cowpie said:
OK ,
I just pulled up the rankings onthe site and see that Maurile Tremblay has Palmer... That is
Carson Palmer... (You know, the guy who had over 3800 passing yards in each of the last 2 years with over 28 td and has great wrs johnson and houshmenzadeh.) ranked #11 out of all qbs. Is there some spot on the site where he justifies this.
I mean he has garcia ranked ahead of him. What gives. Not only that, but he has both chad johnson and TJ Housh.... both ranked in the top 10 in WRS. How can this be?? Is he predicting an injury, or is he smoking crack. If it is an injury prediction, is that what the rankings are meant for?? topredict injuries.
If it is not an injury prediction, Maurile, please lay off the pipe and let us know how you can have 2 top 10 wrs on your team and a #11 qb
and please don't update your ranking showing manning as the #1QB with harrison and wayne being ranked 40th and 41st as wrs.
Hi cowpie,I'm actually a big fan of Palmer and was surprised to see that my projections put him so low. Here's exactly the thinking that went into them.
I do my projections as a combination of bottum-up and top-down thinking.
First I project
team numbers for pass attempts, rush attempts, passing yards, rushing yards, etc.
Then I divide the attempts up for the individual positions and players. For example, I project the Bengals to have 442 rushing attempts, I project the RBs as a group to get a little over 86% of those, and for Rudi Johnson to get about 82% of the RB attempts. So 442 * 0.864 * 0.82 = 313 rushing attempts for Rudi Johnson. I do the same for every other player, for both rushing attempts and (in the case of the passing game), targets.
I then project a reception rate for each receiver, and get receptions by multiplying the projected reception rate by the projected number of targets. I also project a yards per catch and touchdowns per catch for each player, and get yards and touchdowns by multiplying those projections by projected receptions.
For QBs, I divide pass attempts between the starter and backup (usually 15/16 for the starter and 1/16 for the backup unless there's a reason to do something different), then project yards per attempt, touchdowns per attempt, and INTs per attempt for each QB to get yards, TDs, and INTs.
At this point, the total pass attempts will equal the total targets, but the total passing yards will not equal the total receiving yards except by pure luck.
I try to bring the total passing yards (and TDs) into line with the total receiving yards (and TDs) by massaging yards and TDs per attempt (for QBs) and yards and TDs per reception (for receivers), making adjustments in the direction that brings them closer to the overall team projections.
In the Bengals case, my initial team projections were 539 pass attempts for 3845 yards and 22 TDs. I stress the word
initial. There's a lot those initial projections don't take into account, so they are just a starting point. In the case of the Bengals, I think those figures are too low because they don't take into account Palmer's consistency over the last several years, and pedigree. But the team numbers don't come much into play unless the individual passer/receiver numbers disagree, so let's put those aside for now.
My projected YPA for Palmer was initially 7.7. That compares favorably with his career average of 7.4, but is a bit below is 7.8 average from last year. (The NFL average is 6.85.)
On 505 attempts (which is 15/16 of the Bengals' projected 539 attempts), that would give Palmer 3891 yards.
I also had Palmer initially projected for 0.053 TDs per attempt. That compares very favorably to the league average of 0.0395 TDs per attempt, and probably needs to be regressed down a bit.
Skip ahead to receivers . . .
I projected receptions per target, yards per reception, and TDs per reception for each WR, RB, and TE on the Bengals.
When I was finished, I had about 4120 passing yards and 28 passing touchdowns (between Palmer and Johnson) projected, and initially 4027 receiving yards and 23 receiving touchdowns for all receivers combined.
Close enough on the yardage, but TDs were a bit off. Since the receiving TDs were closer to the team projections than the QB projections were, the bulk of the adjustment came from the QBs.
I currently have the Bengals QBs throwing for 4019 yards and 23.9 TDs, and the Bengals players receiving for 4031 yards and 22.6 TDs. (There is a mistake
here in that Gregg Guenther's 7 catches for 69 yards and 0.5 TDs are not showing up.)
The TDs are lower than the Bengals have had in previous years, but Palmer is still projected for 0.045 TDs per attempt, which is significantly higher than the NFL average of 0.0395.
Hope that helps.
Maurile's a smart guy. He'll have a very good reason for this (unless he's made a mistake)
