability of winning each remaining game. But the least-noted element of the Skins' outlook from here is that almost every opponent the team has coming up has gotten better in the last couple of weeks. At this point, every single remaining opponent looks like a pretty good team; even Tampa Bay, given the wins over Cincinnati and Philly.
Combine the Skins getting worse with their opponents getting better, and the probability of beating the Giants at home in December has dropped from 56 percent two weeks ago to 40 percent now. Odds of winning at New Orleans have dropped from 47 percent to 35 percent, and odds of beating St. Louis have fallen from 45 percent to 35 percent.
Add all that up (or get Excel to do so for you), and the probability of the Redskins finishing with a winning record has fallen from 30 percent two weeks ago to 2.8 percent today. Considering that many 9-7 teams don't make the playoffs, the odds of making the postseason is even smaller than that. That seems to me to make a good argument for writing this season off, starting Campbell the rest of the season, and finding out whether he can really be the quarterback of the future.
Neil sent me the spreadsheets that explain all these calculations. If anyone wants to take a closer look, lemme know.
Posted by: jhorstma | October 27, 2006 7:42 AM | Report abuse
Posted by: charlie faulkner | October 27, 2006 9:29 AM | Report abuse
Posted by: B.P. | October 27, 2006 10:06 AM | Report abuse
Posted by: MK | October 27, 2006 11:59 AM | Report abuse
The comments to this entry are closed.