Busting the Predictions

Categories: NFL, Statistics
Comments: No Comments
Published on: December 21, 2012

On a day that will come to represent busted predictions, I thought it would be interesting to talk about how my own predictions can and, likely will, bust.  At least they will bust if we continue to look at the game of football using the same stats we’ve always used.

Most of us are focused on outcomes and use those outcomes to evaluate players.  Fantasy football only makes this tendency worse as the traditional stats a player gains are directly relevant to outcomes that are personally important to us.  We see that Calvin Johnson could accumulate more receiving yards in a season than any other player in NFL history.  Based on this, we assume that Calvin Johnson is a high quality receiver.  I was watching Sportscenter today and it was almost a foregone conclusion that Calvin Johnson is very good at his job.  However, at the end of the segment, the commentator noted something important.  On the show, it was presented almost as an afterthought, but it goes to the heart of why the Lions are 4-10 and not contending.  Through 14 games, the Lions have thrown the ball more times than the team that set the previous record, the 1995 San Francisco 49ers, threw the ball for an entire season.

Yes, Calvin Johnson may be racking up the receiving yards, but that doesn’t mean he’s doing much to help his team win.  A receiver can get credit for more receiving yards by simply being the target of more passes.  Calvin Johnson has been targeted 174 times this season, more than any other receiver in the NFL.  Is he doing anything special with all these balls thrown his way?  No.  My analysis says that any average NFL receiver that was targeted 174 times would have approximately 109 receptions.  Calvin Johnson currently has 106.  He does not have an elite ability to catch a football.  He simply gets more opportunities than anyone else.  This makes the record he will likely break not all that meaningful.  It also sets the Lions up for a vicious cycle of failure.  Anything Calvin Johnson does gets perceived as a reason to continue giving him opportunities.  So the Lions continue to give him opportunities never realizing he is dropping and/or missing just as many catches as any other average NFL receiver. Continuing to use opportunities trying to get a football to an average NFL receiver is not a winning strategy because you then need more opportunities to advance the ball and ultimately score points.  It’s especially unforgivable when better options exist on the same team (or at least they did until Broyles tore his ACL).

Whether a player will be viewed as a success or a failure is heavily dependent on many factors.  Some of those factors are outside the player’s control and may not reflect their actual talent. An average quarterback may appear below average if placed in a bad offensive system (I’m looking at you, Brandon Weeden).   Or, a quarterback having a terrible year may appear to be doing something great if surrounded by the right collection of receivers (I’m looking at you, Andrew Luck).

So, I’m telling you that Geno Smith will be a quarterback bust.  However, being drafted by a team with an above average offensive system or a collection of receivers that are actually talented at catching a football could turn that prediction on its head.  All I can tell you is what a player individually brings to the table.  Geno Smith didn’t complete any more or less passes than we would expect an average quarterback to complete with that offensive system and set of receivers. This is why I say Geno Smith is not a top talent.  That doesn’t mean he won’t be “successful” in the NFL, if we continue to define success in terms of individual outcomes.

In summary, when you look at outcomes, you have to look at all that factors that influence those outcomes.  Average players can have great years by getting lots and lots of opportunities.  That’s all that’s happening in Detroit this year and it will happen in future years.  Focusing on outcomes will get you a high quality fantasy football team and a lot of individual statistical records, but it will also have you leaving the field with a loss.

2013 NFL Draft: Wide Receivers

Hi everyone.  Finals are over, the grading is (mostly) done, and I’m ready to dive back into blogging.

If you’ve been reading this thing the last few months, you might notice something weird.  My tagline says that I use analysis and statistics to evaluate quarterbacks and wide receivers.  But I’ve never breathed a word about any wide receivers.  There are good reasons for this.  First, I can calculate a Completions Away from Average for a wide receiver, just like a can for a quarterback.  This number indicates the difference between this receiver’s completions and the number of completions we would expect an average receiver to have given the same offensive system, defensive competition, and quarterback.  Positive numbers indicate this receiver is catching more than an average receiver and negative numbers indicate this receiver is catching less.

There are problems though.  First off, I ran the calculations on NFL receivers and the data came out all wonky.  What do you do with an analysis that says Calvin Johnson has a CAA of -3.92, indicating that he has 3.92 fewer completions than an average NFL wide receiver would have?  Especially in a year when Calvin Johnson could break the receiving yards record.  What I suspect is happening with Calvin Johnson something similar to the Yay Points thesis over at Wages of Wins.  If you’re unfamiliar, basically thesis says that when basketball evaluators judge talent, they weight points scored too highly and don’t put enough weight on things like shooting percentage or other positive things like rebounds.  I suspect wide receivers might be subject to the same bias.  Call it Yay Yards or Yay Touchdowns or whatever you like.  Calvin Johnson might be racking up the yards, but only because the Lions are constantly throwing the ball in his direction.  Another player might be able to accomplish the same thing more efficiently if they were given the same number of opportunities.  However, without a fully functioning model predicting team wins, we can only speculate.

So, take these recommendations with a whole load of salt.  I believe in them, but more evidence is needed before you should accept these numbers without skepticism.

And now, I give you my top 5 wide receivers of the 2013 draft class.

#1)  Tavon Austin – West Virginia

2012 CAA – 9.35

Career CAA – 18.95

This one should be no surprise.  Many say that Tavon Austin is a great wide receiver, and my analysis is no exception.

#2)  Cody Wilson – Central Michigan

2012 CAA – 10.43

Career CAA – 17.03

This is a surprise.  This guy is almost an exact carbon copy of Tavon Austin in physical size.  My analysis says he’s also a carbon copy in talent.  Value conscious GM’s may want to take this guy in the 6th-7th round over Austin in the 1st.

#3)  Keenan Allen – California

2012 CAA – 11.21

Career CAA – 16.63

Another one that should be no surprise.  Many people have this guy as a top round talent, and I would agree.

#4)  Ryan Swope – Texas A&M

2012 CAA – 4.29

Career CAA – 16.10

Ryan has been the most consistent performer on our list, rising to the number 4 spot using a tortoise strategy rather than the hare strategy of Tavon Austin and Keenan Allen.  2012 was actually his worst season catching the ball.  2011 was his best when he tallied 6.44 CAA.  This guy is a consistent performer across his career.

#5)  Brent Leonard – Louisiana-Monroe

2012 CAA – 8.03

Career CAA – 12.32

Another player that no one is talking about, but that has talent for catching a football.  I don’t even see this guy on draft boards.  Look for this player to be one of the few surprise undrafted free agents to make an impact next year.

Honorable Mention – Eric Page – Toledo

Senior CAA (2011) – 6.72

Career CAA – 10.11

I really, really hope someone gives this guy a chance.  He is the University of Toledo’s all-time leading receiver.  Last year, he went undrafted.  My analysis puts him as the 4th best wide receiver in the class (#1 and #3 are current NFL players Ryan Broyles & Cole Beasley).  He was signed by the Denver Broncos as a free agent.  I was looking at him with hopeful eyes.  Here is a talented player that made the roster with Peyton Manning throwing him the football.  The pieces were aligning for him to have a great season.  And then…torn ACL.  Denver released him and now we might never know how good he could have been.  But…here’s hoping.

New Look

Categories: Uncategorized
Tags: No Tags
Comments: No Comments
Published on: December 13, 2012

The combination of the new WordPress update and my old theme broke my Site Stats.  I structure my entire life based on what numbers can tell me.  I couldn’t be without my Site Stats.  Therefore, I had to find a new theme.  I’m currently using the default Worpress theme, but the look of the site will update from time to time as I check for new themes.

A Viking, a Jet, and a Mountaineer walk into a bar…

As I was watching the Vikings-Packers game last weekend, a strange thing happened.  I agreed completely with Troy Aikman.  He was talking about the Viking’s struggles in the passing game and mentioned that the Viking’s receivers were simply not winning the one-on-one battles.  This meant that Christian Ponder had nowhere to throw the ball and therefore couldn’t complete any passes.

I’ve already mentioned my take on the Viking’s problems, and that assessment hasn’t changed.  With Percy Harvin out for the season, the Viking’s are likely to struggle in the passing game for the rest of the season.  However, the fault won’t be with Christian Ponder, who is still above average in terms of completing passes to the Viking’s set of receivers.

What about the newest quarterback controversy in New York.  The Jets are under considerable fan pressure to bench Mark Sanchez.  What does the model say?  Is Mark Sanchez to blame for the Jets’ struggles?  This question is more difficult to answer than you might think, but only because of historical coincidence.  My model relies on historical data to observe how performance changes when receivers, quarterbacks, and offensive systems change.  Mark Sanchez and the current Jets coaching staff have never been separate from one another.  Rex Ryan’s staff has never had a different quarterback and Mark Sanchez has never worked within a different offensive system in the NFL.  So, while the model can identify that the problem in New York isn’t a receiver problem, it can’t separate the quarterback from the Jets offensive system (if anyone knows of a good site for seeing historical lists of offensive coordinators and play callers, please let me know in the comments).

What does this all have to do with a Mountaineer?  Well, it’s a lot easier to separate offensive system from quarterback in the NCAA because the quarterbacks turn over so quickly.  This turns our attention to Geno Smith, the current consensus #1 pick.  How much of Geno Smith’s success is due to the West Virginia offensive system?  How much is due to Geno’s quarterbacking ability?  Sadly, the answer is that everything that makes Geno Smith an above average NCAA quarterback is attributable to the West Virginia offensive system.  Geno Smith’s Career CAA is only 3.78.  In his entire NCAA career, he has only completed 3.78 passes that an average NCAA quarterback wouldn’t have completed.  The rest of the success comes from the coach.

I find it strange that no one is talking about this, given that Dana Holgorsen made a name for himself as an offensive coaching specialist.  My model says that there is nothing unique about Geno Smith.  Any average NCAA quarterback dropped into the West Virginia offensive system would perform just as well.  I hope someone with decision making power recognizes this before heaping a load of expectations onto him and turning him into the next JaMarcus Russell.  The new rookie wage scale will limit the damage, but Geno Smith is not likely to be the answer for the Kansas City Chiefs or any other team in the market for a quarterback.

2013 Draft Predictions

I’ve added a new section for all the data lovers.  The numbers section has full Completions Away from Average data on all quarterbacks in NCAA FBS football for the 2012 season.  Of course that information will change with bowls, but I won’t update it again until the national championship game is over.

Along with that, I thought it was high time to start talking 2013 draft.  I put up one post that examined talent in the 2013 quarterback class.  That post only looked at data from the 2012 season up through the end of October.  Now we have full career data on all quarterbacks in the draft and can make some good predictions. These picks are based on a regression equation that predicts NFL Passer Rating after 3 years in the NFL.

#1)  Zac Dysert – Miami (OH)

Career CAA – 91.53

2012 CAA – 6.96

Zac has had a difficult 2012.  It’s actually been his worst season of his college career.  That being said, he’s still the most likely of all the draft class to succeed at the next level, according to the equation.  Hopefully he goes to a team that can recognize and use his considerable talents.

#2)  Ryan Aplin – Arkansas State

Career CAA – 83.78

2012 CAA – 29.46

Ryan Aplin will be the steal of the 2013 draft, assuming he gets drafted at all.  No website I look at has him higher than the 11th quarterback and most have him considerably lower.  And all he’s done to get such low ratings is do everything every prognosticator has ever asked of him.  Assuming he gets drafted and assuming he gets some reasonable playing time, look for big things from this guy

#3) Sean Renfree – Duke

Career  CAA – 66.66

2012 CAA – 1.48

Another quarterback that has had a great career, but not a great 2012.  If Sean can regain the form he showed in 2010 and 2011, he is likely to be a success in the NFL.

#4)  Ryan Griffin – Tulane

Career CAA – 47.33

2012 CAA – 22.39

Another quarterback nobody’s talking about, but is likely to be a success in the NFL as long as he gets some playing time.  I wasn’t kidding when I said I thought everyone was looking in the wrong place for quarterback talent this year, eh?

#5)  Ryan Nassib – Syracuse

Career CAA – 44.77

2012 CAA – 6.99

Finally, a player somebody is seriously talking about as a possible first rounder.  Ryan is an interesting player in that most of this prediction comes from his junior season.  The question for Ryan is if the junior season was a one time thing or if he can regain that form.

Honorable Mentions (6-10)

Austin Dantin – Toledo

Dalton Williams – Akron

Aaron Murray – Georgia

B.R. Holbrook – New Mexico

Jeff Tuel – Washington State

Underclassmen to Watch

Keith Price – Washington

In a just world, Keith Price would win a Heisman Trophy for his performance this year, enter the draft, and be a first round pick.  Sadly, the world is not just.  Even if he did enter the draft, he probably wouldn’t go very high, wouldn’t get to play much, and a great talent would be wasted.  Hopefully he starts to generate some buzz next season because his talent is through the roof.

Rakeem Cato – Marshall

Two great years at Marshall and hoping for more.

Brett Hundley – UCLA

As a freshman, Brett posted the second highest CAA for the 2012 season.  Thanks to the large market that UCLA brings, this guy will probably pick up a lot of buzz next season.

page 1 of 1
Welcome , today is Thursday, February 22, 2018