With QB Brett Favre out due to a death in his wife's family, QB Aaron Rodgers played with the starters and completed 13 of 16. He would have had two more completions if his receivers could have held onto the ball. It's only one practice so it is nothing to get excited about, but it does transition nicely into a discussion of Rodgers and the Football Outsiders quarterback projection system.
I'm not convinced Rodgers will ever be a good NFL QB, but I want to see him start for a season or two before he changes my mind or whether the Packers should give up on him. Football Outsiders isn't very encouraged about his future either. In the 2007 Pro Football Prospectus, David Lewin says the best way to project the success of a NFL QB is look at his college completion percentage and number of college games started. It's pretty simple, but Lewin says it's predictive power is accurate 70% of the time for QBs selected in the top 16 and 54% of those picked 17 to 64.
The most powerful predictive stat is completion percentage, but Lewin says number of college starts are important because it gives NFL scouts more games to evaluate and allows them to make a better decision. Lewin doesn't appear to mention the larger sample size, but I would think more games played creates a more accurate completion percentage.
Lewin's quarterback projection expects that Rodgers' best season will only be an average NFL QB season. Rodgers' 63.8% college completion percentage is impressive, but he only started 22 college games, compared to other QBs like Eli Manning who started 37, or the NCAA record holder Phillip Rivers who started an amazing 51 games. With a smaller sample size for Rodgers, his college completion percentage isn't as powerful a predictive statistic as Rivers' similar college 63.5% completion rate. Still the projection is still wrong a lot of the time and Rodgers is only 23 years old, so there are reasons to believe in him. Or it might be a good reason why we hope that Favre will keep postponing his retirement.