Gent
New Member
- Messages
- 115
- Reaction score
- 0
The team at FootballOutsiders.com has just released Profootball Prospectus, their first book giving detailed statistical analysis of all teams along with projections for 05. It's an impressive piece of work and nicely written. Other than KC Joyner's Scientific Football, I'm not aware of anything else like it. And at $13 it's definitely a better buy.
Profootball Prospectus uses a system that tracks the specific conditions under which each play occurs. That way a team/unit/player's performance can be compared to how other teams did in the exact same situation. This is expressed as a Value Over Average (VOA) percentage that is positive for a good offensive performance and negative for a good defensive effort. The system has worked well for them. Last year their predictions were more accurate than any other major sports website or magazine. These results included correctly predicting that the 4-12 Chargers would have one of the top offenses in 2004.
Their 2005 prospectus is high on the Cowboys. The projections break down as follows:
0-4 Wins: 10% probability
5-6 Wins: 15%
7-8 Wins: 23%
9-10 Wins: 24%
11+ Wins: 28%
In other words, a 75% probability that we will win more games than last season and a better than 50% probability that we will make the playoffs. They make us clear favorites to finish second in the division behind the Eagles (who are a lock for the playoffs). Washington is given a 72% chance of winning less than 9 games.
Their analysis of the Boys 2004 defense makes sense. Our run D was 6th in the league for quarters 1-3, and then quite bad in the 4th. This is no doubt due to our lack of rotation. The pass D was uniformly horrible across all quarters. Given our improved depth and secondary, Profootball Outsiders expects our 2005 defense to look much closer to the 2003 version.
On offense there is a surprise. Turns out we weren't really that bad last year, just unlucky. The passing offense was above average on 1st and 2nd down, but terrible on third down. This is a statistical anomaly: the law of averages says we should be equally good (or bad) on all downs, given enough plays. (Remember these averages take into account the specific situations we faced.) Over a 16 game season it is possible to have a bad luck streak on 3rd downs. This turns out to be a strong indicator for a turn-around the next season, and it was the major factor in the Outsiders' prediction of the Chargers offensive performance last year. In other words, even with the same personnel (and injuries), our passing offense should be better this year due simply to better luck.
All in all, this book is a great read. I highly recommend it. Along with KC's Scientific Football, I think Prospectus marks 2005 as a sort of turning point in football analysis.
-Gent
Profootball Prospectus uses a system that tracks the specific conditions under which each play occurs. That way a team/unit/player's performance can be compared to how other teams did in the exact same situation. This is expressed as a Value Over Average (VOA) percentage that is positive for a good offensive performance and negative for a good defensive effort. The system has worked well for them. Last year their predictions were more accurate than any other major sports website or magazine. These results included correctly predicting that the 4-12 Chargers would have one of the top offenses in 2004.
Their 2005 prospectus is high on the Cowboys. The projections break down as follows:
0-4 Wins: 10% probability
5-6 Wins: 15%
7-8 Wins: 23%
9-10 Wins: 24%
11+ Wins: 28%
In other words, a 75% probability that we will win more games than last season and a better than 50% probability that we will make the playoffs. They make us clear favorites to finish second in the division behind the Eagles (who are a lock for the playoffs). Washington is given a 72% chance of winning less than 9 games.
Their analysis of the Boys 2004 defense makes sense. Our run D was 6th in the league for quarters 1-3, and then quite bad in the 4th. This is no doubt due to our lack of rotation. The pass D was uniformly horrible across all quarters. Given our improved depth and secondary, Profootball Outsiders expects our 2005 defense to look much closer to the 2003 version.
On offense there is a surprise. Turns out we weren't really that bad last year, just unlucky. The passing offense was above average on 1st and 2nd down, but terrible on third down. This is a statistical anomaly: the law of averages says we should be equally good (or bad) on all downs, given enough plays. (Remember these averages take into account the specific situations we faced.) Over a 16 game season it is possible to have a bad luck streak on 3rd downs. This turns out to be a strong indicator for a turn-around the next season, and it was the major factor in the Outsiders' prediction of the Chargers offensive performance last year. In other words, even with the same personnel (and injuries), our passing offense should be better this year due simply to better luck.
All in all, this book is a great read. I highly recommend it. Along with KC's Scientific Football, I think Prospectus marks 2005 as a sort of turning point in football analysis.
-Gent
