08 Jun 2009
by Bill Barnwell
I've had an idea for a while now. Inspired by Tangotiger's Scouting Report by the fans, for the fans on defensive range and ability in baseball, I've wanted to use the collected expertise of the readers of this site to learn more about aspects of the game and its players that we simply can't learn through numbers.
My original idea was to try and create some sort of form similar to Tango's page on offensive line play, defining some sort of spectrum of abilities for offensive linemen and then getting fans who watch those players perform regularly to rate them, creating a set of individual offensive line rankings that allow for qualitative analysis.
That's a more complex project that requires significant time and buy-in, though, so I started thinking of simpler projects. One that always came to mind was projecting player performance -- while we use KUBIAK to spit out an expected total based on historical data and team variables, there are inherently things that KUBIAK might not pick up on or fully comprehend.
I'd always wanted to take our projections, as well as the projections from other sites and fantasy experts, and compare them to the thoughts of an informed group of readers and fantasy participants. Truthfully, I've been too busy/lazy when not busy to implement the infrastructure to gather the results over the last couple of summers, but I realized last week that I actually have the infrastructure needed already built for me, with an interested audience already available.
Last week, we started sending out Tweets from our account (@fb_outsiders), asking people to give us their predictions for how a given player will do in the 2009 season. To allow for easier comparison, we asked that people assume that each player participates in 16 games.
After we received a healthy response on the first day for Matt Cassel, we decided to make it a daily feature, asking Twitteronia their thoughts on a different player every day. We received a bevy of responses as the week went along, getting predictions from readers as well as members of the media like Pro Football Talk's Mike Florio, Newsday's Bob Glauber, and Rotoworld's Gregg Rosenthal.
The result is a feature you'll be seeing every Monday on FO: Wisdom Of Crowds, named after James Surowiecki's 2004 book. While we won't be providing KUBIAK figures for the players for a variety of reasons (for example, because KUBIAK isn't done yet), we'll be providing details on each player's projection by the readers and give our thoughts on the figure.
We'll be providing the average projection for each player as well as the maximum and minimum figure provided. We'll draw these from multiple predictions, so if one prediction for Cassel suggests 3000 passing yards and 15 TD, and another suggests 2900 yards with 18 TD, we'll consider the maximum projection to be 3000 yards and 18 touchdowns.
Last week, we decided to go with a Peter King-themed stash of players, focusing on four quarterbacks that caused our favorite SI columnist to spill many a word last season.
Tuesday: Matt Cassel
Average Prediction: 3289 passing yards, 19 TD, 15 INT
Maximum Prediction: 4300 passing yards, 33 TD, 8 INT
Minimum Prediction: 2400 passing yards, 10 TD, 22 INT
I chose Cassel to start because I think he's probably the most interesting player to project in the league this year. Like Edgerrin James in 2006, Cassel is undergoing a dramatic shift in context by moving from New England (the best passing offense in the history of football in 2007) to Kansas City (24th in the league last year while running a gimmick offense). When you throw in the dramatic improvement Cassel made over the second half of the season (as noted by Sean McCormick in our Four Downs series), there's significant variance on what people perceive his actual true level of ability to be. That comes out in the variance of predictions; at his best, Cassel would be among the best players in football; at his worst, he'd be a player the Chiefs wouldn't even want to re-sign.
In my opinion, Cassel's prediction is a little high. I'd expect him to throw a few more interceptions outside of the Patriots' scheme and the comfort of throwing checkdowns to Wes Welker and Kevin Faulk, although his cumulative totals should be high if the Chiefs are losing regularly. Something like 3400/18/20 seems right.
Wednesday: Matt Ryan
Average Prediction: 3589 passing yards, 22 TD, 14 INT
Maximum Prediction: 4000 passing yards, 25 TD, 10 INT
Minimum Prediction: 3100 passing yards, 14 TD, 18 INT
Ryan's another very interesting player because of his context. Last year, the Falcons were one of the unlikely success stories of the league thanks to the play of Ryan as a rookie and the ascension of Michael Turner to an elite back.
One of the other reasons, though, was health -- of the 11 spots on their offensive lineup, only one went through any injury issues. Left tackle Sam Baker went down and was replaced by Todd Weiner, who himself dealt with back issues for most of the year. Everyone else remained healthy, which is unlikely to reoccur in 2009. As a "Curse of 370" candidate, Turner seems the most obvious candidate to break down, but it's easy to imagine how much this offense would struggle were Ryan or Roddy White to suffer an injury. I'd expect Ryan's numbers to be below that average expectation.
Thursday: Tom Brady
Average Prediction: 4013 passing yards, 31 TD, 12 INT
Maximum Prediction: 5000 passing yards, 38 TD, 4 INT
Minimum Prediction: 3000 passing yards, 24 TD, 16 INT
I think the Patriots would take 5000/38/4. While I wasn't surprised that there was such a high yardage total out there, I was very surprised that not a single person picked Brady -- who had 50 TD in 2007 -- to go even above 40 in 2009. I'm not saying they're wrong, just that I was surprised.
I think the prediction generated here is reasonably accurate, actually. It seems weird to peg Brady at 31 TD, but the possibility of Randy Moss being dragged down by an injury along with some regression to the mean from the outlier that was 2007 makes me believe that the most plausible touchdown total for Brady is in the 33-35 range.
Friday: Brett Favre
Average Prediction: 3067 passing yards, 19 TD, 22 INT
Maximum Prediction: 3800 passing yards, 24 TD, 17 INT
Minimum Prediction: 2450 passing yards, 15 TD, 30 INT
Alright -- for this prediction, we asked people to project Favre if he played 16 games as the starting quarterback for the Vikings. Strangely, we got a bunch of responses expecting Favre to get hurt; while he's old and already has a shoulder issue, the guy hasn't missed a game in 14 years. He's probably going to be on the field for all 16 if he makes it to one.
Obviously, hopes for Favre in Minnesota aren't very high -- I'm guessing our readership is more aware of the ridiculously easy schedule Favre struggled against last season, and doesn't have high hopes for an offense that features Bernard Berrian and Percy Harvin as likely starting wideouts. I tend to agree.
This week, we'll be taking predictions on five more quarterbacks (including Tony Romo today), and I'll be next Monday and each one thereafter with another article summating what you guys are thinking about the chances of five new players in the upcoming season.
To participate, all you need to do is follow us on Twitter @fb_outsiders, where we regularly poke holes in Drew Rosenhaus and answer your questions about all things FO.
71 comments, Last at 17 Jun 2009, 5:41am by Marver